CN117367807A - Method, system, equipment and medium for diagnosing faults of aero-engine - Google Patents
Method, system, equipment and medium for diagnosing faults of aero-engine Download PDFInfo
- Publication number
- CN117367807A CN117367807A CN202210784395.3A CN202210784395A CN117367807A CN 117367807 A CN117367807 A CN 117367807A CN 202210784395 A CN202210784395 A CN 202210784395A CN 117367807 A CN117367807 A CN 117367807A
- Authority
- CN
- China
- Prior art keywords
- fault diagnosis
- knowledge
- fault
- result
- diagnosis result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000003745 diagnosis Methods 0.000 claims abstract description 705
- 238000007499 fusion processing Methods 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 81
- 230000004927 fusion Effects 0.000 claims description 44
- 238000007405 data analysis Methods 0.000 claims description 34
- 238000013441 quality evaluation Methods 0.000 claims description 16
- 238000012216 screening Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 5
- 238000007689 inspection Methods 0.000 claims description 5
- 238000013215 result calculation Methods 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims 1
- 238000004364 calculation method Methods 0.000 description 24
- 238000007781 pre-processing Methods 0.000 description 18
- 238000013500 data storage Methods 0.000 description 17
- 238000000605 extraction Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 239000002699 waste material Substances 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 239000010687 lubricating oil Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000001303 quality assessment method Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000295 fuel oil Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a fault diagnosis method, a system, equipment and a medium of an aeroengine, wherein the fault diagnosis method of the aeroengine comprises the steps of obtaining running state parameters and configuration information of the aeroengine to be diagnosed; according to the configuration information, matching in a preset fault diagnosis model library and a preset fault diagnosis knowledge library respectively to obtain a corresponding fault diagnosis model and fault diagnosis knowledge; acquiring a first fault diagnosis result based on the operation state parameters and the fault diagnosis model; acquiring a second fault diagnosis result based on the operation state parameters and the fault diagnosis knowledge; and carrying out fusion processing on the first fault diagnosis result and the second fault diagnosis result to obtain a first target diagnosis result of the aero-engine to be diagnosed, so as to realize providing a universal fault diagnosis flow for the aero-engines with different structural schemes, thereby improving the universality of the high aero-engine fault diagnosis system and the accuracy, reliability and interpretability of the aero-engine fault diagnosis result.
Description
Technical Field
The invention relates to the technical field of fault diagnosis, in particular to a fault diagnosis method, system, equipment and medium of an aeroengine.
Background
Different types of faults such as gas circuit faults, fuel oil system faults, vibration faults and the like can possibly occur in the operation process of the aero-engine, if the faults can not be monitored and removed in time, the aero-engine can be stopped, parts are damaged, and even the aero-engine is scrapped when serious. In order to ensure safe and reliable operation of the aero-engine, the aero-engine is subjected to fault diagnosis through measurement parameters acquired by the aero-engine sensors, and the problems which occur or are likely to occur are evaluated according to important value and significance.
The aero-engine generally has different configurations according to different functional characteristics, physical characteristics, requirements, designs and use information, and the aero-engines with the same configuration can also have different requirements, namely, the characteristics and arrangement positions of sensors are different, so that the fault diagnosis system of the aero-engine generally needs to be configured with special fault diagnosis procedures and methods for the aero-engines with different models, and the fault diagnosis of the aero-engines with different configurations and different sensor schemes cannot be realized based on a single method.
Disclosure of Invention
The invention aims to overcome the defect that the fault diagnosis of aeroengines with different configurations and different sensor schemes cannot be realized in the prior art, and provides a fault diagnosis method, a system, equipment and a medium for the aeroengines.
The invention solves the technical problems by the following technical scheme:
the invention provides a fault diagnosis method of an aeroengine, which comprises the following steps:
acquiring operation state parameters and configuration information of an aeroengine to be diagnosed;
according to the configuration information, matching in a preset fault diagnosis model library and a preset fault diagnosis knowledge library respectively to obtain a corresponding fault diagnosis model and fault diagnosis knowledge; the preset fault diagnosis model library is a fault diagnosis model library of the aeroengine generated based on a data driving algorithm, and the preset fault diagnosis knowledge library is a fault diagnosis knowledge library of the aeroengine generated based on a knowledge reasoning algorithm;
acquiring a first fault diagnosis result based on the running state parameter and the fault diagnosis model;
acquiring a second fault diagnosis result based on the operating state parameters and the fault diagnosis knowledge;
And carrying out fusion processing on the first fault diagnosis result and the second fault diagnosis result to obtain a first target diagnosis result of the aero-engine to be diagnosed.
Preferably, after the matching obtains the corresponding fault diagnosis model and the fault diagnosis knowledge, the fault diagnosis method further comprises:
screening fault diagnosis knowledge according to the state indication from the matched fault diagnosis knowledge;
respectively judging whether the matched fault diagnosis model and the fault diagnosis knowledge indicated according to the state meet preset conditions or not;
if not, deleting the fault diagnosis model which does not meet the preset conditions and fault diagnosis knowledge indicated according to the state from a set diagnosis list;
the preset conditions comprise that first sensor related parameters of the aeroengine to be diagnosed are matched with second sensor related parameters required to be used by the fault diagnosis model or the fault diagnosis knowledge.
Preferably, when there are at least two fault diagnosis models that satisfy the preset condition, the obtaining the first fault diagnosis result based on the operation state parameter and the fault diagnosis model includes:
And respectively inputting the running state parameters into a plurality of fault diagnosis models to obtain a plurality of first intermediate diagnosis results, fusing to obtain the corresponding first fault diagnosis results, and storing the first fault diagnosis results into a fault diagnosis result library.
Preferably, when there are at least two fault diagnosis knowledge according to the status indication that satisfies the preset condition, the obtaining a second fault diagnosis result based on the operating status parameter and the fault diagnosis knowledge includes:
and respectively inputting the operation state parameters into a plurality of fault diagnosis knowledge according to the state indication to obtain a plurality of second intermediate diagnosis results, fusing to obtain the corresponding second fault diagnosis results, and storing the second fault diagnosis results into the fault diagnosis result library.
Preferably, the fusing the first fault diagnosis result and the second fault diagnosis result, and obtaining the first target diagnosis result of the aero-engine to be diagnosed includes:
acquiring a corresponding first fault diagnosis result and a corresponding second fault diagnosis result from the fault diagnosis result library;
determining a corresponding weight relation according to a fault diagnosis model meeting the preset conditions and the diagnosis accuracy of the configuration of the aeroengine to be diagnosed corresponding to the fault diagnosis knowledge indicated by the state;
Carrying out fusion diagnosis result calculation on the first fault diagnosis result, the second fault diagnosis result and the weight relation by using an information fusion algorithm so as to obtain a first target diagnosis result of the aero-engine to be diagnosed;
the first target diagnosis results corresponding to the fault types of the aero-engine to be diagnosed are arranged according to the size of the fusion diagnosis results.
Preferably, the fault diagnosis method further includes:
acquiring phenomenon description information corresponding to at least one condition of the aero-engine to be diagnosed before operation, after operation and in the operation process;
the phenomenon description information comprises at least one of field personnel description, nondestructive inspection results and hole detection result description;
screening fault diagnosis knowledge according to the phenomenon description from the matched fault diagnosis knowledge;
and inputting the phenomenon description into the fault diagnosis knowledge according to the phenomenon description to carry out knowledge reasoning so as to obtain a third fault diagnosis result.
Preferably, the fault diagnosis method further includes:
and carrying out fusion processing on the first fault diagnosis result, the second fault diagnosis result and the third fault diagnosis result to obtain a second target diagnosis result of the aeroengine to be diagnosed.
Preferably, the acquiring the operation state parameters of the aeroengine to be diagnosed includes:
collecting and detecting a data format of original operation parameters of the aeroengine to be diagnosed;
matching in a preset data analysis algorithm library according to the data format to obtain a corresponding data analysis algorithm;
and acquiring the operation state parameters of the aero-engine to be diagnosed based on the original operation parameters and the data analysis algorithm.
Preferably, the fault diagnosis method further includes:
judging whether the running state parameters meet the preset quality evaluation requirements or not; the data quality evaluation requirement comprises at least one of a missing value, an outlier, a redundancy value, a unit standard and a data type standard;
if yes, outputting the running state parameters;
and if not, deleting the running state parameters.
