CN116292245A - Piston pump voiceprint fault detection method - Google Patents
Piston pump voiceprint fault detection method Download PDFInfo
- Publication number
- CN116292245A CN116292245A CN202310190590.8A CN202310190590A CN116292245A CN 116292245 A CN116292245 A CN 116292245A CN 202310190590 A CN202310190590 A CN 202310190590A CN 116292245 A CN116292245 A CN 116292245A
- Authority
- CN
- China
- Prior art keywords
- piston pump
- voiceprint
- optimal
- fault
- model
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 21
- 241000283153 Cetacea Species 0.000 claims abstract description 37
- 238000013528 artificial neural network Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000005457 optimization Methods 0.000 claims abstract description 20
- 238000003062 neural network model Methods 0.000 claims abstract description 11
- 238000010219 correlation analysis Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims abstract description 3
- 238000000034 method Methods 0.000 claims description 12
- 230000006399 behavior Effects 0.000 claims description 11
- 230000003044 adaptive effect Effects 0.000 claims description 10
- 238000009826 distribution Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 2
- 238000012824 chemical production Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 8
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 238000012423 maintenance Methods 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
- 230000008569 process Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B51/00—Testing machines, pumps, or pumping installations
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a piston pump voiceprint fault detection method, which comprises the following steps: s1, carrying out standardization processing on voiceprint data samples of a piston pump; s2, performing correlation analysis on the voiceprint characteristics of the piston pump in the voiceprint data sample and fault categories, and selecting the characteristics with stronger correlation with category labels; s3, constructing a topological structure of the piston pump voiceprint fault detection neural network model by using the selected characteristics as input and the fault type as output of the network, and initializing network weight and a threshold value; s4, inputting the training set into a neural network to train a model, and obtaining a training error as a fitness value; s5, initializing relevant parameters of a whale optimization algorithm HAWOA; s6, performing iterative optimization by utilizing HAWOA improved based on the self-adaptive strategy to obtain optimal parameters; and S7, decoding and assigning the model to a neural network framework, and substituting the model into subsequent training to obtain a final piston pump voiceprint fault model. The invention meets the safety detection requirement of chemical production.
Description
Technical Field
The invention relates to a piston pump voiceprint fault detection method.
Background
A piston pump is a fixed sealing member, and is a pump device for sucking and discharging liquid by the back and forth movement of a piston, and is widely applied to various industrial scenes for delivering fluid.
The piston pump is driven by the reciprocating motion of the piston, so that the working volume of the pump cavity is periodically changed, and liquid is sucked and discharged. During operation, the piston pump can have common fault problems such as abrasion, corrosion, blockage and the like.
At present, manual inspection or various sensors are generally adopted to monitor and detect faults, but the manual inspection mode often causes misjudgment due to insufficient personal experience, the sensors can only monitor the monitoring points in a related mode, the running state of pump equipment cannot be comprehensively represented, and the safety detection requirement of chemical production is difficult to meet.
Disclosure of Invention
The invention aims to provide a method for detecting the voiceprint faults of a piston pump, which aims to solve the technical problem of detecting the faults of the piston pump in industrial production.
Therefore, the invention provides a piston pump voiceprint fault detection method, which comprises the following steps: s1, carrying out standardization processing on voiceprint data samples of a piston pump; s2, performing correlation analysis on the voice print characteristics of the piston pump in the voice print data sample and fault categories, selecting the characteristics with stronger correlation with category labels, and dividing the characteristics into a training set and a testing set; s3, constructing a topological structure of the piston pump voiceprint fault detection neural network model by using the selected characteristics as input and the fault type as output of the network, and initializing network weight and a threshold value; s4, inputting the training set into a neural network to train a model, obtaining training errors, and constructing a neural network model error fitness function; s5, setting relevant parameters of a whale optimization algorithm HAWOA, wherein the relevant parameters comprise a position vector, a population scale, a maximum iteration number and an initial convergence factor, and the position vector is obtained by the initialization network weight and threshold coding; s6, taking the error fitness function as an objective function of an HAWOA algorithm, calculating individual fitness values of whales according to the objective function, and performing iterative optimization by using an improved HAWOA based on an adaptive strategy to obtain optimal parameters; s7, decoding the position vector in the optimal parameters to obtain optimal network weight and threshold value, assigning the optimal network weight and threshold value to a neural network frame, substituting the optimal network weight and threshold value into subsequent training to obtain a final piston pump voiceprint fault model, and using the model for piston pump voiceprint fault detection.
