WO2023044907A1

WO2023044907A1 - Method and device for monitoring equipment health and computer readable storage medium

Info

Publication number: WO2023044907A1
Application number: PCT/CN2021/120981
Authority: WO
Inventors: Xiao Zhou ZHOU; Tian Rui SUN; Xin Bai; Huan Lun LI
Original assignee: Siemens Aktiengesellschaft; Siemens Ltd., China
Priority date: 2021-09-27
Filing date: 2021-09-27
Publication date: 2023-03-30

Abstract

The method comprising: determining a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method; determining a simulation model of the equipment in the possible fault; verifying the possible fault using the operating data based on the simulation model. The possible faults of equipment are determined by artificial intelligence and are verified through a simulation model library, so that causes of the faults can be easily and accurately determined. It can also provide fault-parameter values of possible faults, give quantitative results of faults, and sort possible faults based on multiple dimensions.

Description

Method and device for monitoring equipment health and computer readable storage medium

FIELD

The present invention relates to the technical field of equipment maintenance, in particular to a method and device for monitoring equipment health and computer readable storage medium.

BACKGROUND

Equipment (such as motors, water pumps, compressors, gas turbines, etc. ) that is aging or heavily loaded is prone to fault or low work efficiency. The root causes may be deformation of key components, material degradation, wear, overheating or corrosion, etc. The operator needs to monitor health of the equipment, detect problems, find out root cause and restore the equipment to a normal state.

Currently, equipment health monitoring and fault diagnosis usually rely on sensors installed on the equipment and/or manual inspections of the equipment. However, sometimes sensors for condition monitoring are not installed on key parts of the equipment. For large and complex equipment, such as compressors, the cost of installing sensors to monitor all key components and all key parameters is very high. Due to the lack of sensor data from the equipment and complexity of the equipment, it is difficult to monitor the condition of the equipment and determine root causes of occurred faults. In addition, manual inspection of equipment is both cumbersome and dangerous.

SUMMARY

The embodiment of the present invention proposes a method and device for monitoring equipment health and computer readable storage medium.

In a first aspect, a method for monitoring equipment health is provided. The method comprising:

determining a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method;

determining a simulation model of the equipment in the possible fault;

verifying the possible fault using the operating data based on the simulation model.

Therefore, the embodiment of the present invention uses artificial intelligence to determine a possible fault of equipment, and verifies the possible fault through a simulation model in the possible fault state, so that the cause of the fault can be easily and accurately determined.

Preferably, further comprising:

providing the operation data as an input of a forward-simulation model when the simulation model is the forward-simulation model; running the forward-simulation model; determining a simulation value generated based on the forward-simulation model as a fault-parameter value of the possible fault.

It can be seen that the embodiment of the present invention uses a forward-simulation model to provide fault -parameter values for possible faults, and can give quantitative results of the possible faults.

Preferably, wherein verifying the possible fault using the operating data based on the simulation model comprising:

determining whether the fault-parameter value is within a preset normal-value range, wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.

Therefore, by comparing the fault-parameter value with a normal-value range, a quick verification process for possible faults is performed.

Preferably, further comprising:

providing a fault parameter of the possible fault as an input of a reverse-simulation model when the simulation model is the reverse-simulation model, wherein the fault parameter has a plurality of possible setting values;

running the reverse-simulation model provided with respective setting values to obtain respective simulation values of operating data corresponding to the respective setting values;

determining a minimum value of respective differences between the respective simulation values of operating data and the operating data;

determining the setting value corresponding to the minimum value as a fault-parameter value of the possible fault.

It can be seen that the embodiment of the present invention uses a reverse-simulation model to provide fault -parameter values for possible faults, and can give quantitative results of the possible faults.

Preferably, wherein verifying the possible fault using the operating data based on the simulation model comprising: determining that the possible fault is an unverified fault when the minimum value is greater than a preset threshold value.

Therefore, a quick verification process is performed for possible faults with the respective minimum values.