The invention also provides a fault diagnosis system of the aeroengine, which comprises:
the first acquisition module is used for acquiring running state parameters and configuration information of the aeroengine to be diagnosed;
the matching module is used for respectively matching in a preset fault diagnosis model library and a preset fault diagnosis knowledge library according to the configuration information to obtain a corresponding fault diagnosis model and fault diagnosis knowledge; the preset fault diagnosis model library is a fault diagnosis model library of the aeroengine generated based on a data driving algorithm, and the preset fault diagnosis knowledge library is a fault diagnosis knowledge library of the aeroengine generated based on a knowledge reasoning algorithm;
The first fault diagnosis result acquisition module is used for acquiring a first fault diagnosis result based on the running state parameters and the fault diagnosis model;
the second fault diagnosis result acquisition module is used for acquiring a second fault diagnosis result based on the running state parameters and the fault diagnosis knowledge;
and the first diagnosis result fusion module is used for carrying out fusion processing on the first fault diagnosis result and the second fault diagnosis result to obtain a first target diagnosis result of the aeroengine to be diagnosed.
Preferably, the fault diagnosis system further comprises:
the first screening module is used for screening fault diagnosis knowledge according to the state indication from the matched fault diagnosis knowledge;
the first judging module is used for judging whether the matched fault diagnosis model and the fault diagnosis knowledge indicated according to the state meet preset conditions or not respectively;
if not, deleting the fault diagnosis model which does not meet the preset conditions and fault diagnosis knowledge indicated according to the state from a set diagnosis list;
the preset conditions comprise that first sensor related parameters of the aeroengine to be diagnosed are matched with second sensor related parameters required to be used by the fault diagnosis model or the fault diagnosis knowledge.
Preferably, when at least two fault diagnosis models meeting the preset conditions exist, the first fault diagnosis result obtaining module is further configured to input the running state parameters into a plurality of fault diagnosis models to obtain a plurality of first intermediate diagnosis results, fuse the first fault diagnosis results to obtain corresponding first fault diagnosis results, and store the first fault diagnosis results in a fault diagnosis result library.
Preferably, when there are at least two fault diagnosis knowledge according to the state indication that meets the preset condition, the second fault diagnosis result obtaining module is further configured to input the running state parameter into a plurality of fault diagnosis knowledge according to the state indication to obtain a plurality of second intermediate diagnosis results, fuse the second intermediate diagnosis results to obtain corresponding second fault diagnosis results, and store the second fault diagnosis results in the fault diagnosis result library.
Preferably, the diagnostic result fusion module includes:
the diagnosis result acquisition unit is used for acquiring a corresponding first diagnosis result and a corresponding second diagnosis result from the fault diagnosis result library;
the weight determining unit is used for determining a corresponding weight relation according to a fault diagnosis model meeting the preset condition and the diagnosis accuracy rate of the configuration of the aeroengine to be diagnosed corresponding to the fault diagnosis knowledge indicated by the state;
The fusion diagnosis result calculation unit is used for calculating the fusion diagnosis result of the first fault diagnosis result, the second fault diagnosis result and the weight relation by using an information fusion algorithm so as to obtain a first target diagnosis result of the aeroengine to be diagnosed;
the first target diagnosis results corresponding to the fault types of the aero-engine to be diagnosed are arranged according to the size of the fusion diagnosis results.
Preferably, the fault diagnosis system further comprises:
the second acquisition module is used for acquiring phenomenon description information corresponding to at least one condition of the aero-engine to be diagnosed before operation, after operation and in the operation process;
the phenomenon description information comprises at least one of field personnel description, nondestructive inspection results and hole detection result description;
the second screening module is used for screening out fault diagnosis knowledge described according to the phenomenon from the matched fault diagnosis knowledge;
and the third fault diagnosis result acquisition module is used for inputting the phenomenon description into the fault diagnosis knowledge according to the phenomenon description to perform knowledge reasoning so as to acquire a third fault diagnosis result.
Preferably, the fault diagnosis system further comprises:
and the second diagnosis result fusion module is used for carrying out fusion processing on the first diagnosis result, the second diagnosis result and the third diagnosis result so as to acquire a second target diagnosis result of the aeroengine to be diagnosed.
Preferably, the first obtaining module includes:
the original operation parameter acquisition unit is used for acquiring and detecting the data format of the original operation parameters of the aero-engine to be diagnosed;
the data analysis algorithm matching unit is used for matching in a preset data analysis algorithm library according to the data format to obtain a corresponding data analysis algorithm;
and the running state parameter acquisition unit is used for acquiring the running state parameters of the aeroengine to be diagnosed based on the original running parameters and the data analysis algorithm.
Preferably, the fault diagnosis system further comprises:
the parameter quality judging module is used for judging whether the running state parameters meet the preset quality evaluation requirements or not; the data quality evaluation requirement comprises at least one of a missing value, an outlier, a redundancy value, a unit standard and a data type standard;
If yes, outputting the running state parameters;
and if not, deleting the running state parameters.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method for diagnosing faults of an aeroengine as described above when executing the computer program.
The invention also provides a computer-readable storage medium on which a computer program is stored which, when executed by a processor, implements a method for diagnosing a fault of an aeroengine as described above.
The invention has the positive progress effects that: the method comprises the steps of obtaining running state parameters and configuration information of the aero-engine to be diagnosed, matching corresponding fault diagnosis models and fault diagnosis knowledge according to the configuration information, obtaining corresponding fault diagnosis results according to the fault diagnosis models and the fault diagnosis knowledge which are successfully matched, and carrying out fusion processing on the two different fault diagnosis results to obtain target diagnosis results so as to realize the general fault diagnosis flow provided for the aero-engines with different structural schemes, thereby improving the universality of the high aero-engine fault diagnosis system and the accuracy, reliability and interpretability of the aero-engine fault diagnosis results.
Drawings
Fig. 1 is a flow chart of a fault diagnosis method of an aero-engine provided in embodiment 1 of the present invention.
Fig. 2 is a first flow chart of a fault diagnosis method of an aero-engine according to embodiment 2 of the present invention.
Fig. 3 is a second flow chart of the fault diagnosis method of the aero-engine provided in embodiment 2 of the present invention.
Fig. 4 is a third flow chart of the fault diagnosis method of the aero-engine provided in embodiment 2 of the present invention.
Fig. 5 is a fourth flowchart of a fault diagnosis method for an aero-engine according to embodiment 2 of the present invention.
Fig. 6 is a fifth flowchart of a fault diagnosis method for an aero-engine according to embodiment 2 of the present invention.
Fig. 7 is a system configuration diagram of a fault diagnosis method for an aero-engine according to embodiment 2 of the present invention.
Fig. 8 is a flow chart of an aeroengine fault diagnosis based on knowledge reasoning for the method for diagnosing a fault of an aeroengine according to embodiment 2 of the present invention.
Fig. 9 is a flow chart of an aeroengine fault diagnosis based on a data driving algorithm for the fault diagnosis method of an aeroengine provided in embodiment 2 of the present invention.
Fig. 10 is a schematic block diagram of a fault diagnosis system of an aero-engine according to embodiment 3 of the present invention.
Fig. 11 is a schematic block diagram of a fault diagnosis system of an aero-engine provided in embodiment 4 of the present invention.
Fig. 12 is a schematic block diagram of a first diagnostic result fusion module of the fault diagnosis system of an aero-engine according to embodiment 4 of the present invention.
Fig. 13 is a schematic block diagram of a first obtaining module of a fault diagnosis system of an aero-engine according to embodiment 4 of the present invention.
Fig. 14 is a schematic structural diagram of an electronic device for implementing a fault diagnosis method of an aero-engine according to embodiment 5 of the present invention.
Detailed Description
The invention is further illustrated by means of the following examples, which are not intended to limit the scope of the invention.
Example 1
The present embodiment provides a fault diagnosis method for an aeroengine, as shown in fig. 1, where the fault diagnosis method includes:
s101, acquiring operation state parameters and configuration information of an aeroengine to be diagnosed.