The beneficial effects of the invention are as follows:
according to the invention, the HAWOA-BP model is constructed based on voiceprint recognition and deep learning thought by utilizing voiceprint data of the operation of the piston pump, so that the fault detection efficiency of the piston pump is improved, and the rapid and accurate fault detection of the piston pump according to the voiceprint is realized. In order to improve the convergence speed and the classification accuracy of the BP neural network in the aspect of piston pump voiceprint fault detection, a BP neural network piston pump voiceprint fault detection model based on a whale optimization algorithm is provided, the weight and the threshold of the whale optimization algorithm are adjusted through the self-adaptive strategy, the global searching capacity of population and the ability of jumping out of a local optimal solution are enhanced, and the optimal weight and the threshold of WOA output improvement are adopted as an initial parameter training model through the BP neural network. The invention is beneficial to promoting the fault detection of the piston pump, realizing the rapid and accurate judgment of the operation fault of the piston pump, further providing the basis of fault diagnosis by monitoring the operation state of the equipment, greatly improving the level of equipment management, maintenance and economic operation, improving the operation stability of the equipment, reducing the overall operation and maintenance cost, and being convenient and practical.
In addition to the objects, features and advantages described above, the present invention has other objects, features and advantages. The present invention will be described in further detail with reference to the drawings.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention. In the drawings:
FIG. 1 is a diagram of data enhancement performed on raw data;
FIG. 2 is a flow chart of the use of the present invention;
FIG. 3 is a flow chart of an implementation of the improved whale optimization algorithm based on an adaptive strategy.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
Based on different voiceprint characteristics of the piston pump in a normal state and a fault state, the invention classifies the state of the piston pump by using a neural network model according to voiceprint data acquired during operation of the piston pump. In order to solve the problems that the convergence rate is low and a local optimal value is easy to fall in the training of the BP neural network, a novel whale optimization algorithm (Whale optimization algorithm based on hybrid adaptive strategy, HAWOA) is provided, the initial weight and the threshold value of the BP neural network are optimized, and a good effect of detecting the voiceprint fault of the piston pump is achieved.
Referring to fig. 2, the invention relates to a method for detecting a voiceprint fault of a piston pump, which specifically comprises the following steps S1-S7.
S1: and normalizing the voiceprint data of the piston pump.
In order to process data, a straight-square equalization method is adopted to enhance the data and improve the data quality, and the process of enhancing the data of the original image is drawn as shown in fig. 1.
S2: and carrying out correlation analysis on the voiceprint characteristics and the fault class of the piston pump by using the Pearson correlation coefficient. Specifically, the characteristics with weak correlation or irrelevant characteristics are properly discarded through correlation analysis, the characteristics with strong correlation with the category labels are reserved, and the data set is divided into a training set and a testing set according to the ratio of 7:3.
The invention mainly aims at the hydraulic piston pump, and the fault types mainly comprise: transient pressure spike faults, plugging or restricted faults, pump housing fatigue faults, etc.; for convenient operation, the piston pumps aimed at by the invention belong to the same type, and different types of piston pumps need to retrain the model according to voiceprint characteristics.
The pearson correlation coefficient is used to measure the degree of correlation between two variables X, Y, and the calculation formula is as follows:
wherein ρ is X,Y Representing the correlation coefficient of X, Y, cov (X, Y) representing the covariance of X, Y, and the calculation formula is E [ (X-mu) X )(Y-μ Y )],σ X 、σ Y Representing the standard deviation of X, Y, respectively.
S3: the neural network is initialized. Specifically, the training set divided in the step 2 is used as the input of a neural network, the fault type is used as the output of the network, a piston pump voiceprint fault detection neural network model is constructed, model parameters such as the number of input nodes and output nodes of the network, the maximum iteration number, the learning rate and the like are set according to the selected feature number and the fault state number, and the connection weight and the threshold value are randomly initialized.