Preferably, wherein verifying the possible fault using the operating data based on the simulation model comprising: determining whether the fault-parameter value is within a preset normal-value range; wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.

It can be seen that by comparing the fault parameter value with a normal value range, a quick verification process can be performed on possible faults.

Preferably, comprising sorting the verified faults based on the respective fault-parameter values of the verified faults.

Therefore, the verified faults can be sorted based on the fault-parameter values, which is convenient for users to understand the urgency of the fault.

Preferably, wherein determining a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method comprising: inputting the operating data into a trained artificial-intelligence module adapted to detect faults; receiving a list containing each possible fault and the probability of each possible fault from the artificial-intelligence module; the method further comprising: sorting the verified faults based on the respective probabilities of the verified faults.

It can be seen that the verified faults can be sorted based on probability, which is convenient for users to understand the probabilities of fault occurrence.

Preferably, further comprising: sorting the verified faults based on the respective minimum values of the verified faults.

Therefore, the verified faults can be sorted based on the respective minimum values, so that users can easily find faults that are more likely to occur.

In a second aspect, a device for monitoring equipment health is provided. The device comprising:

a first determining module, configured to determine a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method;

a second determining module, configured to determine a simulation model of the equipment in the possible fault;

a verifying module, configured to verify the possible fault using the operating data based on the simulation model.

Preferably, wherein the verifying module is further configured to provide the operating data as an input of a forward-simulation model when the simulation model is the forward-simulation model; run the forward-simulation model; and determine a simulation value generated based on the forward-simulation model as a fault-parameter value of the possible fault.

It can be seen that the embodiment of the present invention uses a forward-simulation model to provide fault -parameter values for possible faults, and can give a quantitative result of the possible faults.

Preferably, wherein the verifying module is configured to determine whether the fault-parameter value is within a preset normal-value range, wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.

Preferably, wherein the verifying module is further configured to provide a fault parameter of the possible fault as an input of a reverse-simulation model when the simulation model is the reverse-simulation model, wherein the fault parameter has a plurality of possible setting values; run the reverse-simulation model provided with respective setting values to obtain respective simulation values of operating data corresponding to the respective setting values; determine a minimum value of respective differences between the respective simulation values of operating data and the operating data; and determine the setting value corresponding to the minimum value as a fault-parameter value of the possible fault.

It can be seen that the embodiment of the present invention uses a reverse-simulation model to provide fault -parameter values for possible faults, and can give a quantitative result of the possible faults.

Preferably, wherein the verifying module is configured to determine that the possible fault is an unverified fault when the minimum value is greater than a preset threshold value.

Preferably, wherein the verifying module is configured to determine whether the fault-parameter value is within a preset normal-value range; wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.

Preferably, further comprising a sorting module, configured to sort the verified faults based on the respective fault-parameter values of the verified faults.

Preferably, wherein the first determining module is configured to input the operating data into a trained artificial-intelligence module adapted to detect faults; receive a list containing each possible fault and the probability of each possible fault from the artificial-intelligence module; the device further comprising: a sorting module, configured to sort the verified faults based on the respective differences of the verified faults.

Preferably, further comprising: sorting module, configured to sort the verified faults based on respective minimum values of the verified faults.

Therefore, the verified faults can be sorted based on respective minimum values, so that users can easily find faults that are more likely to occur.

In a third aspect, an electronic device is provided. The electronic device comprising a processor and a memory, wherein an application program executable by the processor is stored in the memory for causing the processor to execute a method for monitoring equipment health according to any one of above.

In a fourth aspect, a computer-readable medium comprising computer-readable instructions stored thereon is provided, wherein the computer-readable instructions for executing a method for monitoring equipment health according to any one of above.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to make technical solutions of examples of the present disclosure clearer, accompanying drawings to be used in description of the examples will be simply introduced hereinafter. Obviously, the accompanying drawings to be described hereinafter are only some examples of the present disclosure. Those skilled in the art may obtain other drawings according to these accompanying drawings without creative labor.