The to-be-diagnosed operation state parameters of the aero-engine can comprise at least one of a gas circuit system parameter, a fuel system parameter, a lubricating oil system parameter, a vibration parameter, a state information parameter and a flight parameter of an aero-engine control system of the aero-engine; configuration information of the aircraft engine to be diagnosed includes, but is not limited to, at least one of a double rotor, a triple rotor, an interposed bearing, an unbiased bearing, and the like. The above is by way of example and is not intended to limit the operating state parameters and configuration information of the aircraft engine to be diagnosed.
S102, respectively matching in a preset fault diagnosis model library and a preset fault diagnosis knowledge library according to configuration information to obtain a corresponding fault diagnosis model and fault diagnosis knowledge; the preset fault diagnosis model library is a fault diagnosis model library of the aeroengine generated based on a data driving algorithm, and the preset fault diagnosis knowledge base is a fault diagnosis knowledge base of the aeroengine generated based on a knowledge reasoning algorithm.
Optionally, matching can be performed in a preset fault diagnosis model library and a preset fault diagnosis knowledge base according to configuration information of the aeroengine to be diagnosed, if the matching result of the fault diagnosis model comprises one or more fault diagnosis models, continuing to operate, if no corresponding matching result exists, skipping other steps, and if the diagnosis result based on the fault diagnosis model is null; if the fault diagnosis knowledge matching result contains one or more diagnosis knowledge, the operation can be continued; if no corresponding matching result exists, the rest steps are skipped, and the diagnosis result based on the fault diagnosis knowledge is null.
The fault diagnosis model library based on the data driving algorithm comprises at least one of a statistical analysis model, a machine learning model (including but not limited to a decision tree, a support vector machine, a random forest and other machine learning algorithm models) and a deep learning model (including but not limited to a cyclic neural network, a convolutional neural network and other deep learning algorithm models); the fault diagnosis knowledge base based on the knowledge reasoning algorithm comprises at least one of diagnosis rules, basic facts and other information. The foregoing is illustrative, and the fault diagnosis model library and the fault diagnosis knowledge library are not limited thereto.
S103, acquiring a first fault diagnosis result based on the operation state parameters and the fault diagnosis model.
Optionally, the operation state parameters of the aeroengine to be diagnosed are input into the fault diagnosis model matched in step S102, so as to obtain a first fault diagnosis result.
S104, acquiring a second fault diagnosis result based on the operation state parameters and fault diagnosis knowledge.
Optionally, knowledge reasoning is performed on the operation state parameters of the aeroengine to be diagnosed in combination with the fault diagnosis knowledge matched in step S102, so as to obtain a second fault diagnosis result.
S105, carrying out fusion processing on the first fault diagnosis result and the second fault diagnosis result, and obtaining a first target diagnosis result of the aeroengine to be diagnosed.
According to the method, the device and the system, the running state parameters and configuration information of the aero-engine to be diagnosed are obtained, the corresponding fault diagnosis model and fault diagnosis knowledge are matched according to the configuration information, the corresponding fault diagnosis result is obtained according to the successfully matched fault diagnosis model and fault diagnosis knowledge, and the target diagnosis result is obtained by fusion processing based on two different fault diagnosis results, so that a general fault diagnosis flow is provided for the aero-engine with different structural schemes, and the universality of the high aero-engine fault diagnosis system and the accuracy, reliability and interpretability of the aero-engine fault diagnosis result are improved.
Example 2
The present embodiment provides a fault diagnosis method for an aeroengine, as shown in fig. 2, which is a further improvement of embodiment 1, specifically:
in an alternative implementation manner, after matching to obtain the corresponding fault diagnosis model and the fault diagnosis knowledge, the fault diagnosis method of the present embodiment further includes:
s201, screening out fault diagnosis knowledge according to the state indication from the matched fault diagnosis knowledge.
The state indication is a quantitative indication for judging faults by analyzing the monitoring signals.
S202, judging whether the matched fault diagnosis models and fault diagnosis knowledge indicated according to the state meet preset conditions or not respectively.
If not, step S203 is executed, and if yes, step S204 is executed.
And step S203, deleting the fault diagnosis model which does not meet the preset condition and fault diagnosis knowledge indicated according to the state from the setting diagnosis list.
And step S204, reserving a fault diagnosis model meeting preset conditions and fault diagnosis knowledge according to the state indication.
The preset conditions comprise that first sensor related parameters of the aeroengine to be diagnosed are matched with second sensor related parameters required to be used by a fault diagnosis model or fault diagnosis knowledge.
Alternatively, the first sensor-related parameter may be sensor type, sensor mounting location information including, but not limited to, an aeroengine to be diagnosed, and the second sensor-related parameter may be sensor type, sensor mounting location information used including, but not limited to, a fault diagnosis model, fault diagnosis knowledge.
On the basis of matching the corresponding fault diagnosis model and the fault diagnosis knowledge according to the configuration information, the fault diagnosis model and the fault diagnosis knowledge which are matched according to the sensor are added again, so that the obtained fault diagnosis model and the fault diagnosis knowledge are more matched with the aeroengine to be diagnosed, and the fault diagnosis accuracy of the aeroengine to be diagnosed is improved.
Optionally, if the sensor type required in the matched fault diagnosis model is not available in the aeroengine to be diagnosed, deleting the fault diagnosis model from the set diagnosis list; if the judging result comprises one or more fault diagnosis models, the process is continued to run, and if the judging result does not comprise the fault diagnosis models, the rest steps are skipped, and the diagnosis result based on the fault diagnosis models is empty.
Optionally, if the matched sensor type required in the fault diagnosis knowledge according to the state indication is not available in the aeroengine to be diagnosed, deleting the fault diagnosis knowledge from the set diagnosis list; if the judging result contains one or more pieces of fault diagnosis knowledge, continuing to operate the process, and if the judging result does not contain fault diagnosis knowledge, skipping other steps, wherein the diagnosis result based on the fault diagnosis knowledge is empty.
Preferably, in the judging, the priorities of the sensors required by the fault diagnosis model or the priorities of the sensors required by the fault diagnosis knowledge according to the state indication can be further ordered, so that the judgment is started from the sensor with the highest priority, and the judgment is terminated once the preset condition is not met, so that the judgment efficiency is improved.
In an alternative embodiment, when there is a fault diagnosis model satisfying a preset condition, the fault diagnosis model is directly executed; when there are at least two fault diagnosis models satisfying the preset condition, step S103 includes:
s1031, respectively inputting the operation state parameters into a plurality of fault diagnosis models to obtain a plurality of first intermediate diagnosis results, fusing to obtain corresponding first fault diagnosis results, and storing the first fault diagnosis results in a fault diagnosis result library.
Optionally, when at least two fault diagnosis models meeting the preset conditions exist, feature extraction and state indication after feature extraction need to be counted for each fault diagnosis model, if the state indication after feature extraction of each fault diagnosis model is repeated, the state indication is de-duplicated, repeated calculation possibly caused in the state indication calculation process is reduced, and waste of calculation force and calculation time is reduced. After all the state instructions are calculated, the state instructions are stored, and are used as the process quantity of the fault diagnosis model, and when each fault diagnosis model is operated, the required state instructions are selected from the calculated state instructions to carry out subsequent calculation.
In an alternative embodiment, when there is a diagnosis knowledge according to the state indication that satisfies the preset condition, the inference based on the diagnosis knowledge is directly performed; when there are at least two fault diagnosis knowledge according to the status indication that satisfy the preset condition, step S104 includes:
s1041, respectively inputting the operation state parameters into a plurality of fault diagnosis knowledge according to the state indication to obtain a plurality of second intermediate diagnosis results, fusing to obtain corresponding second fault diagnosis results, and storing the second fault diagnosis results into a fault diagnosis result library.
Optionally, when at least two pieces of fault diagnosis knowledge meeting preset conditions and according to the state indication exist, statistics is needed to be carried out on the state indication which needs to be extracted from each piece of fault diagnosis knowledge, if the state indication after the feature extraction of each piece of fault diagnosis knowledge is repeated, the state indication is de-duplicated, repeated calculation possibly caused in the state indication calculation process is reduced, and the waste of calculation power and calculation time is reduced. After all the state instructions are calculated, the state instructions are stored, and are used as the fault diagnosis knowledge process quantity, and when each fault diagnosis knowledge is operated, a required state instruction is selected from the calculated state instructions to carry out subsequent calculation.