S4: and inputting the training set into a neural network to train a model, obtaining a training error, and taking the error as a fitness value. Specifically, training the BP neural network by using a training set to obtain a training error TrainDataErrorate, and constructing a neural network model error fitness function by taking the error as a fitness value as follows:
f=argmin(TrainDataErrorRate)-------------------------------(2)
s5: HAWOA initialization. Setting relevant parameters of a whale optimization algorithm, and coding the weight and the threshold value of the neural network to obtain an initial population. Specifically, the initial weight and the threshold value in the step 3 are converted into a position vector of HAWOA, other basic parameters such as population scale, maximum iteration times, initial convergence factors and the like are set, and population distribution is initialized randomly.
S6: and (5) carrying out algorithm optimization by utilizing HAWOA. Specifically, the fitness value in the step 4 is used as an objective function, and an adaptive strategy is adopted for optimization until the maximum iteration number is met or the error precision is reached, so that the optimal parameter is obtained.
As shown in FIG. 3, the method for improving the whale optimization algorithm comprises the following specific steps:
s61: setting a formula (2) as an objective function of a whale optimization algorithm, determining the dimension of a search space according to the number of layers of a neural network and the number of nodes of each layer, setting upper and lower limit ranges lb and ub of the space, setting the maximum iteration times T, and initializing the iteration times t=1.
S62: calculating individual fitness value of whale according to objective function, wherein in HAWOA, each whale individual in search space represents a solution, and the optimal individual isAnd the optimal weight and the threshold value of the neural network are used as optimal parameters of a piston pump voiceprint fault detection model.
S63: the population trapping behavior of whales at the head is simulated and mainly divided into two stages of shrinkage predation and random search. Shrink predation is a locally optimal search, with probability 0.5, performing surrounding prey and spiral update actions; the random search is global optimization, and the algorithm randomly optimizes in the current feasible solution space.
Surrounding prey behavior: assuming the currently optimal individualFor the target prey, as the optimal weight and threshold of the neural network, other individuals in the group move to the optimal position, and the position updating formula is as follows:
wherein,,represents the surrounding step size +.>Is a coefficient vector, and is defined by the formula (3) and the formula (4). r represents [0,1]]Random numbers in between; />Represents a linear convergence factor, in [2,0 ]]The number of iterations decreases with increasing number of iterations.
Spiral update behavior: assume that the distance between the ith whale and the current optimal position isb is a constant coefficient used to define a logarithmic spiral form; l is [ -1,1]Random numbers in the space, and a realizing position updating formula is as follows:
to enhance local search capability, a weighting factor is introduced in the location update behavior of the shrink predation phase that can be adaptively changed according to the current population distribution situation. The formula is as follows:
wherein I is 1 、I 2 Is two constants, set I 1 =10 -4 ,I 2 =10 -4 ,P ibest 、P iworst Respectively representing the optimal position and the worst position vector in the ith iteration population;as variable x i Upper and lower bounds of (2).
In the initial stage of algorithm iteration, the second part of the adaptive weightsThe value of which is only subject to the number of iterations T i Can maintain a greater weight to update whale positions; with the increase of the iteration times, the first part I of the weight is adaptively adjusted 1 ·(P ibest -P iworst ) Still can have great weight value, make whale jump out the local optimum solution.
The probability p is introduced to select and implement synchronous updating of two behaviors in the shrinkage predation stage, and a specific formula is as follows:
s64: during the random search stage, when the A is more than or equal to 1, whale is selected randomlyForcing it away from the reference whale to reach a global search to obtain the possible location of the prey. The random search formula is as follows:
in order to improve the optimizing capability of the model to the weight and threshold parameters of the neural network under the global solution space, a self-adaptive adjustment threshold is introduced in a random search stage to select a whale population random movement mode to enhance the diversity of the population, and the movement of all the population towards a local optimal individual is avoided, and the formula is as follows:
wherein,,representing the average fitness value of the iterative population of the present time, f best 、f worst Respectively representing the best and worst fitness values of the current population.