Fig. 1 is a flowchart of a method for monitoring equipment health according to an embodiment of the present invention.

Fig. 2 is an exemplary flow chart of generating a simulation model library according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of generating an artificial-intelligence module adapted to detect faults according to an embodiment of the present invention.

Fig. 4 is an exemplary flow chart of verifying fault according to an embodiment of the present invention.

Fig. 5 is an exemplary flowchart of a fault sorting process according to an embodiment of the present invention.

Fig. 6 is a structural diagram of a device for monitoring equipment health according to an embodiment of the present invention.

FIG. 7 is a structural diagram of an electronic device with a processor-memory architecture according to an embodiment of the present invention.

List of reference numbers:

reference numbers	meanings
100	method for monitoring equipment health
101～103	steps
201～206	steps
301	historical operating data
302	test data
303	data analysis algorithm
304	data model
305	empirical rule
306	artificial-intelligence module
400	simulation model library
401～409	steps
500	sorting process
501～506	steps
600	device for monitoring equipment health
601	first determining module
602	second determining module
603	verifying module
604	sorting module
700	electronic device
701	processor
702	memory

DETAILED DESCRIPTION

In order to make the technical solutions and advantages of the present invention more comprehensible, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the scope of the invention.

In an embodiment of the present invention, artificial intelligence technology and simulation technology are integrated to perform analysis on the operating data of equipment, so as to realize health monitoring and fault diagnosis of the equipment.

As shown in Fig. 1, the method 100 comprising:

Step 101: use an operating data of the device to determine a possible fault of the device, based on an artificial intelligence method.

Here, the equipment is preferably implemented as large-scale complex equipment, such as motors, water pumps, compressors, gas turbines, and so on. Moreover, the operating data of the equipment is different from simulated value, which is true value of the equipment.

In one embodiment, in step 101: input equipment operating data into a trained artificial intelligence module adapted to detect equipment faults, and receive each detected possible fault from the artificial intelligence module. Preferably, in step 101: a list containing each possible fault and the probability of each possible fault is received from the artificial intelligence module.

For example, the artificial-intelligence module contains pre-established empirical rules and data models. Among them: data model is created by applying data analysis algorithms and/or machine learning methods to a historical operating data of the equipment and/or test data of the equipment. Empirical rules and data models can identify possible equipment faults by analyzing the current operating data of the equipment (which can include sensor data) .

As shown in Fig. 3, historical operation data 301 of the equipment and the test data 302 of the equipment are provided to data analysis algorithm 303, thereby generating data model 304. The historical operation data 301 of the equipment is data generated during historical operation of the equipment. The historical operating data 301 of the equipment may include historical control commands of the equipment or historical detection data of sensors, and so on. The test data 302 of the equipment refers to the data generated during tests initiated by the equipment. An artificial intelligence module 306 is generated based on the data model 304 and empirical rules 305. The rule of thumb 305 is regularized expert knowledge.

In one embodiment, the artificial intelligence module 306 is implemented as a neural network model. The neural network model can be trained using training data (for example, labeled data or unlabeled data) , wherein when the accuracy of the output result of the neural network model is greater than a predetermined threshold, a trained artificial intelligence module is obtained. Specifically, the neural network model can be implemented as: feedforward neural network model, radial basis neural network model, long short-term memory (LSTM) network model, echo state network (ESN) , gate recurrent unit (GRU) network model or deep residual Poor network model, etc.

The above exemplarily describes typical embodiments of the neural network model, and those skilled in the art may realize that this description is only exemplary and is not used to limit the protection scope of the embodiments of the present invention.

Step 102: Determine a simulation model of the equipment in the possible fault state.

Here, the simulation model of the equipment in the possible fault state is: a model that simulates the operation process of the equipment in the possible fault state.