In an alternative embodiment, as shown in fig. 3, step S105 includes:
s1051, acquiring a corresponding first fault diagnosis result and a corresponding second fault diagnosis result from a fault diagnosis result library.
Optionally, fault diagnosis results of single operation data of the aeroengine to be diagnosed are obtained from a fault diagnosis result library, wherein the fault diagnosis results comprise a first fault diagnosis result and a second fault diagnosis result, the various fault diagnosis results comprise names of diagnosis models/used knowledge and corresponding results, if a plurality of engine fault types exist in the results, the corresponding probability of each fault type is provided, and the sum of the probabilities of the diagnosis results of each diagnosis model or each knowledge is 100%.
S1052, determining a corresponding weight relation according to the fault diagnosis model meeting the preset condition and the diagnosis accuracy of the configuration of the aeroengine to be diagnosed corresponding to the fault diagnosis knowledge indicated by the state.
The diagnosis accuracy is used as the weight of each fault diagnosis model and fault diagnosis knowledge to participate in the subsequent calculation, and it should be noted that the diagnosis accuracy refers to the confidence of the diagnosis result, and further, the confidence of the diagnosis model and the diagnosis knowledge is information stored in the model/knowledge and is updated according to the historical diagnosis result, if the diagnosis result of the diagnosis knowledge/diagnosis model is correct for 90 times in 100 diagnosis processes, the diagnosis accuracy is 90%, and in addition, the weight is the weight coefficient of the result of the diagnosis model/diagnosis knowledge set according to expert experience, and the weight coefficient of the same diagnosis model or the same piece of diagnosis knowledge is the same.
S1053, performing fusion diagnosis result calculation on the first fault diagnosis result, the second fault diagnosis result and the weight relation by using an information fusion algorithm to obtain a first target diagnosis result of the aeroengine to be diagnosed. The first target diagnosis results corresponding to the fault types of the aero-engine to be diagnosed are arranged according to the size of the fusion diagnosis results.
The fusion diagnosis result is obtained by calculating the diagnosis result (fault type and accuracy) obtained by the diagnosis model and the diagnosis knowledge through an information fusion algorithm.
Optionally, according to the obtained first fault diagnosis result and the obtained second fault diagnosis result and the weight relation, using a D-S theory (D-S evidence theory ) to calculate a fusion diagnosis result, and according to the fusion diagnosis result, arranging possible fault types from large to small as the fusion diagnosis result, and taking the possible fault types as the fault diagnosis result of the operation of the aeroengine and storing the fault diagnosis result. In addition, the information fusion algorithm can also adopt information fusion algorithm based on Bayesian statistical theory, information fusion algorithm based on neural network theory and information fusion algorithm based on fuzzy theory.
In another alternative implementation manner, as shown in fig. 4, the fault diagnosis method of the present embodiment further includes:
s301, acquiring phenomenon description information corresponding to at least one condition of an aeroengine to be diagnosed before operation, after operation and in the operation process.
The phenomenon description information comprises at least one of field personnel description, nondestructive inspection results and hole detection result description.
S302, screening out fault diagnosis knowledge according to the phenomenon description from the matched fault diagnosis knowledge.
S303, inputting the phenomenon description to carry out knowledge reasoning according to the fault diagnosis knowledge of the phenomenon description so as to obtain a third fault diagnosis result.
If the result accords with the reasoning result of the existing phenomenon description, the reasoning result is null according to the knowledge of the phenomenon.
After step S303, the fault diagnosis method of the present embodiment further includes:
s304, carrying out fusion processing on the first fault diagnosis result, the second fault diagnosis result and the third fault diagnosis result to obtain a second target diagnosis result of the aeroengine to be diagnosed.
And the diagnosis results obtained in three different modes of the first fault diagnosis result, the second fault diagnosis result and the third fault diagnosis result are fused, so that the target diagnosis result of the aeroengine to be diagnosed is more accurate.
In an alternative embodiment, as shown in fig. 5, step S101 includes:
s101a, acquiring and detecting a data format of original operation parameters of the aero-engine to be diagnosed.
Alternatively, the original operating parameters of the aircraft engine may be collected by a signal collection device, stored and detected in data format in the form of a real-time data stream or an offline data file.
And S101b, matching in a preset data analysis algorithm library according to the data format to obtain a corresponding data analysis algorithm.
S101c, acquiring the operation state parameters of the aero-engine to be diagnosed based on the original operation parameters and a data analysis algorithm.
And matching the detected data format with a data analysis algorithm in a preset data analysis algorithm library, carrying out data analysis after matching the data analysis algorithm meeting the data format, and sending and storing the analyzed data.
In an alternative implementation, as shown in fig. 6, after step S101, the fault diagnosis method of the present embodiment further includes:
s1011, judging whether the running state parameters meet the preset quality evaluation requirements; wherein the data quality assessment requirements include at least one of a missing value, an outlier, a redundancy value, a unit criterion, and a data type criterion.
If yes, step S1012 is executed, and if not, step S1013 is executed.
Step S1012, outputting operation state parameters;
step S1013, deleting the operation state parameter.
Specifically, the data quality evaluation is performed on the analyzed running state parameters, so that whether the problems of missing values, outliers, redundant values, unit non-standards, data type non-standards and the like exist in the running state parameters is evaluated, noise data in the running state parameters is reduced, and fault diagnosis accuracy is improved.
In addition, for the parameters deleted in step S1013, that is, the data that do not meet the preset quality evaluation requirement, a data preprocessing algorithm that can solve the corresponding data quality problem and can be applied to the corresponding data format may be matched, and the obtained corrected data is saved as an operation state parameter, so as to avoid leaving the operation state parameter that can be used for diagnosing the engine fault.
The implementation principle of the data driving algorithm-based aeroengine fault diagnosis of the embodiment is specifically described below with reference to an example:
referring to fig. 7 and 8, firstly, the signal acquisition end is used for acquiring the operation data of the aeroengine, and the operation data is transmitted to the data transceiver at the server end in the form of a real-time data stream or an off-line data file. The data receiving and transmitting device transmits the received real-time data stream and the offline file to the data preprocessing device, and the data preprocessing device reads the real-time data stream or the offline file and detects the data format. And matching the detected data format with a data analysis algorithm in the data storage device at the server side, after matching the data analysis algorithm meeting the data format, carrying out data analysis by using the matched data analysis algorithm in the data preprocessing device at the server side, and sending the analyzed data to the data storage device at the server side as initial data for storage.
And carrying out data quality evaluation on the analyzed initial data in a data preprocessing device. Step five: according to the data quality evaluation result in the server-side data preprocessing device, matching can solve the corresponding data quality problem in the server-side data storage device and can be applied to the data preprocessing algorithm of the corresponding data format. And running the matched data preprocessing algorithm in the server-side data preprocessing device, and sending the obtained data to a database of the server-side data storage device to be saved as original data.
Matching the configuration of the aeroengine suitable for each fault diagnosis model in the data storage device at the server side with the configuration of the aeroengine to be diagnosed, and if the matching result comprises one or more fault diagnosis models based on a data driving algorithm, continuing to operate the flow; if no matching result exists, the rest steps are skipped, and the diagnosis result of the fault diagnosis model based on the data driving algorithm is null.
And matching according to the type and the position of the sensor required by the successfully matched fault diagnosis model based on the data driving algorithm and the type and the position information of the sensor possessed by the aeroengine to be diagnosed, carrying out the priority ranking of the sensor required by the diagnosis model, and deleting the diagnosis model from the successfully matched fault diagnosis model list if the sensor required by the successfully matched fault diagnosis model is not possessed by the aeroengine to be diagnosed. If the matching result comprises one or more fault diagnosis models based on the data driving algorithm, continuing to run the flow; if no matching result exists, the rest steps are skipped, and the diagnosis result of the fault diagnosis model based on the data driving algorithm is null.
And D, according to the successfully matched fault diagnosis model based on the data driving algorithm, sending the fault diagnosis model to a data analysis device at a server side, and diagnosing the original data generated in the step five. When only one fault diagnosis model based on the data driving algorithm is successfully matched, directly executing the diagnosis model; when two or more fault diagnosis models based on the data driving algorithm are successfully matched, statistics is needed to be carried out on whether the CI (Condition Indicator, state indication) subjected to feature extraction is needed in each diagnosis model, if the CI subjected to feature extraction of each diagnosis model is repeated, the repetition is carried out on the CI, all the CIs are calculated, and then the CI is stored in a server-side data storage device. And taking the CI as the process quantity of the diagnostic model, and selecting a required CI from the calculated CI to carry out subsequent calculation when each diagnostic model is operated.