For each individual whale, a random value q of [0,1] is compared with an adaptive adjustment threshold H calculated for each iteration, if q < H, one individual whale is randomly selected, and according to formula (10), only the position of the whale is updated, and the other individuals remain unchanged. If q > H, the entire whale population location is updated according to equation (8).
X rand =X jlower +r·(X jupper -X jlower )--------------------------(10)
Wherein r is a [0,1]]Random number X of (X) jupper 、X jlower Respectively represent X rand Is the upper and lower bounds of the value of (a).
S65: judging whether the termination condition is reached, if not, turning to S62; otherwise, the algorithm is ended, and the optimal solution is output as the optimal weight and the threshold value of the neural network model.
S7: and decoding, namely assigning the optimal network weight and the threshold value into a BP neural network frame, substituting the optimal network weight and the threshold value into subsequent training to obtain a HAWOA-BP model, and using the model for detecting the voiceprint faults of the piston pump. Specifically, the optimal weight and the threshold value obtained by decoding are applied to a training model in the BP neural network model, so that the accuracy of detecting the fault model of the piston pump by the voiceprint is improved.
The above description is only an example of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. The method for detecting the voiceprint faults of the piston pump is characterized by comprising the following steps of:
s1, carrying out standardization processing on voiceprint data samples of a piston pump;
s2, performing correlation analysis on the voice print characteristics of the piston pump in the voice print data sample and fault categories, selecting the characteristics with stronger correlation with category labels, and dividing the characteristics into a training set and a testing set;
s3, constructing a topological structure of the piston pump voiceprint fault detection neural network model by using the selected characteristics as input and the fault type as output of the network, and initializing network weight and a threshold value;
s4, inputting the training set into a neural network to train a model, obtaining training errors, and constructing a neural network model error fitness function;
s5, setting relevant parameters of a whale optimization algorithm HAWOA, wherein the relevant parameters comprise a position vector, a population scale, a maximum iteration number and an initial convergence factor, and the position vector is obtained by the initialization network weight and threshold coding;
s6, taking the error fitness function as an objective function of the HAWOA algorithm, calculating individual fitness values of whales according to the objective function, and performing iterative optimization by using the HAWOA algorithm to obtain optimal parameters;
s7, decoding the position vector in the optimal parameters to obtain optimal network weight and threshold value, assigning the optimal network weight and threshold value to a neural network frame, substituting the optimal network weight and threshold value into subsequent training to obtain a final piston pump voiceprint fault model, and using the model for piston pump voiceprint fault detection.
2. The piston pump voiceprint fault detection method according to claim 1, wherein the HAWOA algorithm is a whale optimization algorithm modified based on an adaptive strategy.
3. The method for detecting the voiceprint fault of the piston pump according to claim 2, wherein the iterative optimization by using the HAWOA algorithm comprises the steps of:
s61, determining the dimension of a search space according to the number of layers of the neural network and the number of nodes of each layer, setting upper and lower limit ranges lb and ub of the space, maximizing the iteration times T, and initializing the iteration times t=1;
s62, each whale individual in the search space represents a solution, and the optimal individual isThe optimal weight and the threshold value of the neural network are used as optimal parameters of a piston pump voiceprint fault detection model;
s63, simulating population trapping behaviors of whales, wherein the population trapping behaviors mainly comprise two stages of shrinkage predation and random search, and the shrinkage predation is local optimal search; the random search is global optimization, and the algorithm randomly optimizes in the current feasible solution space: surrounding prey behavior: assuming the currently optimal individualFor target prey, as neural network optimal weights and thresholds, +.>And (3) representing the surrounding step length, moving other individuals in the population to the optimal position, and updating the position according to the following formula:
spiral update behavior: assume that the distance between the ith whale and the current optimal position is b is a constant coefficient used to define a logarithmic spiral form; l is a random number between [ -1, and the implementation location update formula is:
to enhance local search capability, a weighting coefficient is introduced in the location update behavior of the shrink predation phase, which can be adaptively changed according to the current population distribution situation:
wherein P is ibest 、P iworst Respectively representing the optimal position and the worst position vector in the ith iteration population; as variable x i Upper and lower bounds of (2); i 1 、I 2 Is two constants, wherein A, C is a coefficient vector, and the introduction probability p is selected to realize synchronous update of two behaviors in the shrink predation stage, and the specific formula is as follows:
s64, randomly searching, wherein when the absolute value A is more than or equal to 1, whales are randomly selectedForcing it away from the reference whale to achieve a global search to obtain the possible locations of the prey, the random search formula is as follows:
the random searching stage introduces an adaptive adjustment threshold to select a whale population random movement mode to enhance the diversity of the population, and the adaptive adjustment threshold has the following formula:
wherein,,representing the average fitness value of the iterative population of the present time, f best 、f worst Respectively representing the best and worst fitness values of the current population,
for each individual whale, a random value q of [0,1] is compared with an adaptive adjustment threshold H calculated for each iteration, and if q < H, one individual whale is randomly selected, and according to formula (8, only the position of the whale is updated, the other individuals remain unchanged, and if q > H, the entire whale population position is updated according to formula (6:
X rand =X jlower +r·(X jupper -X jlower )--------------------------------(8
wherein r is a [0,1] random number, X jupper 、X jlower Respectively represent X rand Is characterized by that its upper and lower boundaries are used for taking values,
s65, judging whether the termination condition is met, and if the termination condition is not met, turning to S62; otherwise, the algorithm is ended, and the optimal solution is output as the optimal weight and the threshold value of the neural network model.
4. The method of claim 1, wherein normalizing the voiceprint dataset of the piston pump comprises data enhancement using a squaring equalization method.
5. The method of claim 1, wherein the correlation analysis of the piston pump voiceprint features and fault categories comprises: and carrying out correlation analysis on the voiceprint characteristics of the piston pump and the fault class of the piston pump by using the Pearson correlation coefficient.
6. The method for detecting the voiceprint fault of the piston pump according to claim 1, wherein the decoding in S7 assigns the optimal network weight and the threshold value to the BP neural network frame, substitutes the optimal network weight and the threshold value into the subsequent training to obtain the HAWOA-BP model, and uses the model for detecting the voiceprint fault of the piston pump.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310190590.8A CN116292245A (en) | 2023-02-27 | 2023-02-27 | Piston pump voiceprint fault detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310190590.8A CN116292245A (en) | 2023-02-27 | 2023-02-27 | Piston pump voiceprint fault detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116292245A true CN116292245A (en) | 2023-06-23 |
Family
ID=86789923
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310190590.8A Pending CN116292245A (en) | 2023-02-27 | 2023-02-27 | Piston pump voiceprint fault detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116292245A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117828403A (en) * | 2024-01-03 | 2024-04-05 | 浙江丰源泵业有限公司 | Water pump fault prediction and diagnosis method based on machine learning |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103994183A (en) * | 2013-02-19 | 2014-08-20 | 住友重机械工业株式会社 | Eccentrically-swinging reduction gear |
CN112700060A (en) * | 2021-01-08 | 2021-04-23 | 佳源科技股份有限公司 | Station terminal load prediction method and prediction device |
CN112766343A (en) * | 2021-01-12 | 2021-05-07 | 郑州轻工业大学 | Network security situation assessment method based on improved WOA-SVM |
CN112943595A (en) * | 2021-02-07 | 2021-06-11 | 三一重工股份有限公司 | Hydraulic pump fault prediction method, hydraulic pump fault prediction device, electronic equipment and storage medium |