The simulation model corresponding to possible fault determined in step 101 can be retrieved from a simulation model library containing simulation models for all possible faults. Among them: the simulation model can be implemented as a forward-simulation model or a reverse-simulation model. For forward-simulation model corresponding to possible fault, the input is the operating data of the equipment and the output simulation result is a fault parameter of the possible fault (for example, temperature at the time of overheating defect) . For reverse-simulation model corresponding to possible fault, the input is a fault parameter of the possible fault (such as the wear thickness of wear defect) , and the output simulation result is the operating data of the equipment. The reverse-simulation model can provide accurate fault parameters through calibration algorithms.

As shown in Fig. 2, the process of generating a simulation model library comprising:

Step 201: Generate a fault-list including all typical faults of the equipment.

Here, for each type of fault, one or more fault parameters can be identified. For example, for crack fault, two fault parameters can be identified, namely the length parameter and the depth parameter of the crack.

Step 202: Establish simulation models under each typical failure separately.

Here, for each type of typical fault, a simulation model of the entire equipment or part of the equipment is established to simulate the impact of this type of fault on the operation of the equipment. Each simulation model may be a forward-simulation model or a reverse-simulation model. Here, the specific method of modeling the simulation model can refer to the common simulation modeling technology in the prior art, which will not be repeated in the embodiment of the present invention.

Step 203: Parameterize each simulation model.

Therefore, each simulation model is parameterized and each simulation model can be easily updated with different fault parameter values and operating data.

Step 204: For each simulation model, determine whether the simulation model is a reverse-simulation model, if yes, perform step 205 and subsequent steps, otherwise perform step 206.

Step 205: For the reverse simulation model, specify a calibration algorithm with fine-tuned parameters.

For each reverse simulation model, a calibration algorithm with fine-tuning parameters is specified separately. Therefore, once the operating data of the equipment is given, the fault parameters can be calculated quickly.

Step 206: Combine simulation models of all faults into a simulation model library.

When a new potential fault is discovered, the simulation model library can be dynamically updated and expanded.

The above describes a typical example of establishing a simulation model library, and those skilled in the art may realize that this description is only exemplary and is not used to limit the protection scope of the embodiments of the present invention.

Step 103: Use the operating data to verify each possible fault based on respective simulation model corresponding to respective possible fault.

Therefore, the embodiment of the present invention uses artificial intelligence to determine the possible faults of the equipment, and also verifies the possible faults through the simulation model library, so that the cause of the faults can be determined conveniently and accurately.

In one embodiment, the method 100 further comprising: when the simulation model is a forward-simulation model, providing the operation data as an input of the forward-simulation model; running the forward-simulation model; determining a simulation value generated based on the forward-simulation model as a fault-parameter value of the possible fault.

Example: The forward-simulation model is a model that simulates operation process of equipment containing a bearing when the bearing wears out. The vibration state of a screw is extracted from the operating data of the equipment, and the vibration state of the screw is provided as input data to the forward-simulation model. Moreover, a bearing wear amount in the simulation result output by the forward-simulation model is determined as the fault-parameter value of bearing wear fault.

It can be seen that the embodiment of the present invention uses a forward-simulation model to provide fault-parameter values of possible faults, and can give a quantitative result of the fault.

In one embodiment, wherein using the operating data to verify the possible fault based on the simulation model in step 103 comprising: judging whether fault-parameter value is within a preset normal value range, and when it is within the normal value range, determining that the possible fault is unverified fault; when it is not within the normal value range, the possible fault is determined to be a verified fault. The normal value range includes the range of values of the parameter when the device is working normally. Therefore, by comparing the fault-parameter value with a normal value range, a quick verification process for possible faults is performed.