And finally, obtaining diagnosis results of the fault diagnosis models based on the data driving algorithm, and storing the diagnosis results and the diagnosis models which are correspondingly used in a data storage device at a server side for later use in the fusion process of the diagnosis results.
The implementation principle of the aeroengine fault diagnosis based on knowledge reasoning of the embodiment is specifically described below with reference to examples:
Referring to fig. 7 and 9, firstly, the signal acquisition end is used for acquiring the operation data of the aeroengine, and the operation data is transmitted to the data transceiver at the server end in the form of a real-time data stream or an off-line data file.
The data receiving and transmitting device transmits the received real-time data stream and the offline file to the data preprocessing device, and the data preprocessing device reads the real-time data stream or the offline file and detects the data format.
And matching the detected data format with a data analysis algorithm in the data storage device at the server side, after matching the data analysis algorithm meeting the data format, carrying out data analysis by using the matched data analysis algorithm in the data preprocessing device at the server side, and sending the analyzed data to the data storage device at the server side as initial data for storage.
And carrying out data quality evaluation on the analyzed initial data in a data preprocessing device.
According to the data quality evaluation result in the server-side data preprocessing device, matching can solve the corresponding data quality problem in the server-side data storage device and can be applied to the data preprocessing algorithm of the corresponding data format. And running the matched data preprocessing algorithm in the server-side data preprocessing device, and sending the obtained data to a database of the server-side data storage device to be saved as original data.
Matching the configuration of the aero-engine suitable for each knowledge in the knowledge base in the data storage device at the server side with the configuration of the aero-engine to be diagnosed, and if the matching result contains one or more pieces of diagnosis knowledge, continuing to operate; if the matching result does not have the matching result, the rest steps are skipped, and the knowledge reasoning diagnosis result is directly output to be null.
And selecting the knowledge according to CI from the knowledge of successful matching. And matching according to the type and the position of the sensor required to be used according to the CI knowledge which is successfully matched and the type and the position information of the sensor which is possessed by the aeroengine to be diagnosed, carrying out the priority ordering of the sensor required by the CI knowledge, and deleting the knowledge from the CI-based knowledge list which is successfully matched if the sensor required by the CI-based knowledge which is successfully matched is not possessed by the aeroengine to be diagnosed. If the matching result contains one or more pieces of diagnosis knowledge, the operation can be continued; if the matching result does not have the matching result, the rest steps are skipped, and the knowledge reasoning diagnosis result is directly output to be null.
And D, sending the original data to a server-side knowledge reasoning device for reasoning according to the knowledge of the CI successfully matched and the original data in the step five. When the matching is successful, only one piece of knowledge based on CI exists, the reasoning based on the rule is directly executed; when the matching is successful, two or more pieces of knowledge according to the CI are needed, statistics is carried out on the CI needing to be extracted in each piece of knowledge, and if the CI extracted by the characteristics of each diagnosis model is repeated, the repetition is carried out on the CI. After all CIs are calculated, the CIs are stored in a server-side data storage device. And taking the CI as the process quantity of the diagnostic model, and selecting a required CI from the calculated CI to carry out subsequent calculation when each diagnostic model is operated.
And finally, obtaining diagnosis results according to the knowledge of each CI, and storing the diagnosis results and the knowledge used correspondingly into a data storage device at a server side for later use in the fusion process of the diagnosis results.
The following specifically describes the implementation principle of the aeroengine fault diagnosis based on knowledge reasoning of the phenomenon description in this embodiment with reference to examples:
referring to fig. 7 and 9, first, the phenomenon description information is collected before, after and during the operation of the aeroengine through the browser, and the phenomenon description information is sent to the data transceiver module of the server, and is sent to the data storage device for storage through the data transceiver module.
And selecting knowledge according to the phenomenon from the knowledge of successful matching. And sending the existing phenomenon description and knowledge according to the phenomenon to a knowledge reasoning device at a server side to perform reasoning according to the phenomenon.
Finally, the diagnosis results of the knowledge according to the phenomena are obtained, the diagnosis results and the knowledge used correspondingly are stored in a data storage device at the server side for use in the subsequent diagnosis result fusion process, and the diagnosis results and the knowledge used correspondingly are sent to the data storage device at the server side. If the result of reasoning according with the existing phenomenon description is met, the result of reasoning is null according to the knowledge of the phenomenon.
According to the method, the device and the system, the running state parameters and configuration information of the aero-engine to be diagnosed are obtained, the corresponding fault diagnosis model and fault diagnosis knowledge are matched according to the configuration information, the corresponding fault diagnosis model and fault diagnosis knowledge are matched according to the sensor type, fusion processing is carried out on the basis of two different fault diagnosis results to obtain the target diagnosis result, and a universal fault diagnosis flow is provided for the aero-engine with different structural schemes and different sensor schemes, so that the accuracy and the interpretability of the fault diagnosis result of the aero-engine are improved.
Example 3
The present embodiment provides a fault diagnosis system of an aeroengine, as shown in fig. 10, the fault diagnosis system of the present embodiment includes:
the first acquisition module 1 is used for acquiring operation state parameters and configuration information of the aeroengine to be diagnosed.
The to-be-diagnosed operation state parameters of the aero-engine can comprise at least one of a gas circuit system parameter, a fuel system parameter, a lubricating oil system parameter, a vibration parameter, a state information parameter and a flight parameter of an aero-engine control system of the aero-engine; configuration information of the aircraft engine to be diagnosed includes, but is not limited to, at least one of a double rotor, a triple rotor, an interposed bearing, an unbiased bearing, and the like. The above is by way of example and is not intended to limit the operating state parameters and configuration information of the aircraft engine to be diagnosed.
The matching module 2 is used for respectively matching in a preset fault diagnosis model library and a preset fault diagnosis knowledge library according to the configuration information to obtain a corresponding fault diagnosis model and fault diagnosis knowledge; the preset fault diagnosis model library is a fault diagnosis model library of the aeroengine generated based on a data driving algorithm, and the preset fault diagnosis knowledge base is a fault diagnosis knowledge base of the aeroengine generated based on a knowledge reasoning algorithm.
Optionally, matching can be performed in a preset fault diagnosis model library and a preset fault diagnosis knowledge base according to configuration information of the aeroengine to be diagnosed, if the matching result of the fault diagnosis model comprises one or more fault diagnosis models, continuing to operate, if no corresponding matching result exists, skipping other steps, and if the diagnosis result based on the fault diagnosis model is null; if the fault diagnosis knowledge matching result contains one or more diagnosis knowledge, the operation can be continued; if no corresponding matching result exists, the rest steps are skipped, and the diagnosis result based on the fault diagnosis knowledge is null.
The fault diagnosis model library based on the data driving algorithm comprises at least one of a statistical analysis model, a machine learning model (including but not limited to a decision tree, a support vector machine, a random forest and other machine learning algorithm models) and a deep learning model (including but not limited to a cyclic neural network, a convolutional neural network and other deep learning algorithm models); the fault diagnosis knowledge base based on the knowledge reasoning algorithm comprises at least one of diagnosis rules, basic facts and other information. The foregoing is illustrative, and the fault diagnosis model library and the fault diagnosis knowledge library are not limited thereto.
A first fault diagnosis result acquisition module 3, configured to acquire a first fault diagnosis result based on the operation state parameter and the fault diagnosis model.
Optionally, the operation state parameters of the aeroengine to be diagnosed are input into the matched fault diagnosis model to obtain a first fault diagnosis result.
And a second fault diagnosis result acquisition module 4, configured to acquire a second fault diagnosis result based on the operating state parameter and the fault diagnosis knowledge.
Optionally, knowledge reasoning is performed on the operation state parameters of the aeroengine to be diagnosed in combination with the fault diagnosis knowledge, so as to obtain a second fault diagnosis result.