CN113281620A (en) * | 2021-07-22 | 2021-08-20 | 广东电网有限责任公司惠州供电局 | Fault section positioning method, system and medium based on self-adaptive whale algorithm |
CN115049041A (en) * | 2022-04-15 | 2022-09-13 | 西安石油大学 | Moisture pipe liquid holding rate prediction method based on WOA-BP neural network |
CN115238961A (en) * | 2022-06-23 | 2022-10-25 | 合肥工业大学 | Tire vulcanization quality prediction method based on whale algorithm optimization BP neural network |
-
2023
- 2023-02-27 CN CN202310190590.8A patent/CN116292245A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103994183A (en) * | 2013-02-19 | 2014-08-20 | 住友重机械工业株式会社 | Eccentrically-swinging reduction gear |
CN112700060A (en) * | 2021-01-08 | 2021-04-23 | 佳源科技股份有限公司 | Station terminal load prediction method and prediction device |
CN112766343A (en) * | 2021-01-12 | 2021-05-07 | 郑州轻工业大学 | Network security situation assessment method based on improved WOA-SVM |
CN112943595A (en) * | 2021-02-07 | 2021-06-11 | 三一重工股份有限公司 | Hydraulic pump fault prediction method, hydraulic pump fault prediction device, electronic equipment and storage medium |
CN113281620A (en) * | 2021-07-22 | 2021-08-20 | 广东电网有限责任公司惠州供电局 | Fault section positioning method, system and medium based on self-adaptive whale algorithm |
CN115049041A (en) * | 2022-04-15 | 2022-09-13 | 西安石油大学 | Moisture pipe liquid holding rate prediction method based on WOA-BP neural network |
CN115238961A (en) * | 2022-06-23 | 2022-10-25 | 合肥工业大学 | Tire vulcanization quality prediction method based on whale algorithm optimization BP neural network |
Non-Patent Citations (2)
Title |
---|
欧阳城添;刘青云;: "IWOA和BP神经网络的压缩机故障检测算法", 井冈山大学学报(自然科学版), no. 03, 15 May 2020 (2020-05-15), pages 45 - 53 * |
欧阳城添等: "IWOA和BP神经网络的压缩机故障检测算法", 井冈山大学学报(自然科学版), vol. 41, no. 3, 31 May 2020 (2020-05-31), pages 45 - 53 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117828403A (en) * | 2024-01-03 | 2024-04-05 | 浙江丰源泵业有限公司 | Water pump fault prediction and diagnosis method based on machine learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112257341B (en) | Customized product performance prediction method based on heterogeneous data difference compensation fusion | |
CN112577747A (en) | Rolling bearing fault diagnosis method based on space pooling network | |
CN107392919B (en) | Adaptive genetic algorithm-based gray threshold acquisition method and image segmentation method | |
CN108710771B (en) | Based on the integrated mechanized equipment service reliability appraisal procedure extracted of depth characteristic | |
CN111967486A (en) | Complex equipment fault diagnosis method based on multi-sensor fusion | |
CN110929848B (en) | Training and tracking method based on multi-challenge perception learning model | |
CN112183742B (en) | Neural network hybrid quantization method based on progressive quantization and Hessian information | |
CN110458213A (en) | A kind of disaggregated model robust performance appraisal procedure | |
CN111414849A (en) | Face recognition method based on evolution convolutional neural network | |
CN109657147B (en) | Microblog abnormal user detection method based on firefly and weighted extreme learning machine | |
CN116292245A (en) | Piston pump voiceprint fault detection method | |
CN115580445B (en) | Unknown attack intrusion detection method, unknown attack intrusion detection device and computer readable storage medium | |
CN112199670B (en) | Log monitoring method for improving IFOREST (entry face detection sequence) to conduct abnormity detection based on deep learning | |
CN115412301B (en) | Predictive analysis method and system for network security | |
CN115575125A (en) | Bearing fault diagnosis method based on GADF-GAN-AVOA-CNN | |
CN114861880A (en) | Industrial equipment fault prediction method and device based on cavity convolutional neural network | |
CN116015967A (en) | Industrial Internet intrusion detection method based on improved whale algorithm optimization DELM | |
CN115905855A (en) | Improved meta-learning algorithm MG-copy | |
CN115344693A (en) | Clustering method based on fusion of traditional algorithm and neural network algorithm | |
CN113486569A (en) | Method for predicting residual life of centrifugal pump | |
CN107480724A (en) | A kind of determination method of cluster centre, determine system and a kind of clustering method | |
CN116503147A (en) | Financial risk prediction method based on deep learning neural network | |
US20230186091A1 (en) | Method and device for determining task-driven pruning module, and computer readable storage medium | |
CN113610148B (en) | Fault diagnosis method based on bias weighted AdaBoost | |
CN112529637B (en) | Service demand dynamic prediction method and system based on context awareness |
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 |