In one embodiment, the method 100 further comprising: providing a fault parameter of the possible fault as an input of a reverse-simulation model when the simulation model is the reverse-simulation model, wherein the fault parameter has a plurality of possible setting values; running the reverse-simulation model provided with respective setting values to obtain respective simulation values of operating data corresponding to the respective setting values; determining a minimum value of respective differences between the respective simulation values of operating data and the operating data; determining the setting value corresponding to the minimum value as a fault-parameter value of the possible fault. An exemplary calibration algorithm of the reverse simulation model is described in detail above. In fact, other calibration algorithms can also be used, such as using an optimization algorithm to automatically find a certain setting value in a specified interval, so as to minimize the difference between the simulated value and the running data.

Example: The reverse-simulation model is a model for simulating operation process of equipment containing a support when the support has a crack fault. The vibration amplitude of the support is extracted from the operating data of the equipment. The fault parameter is the crack length of the support. The crack length has multiple preset possible settings. Each possible setting value is used as input of the reverse-simulation model and provided to the reverse simulation-model, so that the reverse-simulation model simulates the respective operation processes of the equipment when the crack length is equal to respective setting values, and obtains respective simulation results corresponding to the respective setting values, that is, the respective simulation values of the vibration amplitude of the support. Then, the differences between respective simulation values and the vibration amplitude of the support member extracted from the operating data of the equipment are calculated. Determine the minimum value of the differences (referred to as the minimum difference) , and determine the setting value corresponding to the minimum value as the crack length at the time of crack fault.

It can be seen that the embodiment of the present invention uses a reverse-simulation model to provide fault-parameter values of possible faults, and can give a quantitative result of the fault.

In one embodiment, in step 103, wherein using operating data to verify possible faults based on the simulation model comprising: When the minimum value (that is, the minimum value of the differences between simulation values of operating data and the real operating data) is greater than a preset threshold value, it is determined that the possible fault is an unverified fault. Therefore, when the difference is relatively large, the reliability of the possible fault is relatively low and the possible fault is determined to be an unverified fault.

In one embodiment, the method further comprising: sorting the verified faults based on the smallest difference between the simulation values of the operating data and the real operating data (called the minimum calibration error) . Therefore, the verified faults can be sorted based on the difference, which is convenient for users to understand the urgency of the fault.

In one embodiment, based on the simulation model in step 103, using the operating data to verify the possible fault comprising: judging whether the fault-parameter value is within a preset normal-value range, and when it is within the normal-value range, determining that the possible fault is unverified fault; when it is not within the normal value range, the possible fault is determined to be a verified fault. Therefore, by comparing the fault-parameter value with a normal value range, a quick verification process is performed for possible faults.

As shown in Figure 4, the verification fault processing process comprising:

Step 401: Input real operating data of the equipment into a trained artificial intelligence module adapted to detect possible faults, and receive a list containing each possible fault and the probability of each possible fault from the artificial intelligence module.

Step 402: Retrieve simulation model of the equipment in the possible fault state from simulation model library 400. Wherein when multiple possible faults are received in step 401, the number of corresponding simulation models is also multiple.

Step 403: Determine whether the simulation model is a reverse-simulation model. If it is a reverse -simulation model (corresponding to the "Y" branch in Fig. 4) , perform step 404 and subsequent steps; otherwise (correspond to the "N" branch) perform step 406 and subsequent steps.

Step 404: Provide the fault parameter as an input of the reverse-simulation model, where the fault parameter has multiple possible setting values. Run the simulation model provided with the respective setting values to obtain respective operating data simulation values corresponding to the respective setting values. Determine respective differences between respective simulation values of operating data and the real operating data. Determine the minimum of all differences. Determine setting value corresponding to the minimum difference as fault-parameter value of the possible fault.

Step 405: Determine whether the minimum difference is greater than a preset difference threshold, if it is (corresponding to the "Y" branch in FIG. 4) , go to step 410 and end the process; otherwise (correspond to the "Y" branch in FIG. N” branch) go to step 407 and subsequent steps.

Step 406: Provide the real operating data as the input of the forward-simulation model; run the forward -simulation model; determine simulation value generated based on the forward-simulation model as a fault parameter-value of the possible fault.