The first diagnosis result fusion module 5 is configured to fuse the first fault diagnosis result and the second fault diagnosis result, and obtain a first target diagnosis result of the aeroengine to be diagnosed.
According to the method, the operation state parameters and configuration information of the aero-engine to be diagnosed are obtained through the first obtaining module, the corresponding fault diagnosis model and fault diagnosis knowledge are matched through the matching module according to the configuration information, the corresponding fault diagnosis result is obtained according to the fault diagnosis model and fault diagnosis knowledge which are successfully matched, the corresponding fault diagnosis result is obtained through the first fault diagnosis result obtaining module and the second fault diagnosis result obtaining module, and the target diagnosis result is obtained through fusion processing of the first diagnosis result fusion module, so that a universal fault diagnosis flow is provided for the aero-engines with different configuration schemes and different sensor schemes, and therefore the universality of a high aero-engine fault diagnosis system is improved, and the accuracy, reliability and interpretability of the aero-engine fault diagnosis result are improved.
Example 4
The present embodiment provides a fault diagnosis system for an aeroengine, as shown in fig. 11, which is a further improvement of embodiment 3, specifically:
the fault diagnosis system of the present embodiment further includes:
and the first screening module 6 is used for screening out fault diagnosis knowledge according to the state indication from the matched fault diagnosis knowledge.
The state indication is a quantitative indication for judging faults by analyzing the monitoring signals.
The first judging module 7 is configured to respectively judge whether the matched fault diagnosis model and the fault diagnosis knowledge indicated according to the state meet the preset condition.
If not, deleting the fault diagnosis model which does not meet the preset conditions and fault diagnosis knowledge indicated according to the state from the setting diagnosis list. If yes, a fault diagnosis model meeting preset conditions and fault diagnosis knowledge according to state indication are reserved.
The preset conditions comprise that first sensor related parameters of the aeroengine to be diagnosed are matched with second sensor related parameters required to be used by a fault diagnosis model or fault diagnosis knowledge.
Alternatively, the first sensor-related parameter may be sensor type, sensor mounting location information including, but not limited to, an aeroengine to be diagnosed, and the second sensor-related parameter may be sensor type, sensor mounting location information used including, but not limited to, a fault diagnosis model, fault diagnosis knowledge.
On the basis of matching the corresponding fault diagnosis model and the fault diagnosis knowledge according to the configuration information, the fault diagnosis model and the fault diagnosis knowledge which are matched according to the sensor are added again, so that the obtained fault diagnosis model and the fault diagnosis knowledge are more matched with the aeroengine to be diagnosed, and the fault diagnosis accuracy of the aeroengine to be diagnosed is improved.
Optionally, if the sensor type required in the matched fault diagnosis model is not available in the aeroengine to be diagnosed, deleting the fault diagnosis model from the set diagnosis list; if the judging result comprises one or more fault diagnosis models, the process is continued to run, and if the judging result does not comprise the fault diagnosis models, the rest steps are skipped, and the diagnosis result based on the fault diagnosis models is empty.
Optionally, if the matched sensor type required in the fault diagnosis knowledge according to the state indication is not available in the aeroengine to be diagnosed, deleting the fault diagnosis knowledge from the set diagnosis list; if the judging result contains one or more pieces of fault diagnosis knowledge, continuing to operate the process, and if the judging result does not contain fault diagnosis knowledge, skipping other steps, wherein the diagnosis result based on the fault diagnosis knowledge is empty.
Preferably, in the judging, the priorities of the sensors required by the fault diagnosis model or the priorities of the sensors required by the fault diagnosis knowledge according to the state indication can be further ordered, so that the judgment is started from the sensor with the highest priority, and the judgment is terminated once the preset condition is not met, so that the judgment efficiency is improved.
In an alternative embodiment, when there is a fault diagnosis model satisfying a preset condition, the fault diagnosis model is directly executed; when at least two fault diagnosis models meeting the preset conditions exist, the first fault diagnosis result obtaining module 3 is further configured to input the operation state parameters into a plurality of fault diagnosis models to obtain a plurality of first intermediate diagnosis results, fuse the first intermediate diagnosis results to obtain corresponding first fault diagnosis results, and store the first fault diagnosis results in a fault diagnosis result library.
Optionally, when at least two fault diagnosis models meeting the preset conditions exist, feature extraction and state indication after feature extraction need to be counted for each fault diagnosis model, if the state indication after feature extraction of each fault diagnosis model is repeated, the state indication is de-duplicated, repeated calculation possibly caused in the state indication calculation process is reduced, and waste of calculation force and calculation time is reduced. After all the state instructions are calculated, the state instructions are stored, and are used as the process quantity of the fault diagnosis model, and when each fault diagnosis model is operated, the required state instructions are selected from the calculated state instructions to carry out subsequent calculation.
In an alternative embodiment, when there is a diagnosis knowledge according to the state indication that satisfies the preset condition, the inference based on the diagnosis knowledge is directly performed; when there are at least two fault diagnosis knowledge according to the state indication that meets the preset condition, the second fault diagnosis result obtaining module 4 is further configured to input the operation state parameter into a plurality of fault diagnosis knowledge according to the state indication to obtain a plurality of second intermediate diagnosis results, fuse the second intermediate diagnosis results to obtain corresponding second fault diagnosis results, and store the second fault diagnosis results in the fault diagnosis result library.
Optionally, when at least two pieces of fault diagnosis knowledge meeting preset conditions and according to the state indication exist, statistics is needed to be carried out on the state indication which needs to be extracted from each piece of fault diagnosis knowledge, if the state indication after the feature extraction of each piece of fault diagnosis knowledge is repeated, the state indication is de-duplicated, repeated calculation possibly caused in the state indication calculation process is reduced, and the waste of calculation power and calculation time is reduced. After all the state instructions are calculated, the state instructions are stored, and are used as the fault diagnosis knowledge process quantity, and when each fault diagnosis knowledge is operated, a required state instruction is selected from the calculated state instructions to carry out subsequent calculation.
In an alternative implementation manner, as shown in fig. 12, the first diagnostic result fusion module 5 of the present embodiment includes:
the diagnosis result obtaining unit 51 is configured to obtain a corresponding first diagnosis result and a corresponding second diagnosis result from the diagnosis result library.
Optionally, fault diagnosis results of single operation data of the aeroengine to be diagnosed are obtained from a fault diagnosis result library, wherein the fault diagnosis results comprise a first fault diagnosis result and a second fault diagnosis result, the various fault diagnosis results comprise names of diagnosis models/used knowledge and corresponding results, if a plurality of engine fault types exist in the results, the corresponding probability of each fault type is provided, and the sum of the probabilities of the diagnosis results of each diagnosis model or each knowledge is 100%.
The weight determining unit 52 is configured to determine a corresponding weight relationship according to a fault diagnosis model that satisfies a preset condition and a diagnosis accuracy corresponding to a configuration of the aeroengine to be diagnosed according to the fault diagnosis knowledge indicated by the status.
The diagnosis accuracy is used as the weight of each fault diagnosis model and fault diagnosis knowledge to participate in the subsequent calculation, and it should be noted that the diagnosis accuracy refers to the confidence of the diagnosis result, and further, the confidence of the diagnosis model and the diagnosis knowledge is information stored in the model/knowledge and is updated according to the historical diagnosis result, if the diagnosis result of the diagnosis knowledge/diagnosis model is correct for 90 times in 100 diagnosis processes, the diagnosis accuracy is 90%, and in addition, the weight is the weight coefficient of the result of the diagnosis model/diagnosis knowledge set according to expert experience, and the weight coefficient of the same diagnosis model or the same piece of diagnosis knowledge is the same.
A fusion diagnosis result calculation unit 53, configured to calculate a fusion diagnosis result by using an information fusion algorithm from the first fault diagnosis result, the second fault diagnosis result, and the weight relationship, so as to obtain a first target diagnosis result of the aero-engine to be diagnosed; the first target diagnosis results corresponding to the fault types of the aero-engine to be diagnosed are arranged according to the size of the fusion diagnosis results.
The fusion diagnosis result is obtained by calculating the diagnosis result (fault type and accuracy) obtained by the diagnosis model and the diagnosis knowledge through an information fusion algorithm.
Optionally, calculating a fusion diagnosis result by using a D-S theory according to the acquired first fault diagnosis result, the second fault diagnosis result and the weight relation, and arranging possible fault types from large to small according to the fusion diagnosis result as a fusion diagnosis result, and storing the possible fault types as the fault diagnosis result of the operation of the aeroengine. In addition, the information fusion algorithm can also adopt information fusion algorithm based on Bayesian statistical theory, information fusion algorithm based on neural network theory and information fusion algorithm based on fuzzy theory.
In another alternative implementation manner, the fault diagnosis system of the present embodiment further includes:
the second obtaining module 8 is configured to obtain phenomenon description information corresponding to at least one of before, after, and during operation of the aeroengine to be diagnosed.
The phenomenon description information comprises at least one of field personnel description, nondestructive inspection results and hole detection result description.
And the second screening module 9 is used for screening out fault diagnosis knowledge described according to the phenomenon from the matched fault diagnosis knowledge.
The third fault diagnosis result obtaining module 10 is configured to input the phenomenon description into the knowledge reasoning according to the fault diagnosis knowledge of the phenomenon description, so as to obtain a third fault diagnosis result.
If the result accords with the reasoning result of the existing phenomenon description, the reasoning result is null according to the knowledge of the phenomenon.
In an alternative implementation manner, the fault diagnosis system of the present embodiment further includes:
and the second diagnosis result fusion module 11 is configured to fuse the first diagnosis result, the second diagnosis result, and the third diagnosis result to obtain a second target diagnosis result of the aeroengine to be diagnosed.
And the diagnosis results obtained in three different modes of the first fault diagnosis result, the second fault diagnosis result and the third fault diagnosis result are fused, so that the target diagnosis result of the aeroengine to be diagnosed is more accurate.
In an alternative embodiment, as shown in fig. 13, the first acquisition module 1 includes:
the primary operation parameter acquisition unit 101 is used for acquiring and detecting a data format of primary operation parameters of the aeroengine to be diagnosed.
Alternatively, the original operating parameters of the aircraft engine may be collected by a signal collection device, stored and detected in data format in the form of a real-time data stream or an offline data file.
The data analysis algorithm matching unit 102 is configured to match in a preset data analysis algorithm library according to a data format to obtain a corresponding data analysis algorithm.
An operation state parameter obtaining unit 103, configured to obtain an operation state parameter of the aeroengine to be diagnosed based on the original operation parameter and the data analysis algorithm.
And matching the detected data format with a data analysis algorithm in a preset data analysis algorithm library, carrying out data analysis after matching the data analysis algorithm meeting the data format, and sending and storing the analyzed data.
In an alternative embodiment, the fault diagnosis system further includes:
a parameter quality judging module 12 for judging whether the running state parameter meets the preset quality evaluation requirement; wherein the data quality assessment requirements include at least one of a missing value, an outlier, a redundancy value, a unit criterion, and a data type criterion.
And if so, outputting the running state parameters.
If not, deleting the running state parameters.
Specifically, the data quality evaluation is performed on the analyzed running state parameters, so that whether the problems of missing values, outliers, redundant values, unit non-standards, data type non-standards and the like exist in the running state parameters is evaluated, noise data in the running state parameters is reduced, and fault diagnosis accuracy is improved.
In addition, for the deleted running state parameters, namely the data which does not meet the preset quality evaluation requirement, a data preprocessing algorithm which can solve the corresponding data quality problem and can be applied to the corresponding data format can be matched, and the obtained corrected data is stored as the running state parameters so as to avoid leaving the running state parameters which can be used for diagnosing the engine faults.
According to the method, the device and the system, the running state parameters and configuration information of the aero-engine to be diagnosed are obtained, the corresponding fault diagnosis model and fault diagnosis knowledge are matched according to the configuration information, the corresponding fault diagnosis model and fault diagnosis knowledge are matched according to the sensor type, and the fusion processing is carried out on the basis of two different fault diagnosis results to obtain the target diagnosis result, so that a universal fault diagnosis flow is provided for the aero-engine with different structural schemes and different sensor schemes, and therefore the universality of a high-aero-engine fault diagnosis system is improved, and the accuracy, reliability and interpretability of the aero-engine fault diagnosis result are improved.
Example 5
Fig. 14 is a schematic structural diagram of an electronic device according to embodiment 5 of the present invention. The electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the fault diagnosis method of the aeroengine of embodiment 1 or 2 when executing the program. The electronic device 30 shown in fig. 14 is merely an example, and should not be construed as limiting the functionality and scope of use of embodiments of the present invention.
As shown in fig. 14, the electronic device 30 may be embodied in the form of a general purpose computing device, which may be a server device, for example. Components of electronic device 30 may include, but are not limited to: the at least one processor 31, the at least one memory 32, a bus 33 connecting the different system components, including the memory 32 and the processor 31.
The bus 33 includes a data bus, an address bus, and a control bus.
Memory 32 may include volatile memory such as Random Access Memory (RAM) 321 and/or cache memory 322, and may further include Read Only Memory (ROM) 323.
Memory 32 may also include a program/utility 325 having a set (at least one) of program modules 324, such program modules 324 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.
The processor 31 executes various functional applications and data processing, such as the fault diagnosis method of the aeroengine of embodiment 1 or 2 of the present invention, by running a computer program stored in the memory 32.
The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). Such communication may be through an input/output (I/O) interface 35. Also, model-generating device 30 may also communicate with one or more networks, such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet, via network adapter 36. As shown in fig. 14, network adapter 36 communicates with the other modules of model-generating device 30 via bus 33. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in connection with the model-generating device 30, including, but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, data backup storage systems, and the like.
It should be noted that although several units/modules or sub-units/modules of an electronic device are mentioned in the above detailed description, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more units/modules described above may be embodied in one unit/module in accordance with embodiments of the present invention. Conversely, the features and functions of one unit/module described above may be further divided into ones that are embodied by a plurality of units/modules.
Example 6
The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps in the fault diagnosis method of the aeroengine of embodiment 1 or 2.
More specifically, among others, readable storage media may be employed including, but not limited to: portable disk, hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.
In a possible embodiment, the invention may also be realized in the form of a program product comprising program code for causing a terminal device to carry out the steps of the fault diagnosis method of an aeroengine implementing embodiment 1 or 2, when the program product is run on the terminal device.
Wherein the program code for carrying out the invention may be written in any combination of one or more programming languages, the program code may execute entirely on the user device, partly on the user device, as a stand-alone software package, partly on the user device, partly on a remote device or entirely on the remote device.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the principles and spirit of the invention, but such changes and modifications fall within the scope of the invention.
Claims (12)
1. A method of diagnosing a fault in an aircraft engine, the method comprising:
acquiring operation state parameters and configuration information of an aeroengine to be diagnosed;
according to the configuration information, matching in a preset fault diagnosis model library and a preset fault diagnosis knowledge library respectively to obtain a corresponding fault diagnosis model and fault diagnosis knowledge; the preset fault diagnosis model library is a fault diagnosis model library of the aeroengine generated based on a data driving algorithm, and the preset fault diagnosis knowledge library is a fault diagnosis knowledge library of the aeroengine generated based on a knowledge reasoning algorithm;
acquiring a first fault diagnosis result based on the running state parameter and the fault diagnosis model;
Acquiring a second fault diagnosis result based on the operating state parameters and the fault diagnosis knowledge;
and carrying out fusion processing on the first fault diagnosis result and the second fault diagnosis result to obtain a first target diagnosis result of the aero-engine to be diagnosed.
2. The method for diagnosing a fault in an aircraft engine according to claim 1, wherein after matching the corresponding fault diagnosis model and the fault diagnosis knowledge, the method for diagnosing a fault further comprises:
screening fault diagnosis knowledge according to the state indication from the matched fault diagnosis knowledge;
respectively judging whether the matched fault diagnosis model and the fault diagnosis knowledge indicated according to the state meet preset conditions or not;
if not, deleting the fault diagnosis model which does not meet the preset conditions and fault diagnosis knowledge indicated according to the state from a set diagnosis list;
the preset conditions comprise that first sensor related parameters of the aeroengine to be diagnosed are matched with second sensor related parameters required to be used by the fault diagnosis model or the fault diagnosis knowledge.