Step 407: Determine whether the fault-parameter value is within a preset normal value range, if it is (corresponding to the "Y" branch in FIG. 4) , go to step 408 and end the process; otherwise (correspond to the "N" in FIG. 4 Branch) execute step 409 and end the process.

Step 408: It is determined that the possible fault is an unverified fault and this process is ended.

Step 409: It is determined that the possible fault is a verified fault and this process is ended.

Step 410: It is determined that the possible fault is an unverified fault and this process is ended.

In one embodiment, the method 100 further comprising: sorting the verified faults based on the fault-parameter values of the verified faults. For example, each difference between the fault-parameter value of each verified fault and a preset normal parameter value is calculated and the verified faults are sorted based on the respective differences. The greater the difference between fault-parameter value and the preset normal parameter value, the more serious the fault. This sorting method is convenient for users to understand the urgency of the fault.

In one embodiment, step 101 comprising: inputting the operating data into a trained artificial-intelligence module adapted to detect faults; receiving a list containing each possible fault and the probability of each possible fault from the artificial-intelligence module. The method 100 further comprising: sorting the verified faults based on the respective probabilities of the verified faults. It can be seen that the verified faults can be sorted based on the probability, which is convenient for users to understand the probability of fault.

As shown in Fig. 5, the fault sorting process includes:

Step 501: Determine whether the verified fault has been found, if it is (corresponding to the "Y" branch in Fig. 5) , perform step 502 and subsequent steps; otherwise (corresponding to the "N" branch in Fig. 5) Step 504 and subsequent steps are performed.

Step 502: Invoke sorting process 500 to sort the verified faults, where the sorting basis used in the sorting process 500 may include: fault probability; fault parameter values; calibration errors, and so on.

Step 503: Provide a health monitoring status and/or maintenance plan based on the sorting result, and end this process.

For example, for several verified faults at the top of the ranking, preset maintenance plans are given respectively.

Step 504: Perform manual equipment fault analysis.

Step 505: Update the fault list based on the result of manual equipment fault analysis.

Step 506: Update the simulation model library based on the result of manual equipment fault analysis.

It can be seen that when no verified fault is found, the possible reason is that the artificial intelligence module and simulation model failed to accurately identify or verify the fault. At this time, manual fault analysis process is introduced, and the feedback result of the manual fault analysis process can improve the artificial intelligence module and simulation model.

The embodiment of the present invention also provides a health monitoring device for equipment. Fig. 6 is a structural diagram of a device for monitoring equipment health according to an embodiment of the present invention.

As shown in FIG. 6, the device 600 for monitoring equipment health comprising:

a first determining module 601 is configured to determine a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method;

a second determining module 602 is configured to determine a simulation model of the equipment in the possible fault;

a verification module 603 is configured to verify the possible fault using the operating data based on the simulation model.

In one embodiment, wherein the verifying module 603 is further configured to provide the operating data as an input of a forward-simulation model when the simulation model is the forward-simulation model; run the forward-simulation model; and determine a simulation value generated based on the forward-simulation model as a fault-parameter value of the possible fault.

In one embodiment, wherein the verifying module 603 is configured to determine whether the fault-parameter value is within a preset normal-value range, wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.

In one embodiment, wherein the verifying module 603 is further configured to provide a fault parameter of the possible fault as an input of a reverse-simulation model when the simulation model is the reverse-simulation model, wherein the fault parameter has a plurality of possible setting values; run the reverse-simulation model provided with respective setting values to obtain respective simulation values of operating data corresponding to the respective setting values; determine a minimum value of respective differences between the respective simulation values of operating data and the operating data; and determine the setting value corresponding to the minimum value as a fault-parameter value of the possible fault.

In one embodiment, wherein the verifying module 603 is configured to determine that the possible fault is an unverified fault when the minimum value is greater than a preset threshold value.

In one embodiment, wherein the verifying module 603 is configured to determine whether the fault-parameter value is within a preset normal-value range; wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.

In one embodiment, the device 600 further comprising: a sorting module 604, configured to sort the verified faults based on the respective fault-parameter values of the verified faults.

In one embodiment, the first determining module 601 is configured to input the operating data into a trained artificial-intelligence module adapted to detect faults; receive a list containing each possible fault and the probability of each possible fault from the artificial-intelligence module. The device 600 further comprising: a sorting module 604, configured to sort the verified faults based on the respective differences of the verified faults.

In one embodiment, the device 600 further comprising: a sorting module 604, configured to sort the verified faults based on respective minimum values of the verified faults.

In summary, the embodiment of the present invention combines data analysis and simulation technology to solve the problem of equipment fault diagnosis. The embodiment of the present invention can also provide a sequence of possible faults according to estimated values of the fault parameters, clearly display health status of the equipment and provide guidance for maintenance. The embodiment of the present invention also prepares a respective simulation model for each potential fault, and can automatically select appropriate simulation model according to the recognition result of the artificial intelligence diagnosis, which can significantly save calculation cost, because only the simulation model with most possible faults can be run. In addition, the embodiment of the present invention prepares an automatic calibration algorithm for each reverse -simulation model, and uses the automatic calibration algorithm to calibrate the reverse simulation model to obtain accurate fault parameter estimation.

The embodiment of the present invention also proposes an electronic device with a processor-memory architecture and adapted to perform health monitoring of the device. Fig. 7 is a structural diagram of an electronic device with a processor-memory architecture according to an embodiment of the present invention.

As shown in Fig. 7, the electronic device 700 includes a processor 701, a memory 702, and a computer program stored on the memory 702 and running on the processor 701. When the computer program is executed by the processor 701, it realizes the operation of any of the method for monitoring equipment health. Among them, the memory 702 may be specifically implemented as various storage media such as an electrically erasable programmable read-only memory (EEPROM) , a flash memory (Flash memory) , and a programmable program read-only memory (PROM) . The processor 701 may be implemented to include one or more central processing units or one or more field programmable gate arrays, where the field programmable gate array integrates one or more central processing unit cores. Specifically, the central processing unit or central processing unit core may be implemented as a CPU, MCU, or DSP, and so on.

It should be noted that not all steps and modules in the above-mentioned processes and structural diagrams are necessary, and some steps or modules can be omitted according to actual needs. The order of execution of each step is not fixed and can be adjusted as needed. The division of each module is just to facilitate the description of the functional division. In actual implementation, a module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be located in the same device. It can also be located in a different device.

The hardware modules in each embodiment can be implemented in a mechanical way or an electronic way. For example, a hardware module may include specially designed permanent circuits or logic devices (such as dedicated processors, such as FPGAs or ASICs) to complete specific operations. The hardware module may also include programmable logic devices or circuits temporarily configured by software (for example, including general-purpose processors or other programmable processors) for performing specific operations. As for the specific use of mechanical methods, or the use of dedicated permanent circuits, or the use of temporarily configured circuits (such as software configuration) to implement hardware modules, it can be determined according to cost and time considerations.

The above are only the preferred embodiments of the present invention, and are not used to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

A method (100) for monitoring equipment health, comprising:

determining (101) a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method;

determining (102) a simulation model of the equipment in the possible fault;

verifying (103) the possible fault using the operating data based on the simulation model.
The method (100) according to claim 1, further comprising:

providing the operation data as an input of a forward-simulation model when the simulation model is the forward-simulation model;

running the forward-simulation model;

determining a simulation value generated based on the forward-simulation model as a fault-parameter value of the possible fault.
The method (100) according to claim 2, wherein verifying (103) the possible fault using the operating data based on the simulation model comprising:

determining whether the fault-parameter value is within a preset normal-value range, wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.
The method (100) according to claim 1, further comprising:

providing a fault parameter of the possible fault as an input of a reverse-simulation model when the simulation model is the reverse-simulation model, wherein the fault parameter has a plurality of possible setting values;

running the reverse-simulation model provided with respective setting values to obtain respective simulation values of operating data corresponding to the respective setting values;

determining a minimum value of respective differences between the respective simulation values of operating data and the operating data;

determining the setting value corresponding to the minimum value as a fault-parameter value of the possible fault.
The method (100) according to claim 4, wherein verifying (103) the possible fault using the operating data based on the simulation model comprising:

determining that the possible fault is an unverified fault when the minimum value is greater than a preset threshold value.
The method (100) according to claim 4, wherein verifying (103) the possible fault using the operating data based on the simulation model comprising:

determining whether the fault-parameter value is within a preset normal-value range; wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.
The method (100) according to claim 3, 5 or 6, comprising:

sorting the verified faults based on the respective fault-parameter values of the verified faults.
The method (100) according to claim 3, 5 or 6, wherein determining (101) a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method comprising:

inputting the operating data into a trained artificial-intelligence module adapted to detect faults;

receiving a list containing each possible fault and the probability of each possible fault from the artificial-intelligence module;

the method (100) further comprising:

sorting the verified faults based on the respective probabilities of the verified faults.
The method (100) according to claim 5 or 6, further comprising:

sorting the verified faults based on the respective minimum values of the verified faults.
A device (600) for monitoring equipment health, comprising:

a first determining module (601) , configured to determine a possible fault of equipment using an operating data of the equipment based on an artificial intelligence method;

a second determining module (602) , configured to determine a simulation model of the equipment in the possible fault;

a verifying module (603) , configured to verify the possible fault using the operating data based on the simulation model.
The device (600) according to claim 10, wherein the verifying module (603) is further configured to provide the operating data as an input of a forward-simulation model when the simulation model is the forward-simulation model; run the forward-simulation model; and determine a simulation value generated based on the forward-simulation model as a fault-parameter value of the possible fault.
The device (600) according to claim 11, wherein the verifying module (603) is configured to determine whether the fault-parameter value is within a preset normal-value range, wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.
The device (600) according to claim 10, wherein the verifying module (603) is further configured to provide a fault parameter of the possible fault as an input of a reverse-simulation model when the simulation model is the reverse-simulation model, wherein the fault parameter has a plurality of possible setting values; run the reverse-simulation model provided with respective setting values to obtain respective simulation values of operating data corresponding to the respective setting values; determine a minimum value of respective differences between the respective simulation values of operating data and the operating data; and determine the setting value corresponding to the minimum value as a fault-parameter value of the possible fault.
The device (600) according to claim 13, wherein the verifying module (603) is configured to determine that the possible fault is an unverified fault when the minimum value is greater than a preset threshold value.
The device (600) according to claim 13, wherein the verifying module (603) is configured to determine whether the fault-parameter value is within a preset normal-value range; wherein determining that the possible fault is an unverified fault when it is within the normal-value range and determining that the possible fault is a verified fault when it is not within the normal-value range.
The device (600) according to claim 12, 14 or 15, further comprising:

a sorting module (604) , configured to sort the verified faults based on the respective fault-parameter values of the verified faults.
The device (600) according to claim 12, 14 or 15, wherein the first determining module (601) is configured to input the operating data into a trained artificial-intelligence module adapted to detect faults; receive a list containing each possible fault and the probability of each possible fault from the artificial-intelligence module; the device (600) further comprising:

a sorting module (604) , configured to sort the verified faults based on the respective differences of the verified faults.
The device (600) according to claim 14 or 15, further comprising:

sorting module (604) , configured to sort the verified faults based on respective minimum values of the verified faults.
An electronic device (700) , comprising a processor (701) and a memory (702) , wherein an application program executable by the processor (701) is stored in the memory (702) for causing the processor (701) to execute a method (100) for monitoring equipment health according to any one of claims 1 to 9.
A computer-readable medium comprising computer-readable instructions stored thereon, wherein the computer-readable instructions for executing a method (100) for monitoring equipment health according to any one of claims 1 to 9.