3. The method for diagnosing a fault in an aircraft engine according to claim 2, wherein when there are at least two fault diagnosis models satisfying the preset condition, the obtaining a first fault diagnosis result based on the operating state parameter and the fault diagnosis model includes:
And respectively inputting the running state parameters into a plurality of fault diagnosis models to obtain a plurality of first intermediate diagnosis results, fusing to obtain the corresponding first fault diagnosis results, and storing the first fault diagnosis results into a fault diagnosis result library.
4. A method of diagnosing a fault in an aircraft engine as claimed in claim 3, wherein said obtaining a second fault diagnosis based on said operating condition parameters and said fault diagnosis knowledge when there are at least two fault diagnosis knowledge dependent on a condition indication that satisfies said preset condition comprises:
and respectively inputting the operation state parameters into a plurality of fault diagnosis knowledge according to the state indication to obtain a plurality of second intermediate diagnosis results, fusing to obtain the corresponding second fault diagnosis results, and storing the second fault diagnosis results into the fault diagnosis result library.
5. The method for diagnosing a fault in an aircraft engine according to claim 4, wherein the fusing the first fault diagnosis result and the second fault diagnosis result to obtain a first target diagnosis result of the aircraft engine to be diagnosed comprises:
acquiring a corresponding first fault diagnosis result and a corresponding second fault diagnosis result from the fault diagnosis result library;
Determining a corresponding weight relation according to a fault diagnosis model meeting the preset conditions and the diagnosis accuracy of the configuration of the aeroengine to be diagnosed corresponding to the fault diagnosis knowledge indicated by the state;
carrying out fusion diagnosis result calculation on the first fault diagnosis result, the second fault diagnosis result and the weight relation by using an information fusion algorithm so as to obtain a first target diagnosis result of the aero-engine to be diagnosed;
the first target diagnosis results corresponding to the fault types of the aero-engine to be diagnosed are arranged according to the size of the fusion diagnosis results.
6. The method for diagnosing a fault in an aircraft engine according to claim 5, further comprising:
acquiring phenomenon description information corresponding to at least one condition of the aero-engine to be diagnosed before operation, after operation and in the operation process;
the phenomenon description information comprises at least one of field personnel description, nondestructive inspection results and hole detection result description;
screening fault diagnosis knowledge according to the phenomenon description from the matched fault diagnosis knowledge;
And inputting the phenomenon description into the fault diagnosis knowledge according to the phenomenon description to carry out knowledge reasoning so as to obtain a third fault diagnosis result.
7. The method for diagnosing a fault in an aircraft engine as recited in claim 6, wherein the method for diagnosing a fault further comprises:
and carrying out fusion processing on the first fault diagnosis result, the second fault diagnosis result and the third fault diagnosis result to obtain a second target diagnosis result of the aeroengine to be diagnosed.
8. The method for diagnosing faults of an aircraft engine according to any of claims 1 to 7, wherein the obtaining of operating state parameters of the aircraft engine to be diagnosed comprises:
collecting and detecting a data format of original operation parameters of the aeroengine to be diagnosed;
matching in a preset data analysis algorithm library according to the data format to obtain a corresponding data analysis algorithm;
and acquiring the operation state parameters of the aero-engine to be diagnosed based on the original operation parameters and the data analysis algorithm.
9. The method for diagnosing a fault in an aircraft engine as recited in claim 8, wherein said method for diagnosing a fault further comprises:
Judging whether the running state parameters meet the preset quality evaluation requirements or not; the data quality evaluation requirement comprises at least one of a missing value, an outlier, a redundancy value, a unit standard and a data type standard;
if yes, outputting the running state parameters;
and if not, deleting the running state parameters.
10. A fault diagnosis system for an aircraft engine, the fault diagnosis system comprising:
the first acquisition module is used for acquiring running state parameters and configuration information of the aeroengine to be diagnosed;
the matching module is used for respectively matching in a preset fault diagnosis model library and a preset fault diagnosis knowledge library according to the configuration information to obtain a corresponding fault diagnosis model and fault diagnosis knowledge; the preset fault diagnosis model library is a fault diagnosis model library of the aeroengine generated based on a data driving algorithm, and the preset fault diagnosis knowledge library is a fault diagnosis knowledge library of the aeroengine generated based on a knowledge reasoning algorithm;
the first fault diagnosis result acquisition module is used for acquiring a first fault diagnosis result based on the running state parameters and the fault diagnosis model;
The second fault diagnosis result acquisition module is used for acquiring a second fault diagnosis result based on the running state parameters and the fault diagnosis knowledge;
and the first diagnosis result fusion module is used for carrying out fusion processing on the first fault diagnosis result and the second fault diagnosis result to obtain a first target diagnosis result of the aeroengine to be diagnosed.
11. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method for fault diagnosis of an aeroengine as claimed in any of claims 1-9 when executing the computer program.
12. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements a method for fault diagnosis of an aircraft engine according to any one of claims 1-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210784395.3A CN117367807A (en) | 2022-06-28 | 2022-06-28 | Method, system, equipment and medium for diagnosing faults of aero-engine |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210784395.3A CN117367807A (en) | 2022-06-28 | 2022-06-28 | Method, system, equipment and medium for diagnosing faults of aero-engine |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117367807A true CN117367807A (en) | 2024-01-09 |
Family
ID=89391662
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210784395.3A Pending CN117367807A (en) | 2022-06-28 | 2022-06-28 | Method, system, equipment and medium for diagnosing faults of aero-engine |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117367807A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118135143A (en) * | 2024-05-07 | 2024-06-04 | 成都市技师学院(成都工贸职业技术学院、成都市高级技工学校、成都铁路工程学校) | AR-based aeroengine maintenance modeling method |
-
2022
- 2022-06-28 CN CN202210784395.3A patent/CN117367807A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118135143A (en) * | 2024-05-07 | 2024-06-04 | 成都市技师学院(成都工贸职业技术学院、成都市高级技工学校、成都铁路工程学校) | AR-based aeroengine maintenance modeling method |
CN118135143B (en) * | 2024-05-07 | 2024-07-23 | 成都市技师学院(成都工贸职业技术学院、成都市高级技工学校、成都铁路工程学校) | AR-based aeroengine maintenance modeling method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6622264B1 (en) | Process and system for analyzing fault log data from a machine so as to identify faults predictive of machine failures | |
US6543007B1 (en) | Process and system for configuring repair codes for diagnostics of machine malfunctions | |
US8001423B2 (en) | Prognostic diagnostic capability tracking system | |
US6834256B2 (en) | Method and system for determining motor reliability | |
US7873581B2 (en) | Method and system for determining the reliability of a DC motor system | |
EP2323005B1 (en) | Method and system for making diagnosis and prognosis of the status of complex systems via health indicators | |
US20050096759A1 (en) | Distributed power generation plant automated event assessment and mitigation plan determination process | |
US20080177683A1 (en) | Method and apparatus for mobile intelligence | |
US20230013544A1 (en) | Method, Apparatus and System for Detecting Abnormal Operating States of a Device | |
JP2009053938A (en) | Equipment diagnosing system and equipment-diagnosing method on the basis of multiple model | |
CN112631240A (en) | Spacecraft fault active detection method and device | |
JP2011529220A (en) | Method and apparatus for acquiring vehicle data | |
KR102102346B1 (en) | System and method for condition based maintenance support of naval ship equipment | |
CN113931258B (en) | Self-diagnosis method and non-negative pressure-superposed water supply equipment | |
US7996337B2 (en) | Method and system for remotely determining the reliability of a DC motor system | |
CN112697267A (en) | Abnormal vibration detection device for industrial equipment | |
CN113411229A (en) | Data processing method, playback data acquisition device and movable platform | |
CN117367807A (en) | Method, system, equipment and medium for diagnosing faults of aero-engine | |
CN114139274A (en) | Health management system | |
CN114779747A (en) | Vehicle fault cause determination system and method | |
CN113919207A (en) | Top-level open type electrical intelligent health monitoring and management system | |
CN117592975A (en) | Operation and maintenance decision processing method and system for electromechanical equipment of expressway based on cloud computing | |
JP2023536677A (en) | Vehicle level failure prediction for improved maintenance of vehicles | |
KR102108975B1 (en) | Apparatus and method for condition based maintenance support of naval ship equipment | |
CN115599077B (en) | Vehicle fault delimiting method and device, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |