Self-starting permanent magnet synchronous motor fault identification method and system based on timing analysis
Through the self-starting method based on timing analysis, the problem of low intelligence of permanent magnet synchronous motor fault recognition is solved, the effect and reliability of complex motor fault recognition is improved, and the accurate identification of motor operating status faults is achieved.
Patent Information
- Application Number
- CN202410950870.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-07-16
AI Technical Summary
In the prior art, the fault identification method of permanent magnet synchronous motor is relatively low, resulting in poor fault identification effect in the operating state of complex motors, and it is difficult to improve the reliability of fault identification.
Using a self-start method based on timing analysis, the target motor is divided into nodes, generates a running window, extracts historical running data, performs parameter situation identification and timing search, builds operation constraints, monitors in real time and starts sequence constraint authentication. Generate fault identification instructions and results.
It improves the fault identification effect and reliability of complex motor operating states, and realizes accurate identification of faults in various operating states during motor operation.
Smart Images

Figure CN118508829B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field related to electrical variable measurement technology, and specifically to a self-starting permanent magnet synchronous motor fault identification method and system based on timing analysis. Background Art
[0002] With the rapid development of industrial automation and intelligent manufacturing, the monitoring and management of the operating status of motors, as the core power equipment in industrial systems, has become particularly important. Traditional motor monitoring systems often rely on simple threshold judgments or empirical rules. The methods in the prior art have certain limitations in identifying faults in complex motor operating states, and the fault identification effect is poor and the accuracy is low.
[0003] Therefore, in the prior art, the permanent magnet synchronous motor fault identification method has low intelligence and poor fault identification effect for complex motor operating states, resulting in a technical problem that it is difficult to improve the reliability of motor fault identification. Summary of the invention
[0004] The present application provides a method and system for self-starting permanent magnet synchronous motor fault identification based on timing analysis, which solves the technical problem that the permanent magnet synchronous motor fault identification method in the prior art is low in intelligence, has poor fault identification effect for complex motor operating states, and makes it difficult to improve the reliability of motor fault identification. The corresponding identification of faults existing in each operating state during the operation of the motor is achieved, which improves the fault identification effect and identification reliability of complex motor operating states.
[0005] The present application provides a self-starting permanent magnet synchronous motor fault identification method based on timing analysis, the method comprising: performing node division of the start-stop cycle based on the model of the target motor, and generating M operation windows. Based on the target motor, historical operation data of M operation windows are extracted to obtain M historical operation data sets. Parameter situation identification is performed on the M historical operation data sets respectively to generate M operation index parameter value sets. With the operation index as the index, the M operation index parameter value sets are time-series searched to obtain K operation index parameter value sequences. A first operation constraint is constructed, wherein the first operation constraint is that the parameter value difference between two adjacent operation windows of the K operation indexes needs to satisfy the parameter value difference between two adjacent operation windows in the K operation index parameter value sequence. The startup process of the target motor is monitored in real time to generate a real-time monitoring data set, wherein the real-time monitoring data set includes a real-time operation window. Based on the first operation constraint and the real-time monitoring data set, a startup sequence constraint authentication is performed, and when the authentication passes, a fault identification instruction is obtained. According to the fault identification instruction, a real-time operation index parameter set corresponding to the real-time operation window is retrieved, fault identification is performed on the real-time monitoring data set, and a fault identification result is generated.
[0006] In an implementation, the method includes: retrieving monitoring data of the target motor in a previous operation window of the real-time operation window to generate a previous window monitoring data set. Using the first operation constraint, the previous window monitoring data set and the real-time monitoring data set are subjected to startup sequence constraint authentication, and if the authentication is passed, a fault identification instruction is obtained.
[0007] In the implementation, the method includes: extracting the parameter values of K operating indicators in the previous window monitoring data set and the real-time monitoring data set respectively, and generating K previous window operating indicator parameter values and K real-time operating indicator parameter values. Determine respectively whether the difference between the K previous window operating indicator parameter values and the K real-time operating indicator parameter values meets the first operating constraint, and if so, add them to the qualified indicator set. Calculate the ratio of the number of indicators in the qualified indicator set to the K operating indicators, and generate a real-time indicator pass rate. When the real-time indicator pass rate meets the preset indicator pass rate, the authentication is passed.
[0008] In an implementation, the method includes: performing consistency identification on the M historical operation data sets respectively to generate M consistency factors. Determining whether the M consistency factors are greater than or equal to a preset consistency factor, and if so, generating a data supplement instruction. Based on the data supplement instruction, performing data retrieval from the cloud to obtain M supplementary operation data sets. Using the M supplementary operation data sets to update the M historical operation data sets.
[0009] In an implementation, the method includes: extracting a first historical operation data set from the M historical operation data sets. Using K operation indicators as indexes, searching the first historical operation data set to generate K first historical operation indicator parameter value sets. Performing parameter situation identification on the K first historical operation indicator parameter value sets respectively to generate a first operation indicator parameter value set, wherein the first operation indicator parameter value set includes K first operation indicator parameter values. Performing parameter situation identification on the M historical operation data sets to generate M operation indicator parameter value sets.
[0010] In an implementation, the method includes: extracting a first first historical operation indicator parameter set from the K first historical operation indicator parameter sets. Calculating the parameter mean of the first first historical operation indicator parameter set to generate a first first historical parameter mean. Taking the first first historical parameter mean as the interval center and combining with a preset interval radius to construct a mean interval. Counting the density of the mean interval to generate a mean interval density.
[0011] In the implementation, a mean interval density is generated, and then the method further includes: taking the two endpoints of the mean interval as the interval center, constructing a left interval and a right interval, wherein the left interval has a left interval density, and the right interval has a right interval density. Density optimization is performed according to the mean interval density, the left interval density, and the right interval density, and the interval corresponding to the maximum interval density is used as the stage interval. According to the stage interval, iterative optimization is performed in the first first historical operation indicator parameter set until a preset number of iterations is met, and the stage interval corresponding to the maximum interval density is used as the target interval. Calculate the mean of multiple first historical operation indicator parameters within the target interval to obtain the first first operation indicator parameter value. Analyze the K first historical operation indicator parameter sets to generate K first operation indicator parameter values.
[0012] The present application also provides a self-starting permanent magnet synchronous motor fault identification system based on timing analysis, including:
[0013] The operation window acquisition module is used to divide the nodes of the start-stop cycle based on the model of the target motor and generate M operation windows.
[0014] The historical data acquisition module is used to extract historical operation data of M operation windows based on the target motor to obtain M historical operation data sets.
[0015] The indicator parameter value acquisition module is used to perform parameter situation identification on the M historical operation data sets respectively to generate M operation indicator parameter value sets.
[0016] The parameter value sequence acquisition module is used to perform time-series retrieval on the M operating indicator parameter value sets with the operating indicator as an index to obtain K operating indicator parameter value sequences.
[0017] The operation constraint acquisition module is used to construct a first operation constraint, wherein the first operation constraint is that the parameter value difference between two adjacent operation windows of K operation indicators must satisfy the parameter value difference between two adjacent operation windows in the K operation indicator parameter value sequence.
[0018] The real-time operation monitoring module is used to monitor the startup process of the target motor in real time and generate a real-time monitoring data set, wherein the real-time monitoring data set includes a real-time operation window.
[0019] The constraint authentication module is used to perform a startup sequence constraint authentication based on the first operation constraint and the real-time monitoring data set, and obtain a fault identification instruction when the authentication passes.
[0020] The fault identification module is used to retrieve the real-time operation index parameter set corresponding to the real-time operation window according to the fault identification instruction, perform fault identification on the real-time monitoring data set, and generate a fault identification result.
[0021] The self-starting permanent magnet synchronous motor fault identification method and system based on timing analysis proposed in this application is proposed to generate M operation windows by dividing the nodes of the start-stop cycle based on the model of the target motor. The historical operation data of the M operation windows are extracted based on the target motor to obtain M historical operation data sets. The parameter situation identification is performed on the M historical operation data sets respectively to generate M operation index parameter value sets. With the operation index as the index, the M operation index parameter value sets are retrieved in time series to obtain K operation index parameter value sequences. A first operation constraint is constructed, wherein the first operation constraint is that the parameter value difference between two adjacent operation windows of the K operation indicators needs to satisfy the parameter value difference between two adjacent operation windows in the K operation index parameter value sequence. The startup process of the target motor is monitored in real time to generate a real-time monitoring data set, wherein the real-time monitoring data set includes a real-time operation window. The startup sequence constraint authentication is performed based on the first operation constraint and the real-time monitoring data set, and when the authentication passes, a fault identification instruction is obtained. According to the fault identification instruction, the real-time operation index parameter set corresponding to the real-time operation window is retrieved, the real-time monitoring data set is subjected to fault identification, and a fault identification result is generated. The technical problem that the fault identification method of permanent magnet synchronous motor in the prior art is low in intelligence, the fault identification effect for complex motor operation states is poor, and the reliability of motor fault identification is difficult to improve is solved. The corresponding identification of faults existing in each operation state during the operation of the motor is achieved, and the fault identification effect and identification reliability of complex motor operation states are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.
[0023] Figure 1 A schematic flow chart of a method for identifying faults of a self-starting permanent magnet synchronous motor based on timing analysis provided in an embodiment of the present application;
[0024] Figure 2 A schematic diagram of the structure of a self-starting permanent magnet synchronous motor fault identification system based on timing analysis provided in an embodiment of the present application.
[0025] Explanation of the reference numerals: operation window acquisition module 11, historical data acquisition module 12, indicator parameter value acquisition module 13, parameter value sequence acquisition module 14, operation constraint acquisition module 15, real-time operation monitoring module 16, constraint authentication module 17, fault identification module 18. DETAILED DESCRIPTION
[0026] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.
[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.
[0028] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the term "first" involved is merely to distinguish similar objects and does not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.
[0029] The present application provides a method and system for identifying faults of a self-starting permanent magnet synchronous motor based on timing analysis, such as Figure 1 As shown, the method includes:
[0030] The nodes of the start-stop cycle are divided based on the model of the target motor to generate M operation windows.
[0031] Historical operation data of M operation windows are extracted based on the target motor to obtain M historical operation data sets.
[0032] Parameter situation identification is performed on each of the M historical operation data sets to generate M operation indicator parameter value sets.
[0033] Taking the operation index as an index, a time series search is performed on the M operation index parameter value sets to obtain K operation index parameter value sequences.
[0034] Based on the model of the target motor, the node division of the start-stop cycle is carried out, and the specific model of the target motor is analyzed in detail to obtain the technical parameters, motor characteristic curve, load characteristics, etc. of the target motor. M operation windows are generated based on the motor characteristic curve. The M operation windows are applied to the actual operation of the motor, and the start and stop of the motor are managed by the automatic control system. The M operation windows include the operation stages from the start to the normal operation of the motor, and the final stop, including: the start stage, the constant torque zone, the constant power zone, the rated operation point, etc. Based on the target motor, the historical operation data of the M operation windows are extracted, and the historical operation data includes the current, voltage, power, torque, speed, temperature, operation time and other parameter data generated by the motor during the historical operation of different operation windows, and M historical operation data sets are obtained, each of which corresponds to an operation window. Further, the M historical operation data sets are respectively identified with parameter status, that is, the operation index parameter values of the corresponding parameters of the operation windows in the M historical operation data sets are obtained, and M operation index parameter value sets are generated based on the operation index parameters corresponding to the M operation windows. Finally, taking the operation index as an index, the M operation index parameter value sets are time-series searched and arranged based on the sequence of the M operation windows to obtain K operation index parameter value sequences.
[0035] A first operation constraint is constructed, wherein the first operation constraint is that a parameter value difference between two adjacent operation windows of the K operation indicators must satisfy a parameter value difference between two adjacent operation windows in the parameter value sequence of the K operation indicators.
[0036] The startup process of the target motor is monitored in real time to generate a real-time monitoring data set, wherein the real-time monitoring data set includes a real-time operation window.
[0037] A startup sequence constraint authentication is performed based on the first operation constraint and the real-time monitoring data set, and when the authentication passes, a fault identification instruction is obtained.
[0038] According to the fault identification instruction, a real-time operation index parameter set corresponding to the real-time operation window is retrieved, fault identification is performed on the real-time monitoring data set, and a fault identification result is generated.
[0039] Construct a first operation constraint, wherein the first operation constraint is that the parameter value difference between two adjacent operation windows of the K operation indicators needs to satisfy the parameter value difference between two adjacent operation windows in the K operation indicator parameter value sequence, that is, the corresponding indicator parameters in different operation windows should be in different states, therefore, the parameter value difference between two adjacent operation windows should satisfy the parameter value difference between two adjacent operation windows in the acquired operation indicator parameter value sequence. When the parameter value difference between two adjacent operation windows in the operation indicator parameter value sequence is not satisfied, the parameters corresponding to the two adjacent operation windows of the K operation indicators actually obtained may be abnormal, thereby making a preliminary screening and judgment of the motor operation fault, thereby improving the fault identification efficiency, and storing the standard range value of the parameter value difference between adjacent different operation windows in the first operation constraint. Monitor the startup process of the target motor in real time, obtain the real-time monitored current, voltage, power, torque, speed, temperature, operation time and other parameters, generate a real-time monitoring data set, and the real-time monitoring data set contains a specific operation window. Based on the first operation constraint and the real-time monitoring data set, the startup sequence constraint authentication is performed to determine whether the parameter value difference between the operation windows in the real-time monitoring data set meets the first operation constraint requirement. When the parameter value difference between two adjacent operation windows in the corresponding indicator parameter value sequence in the first operation constraint requirement is met, the authentication is passed and a fault identification instruction is obtained. According to the fault identification instruction, the real-time operation index parameter set corresponding to the real-time operation window is retrieved, and the real-time monitoring data set is fault identified to generate a fault identification result. When performing fault identification, the real-time monitoring data set and the corresponding real-time operation window are obtained, and the real-time monitoring data set and the corresponding real-time operation window are input into the fault identification model for fault identification, and the result of the model output is obtained to obtain the fault identification result. Among them, the fault identification model is constructed based on the neural network model. When constructing, the historical motor operation parameters, the corresponding operation window and the fault identification data are used as training data, and the neural network model is supervised and trained until the output result of the model meets the preset accuracy. The training of the model is completed and the fault identification model is obtained. The fault identification model realizes the identification of motor operation faults by inputting real-time monitoring data and the corresponding operation window. The invention solves the technical problem that the fault identification method of permanent magnet synchronous motor in the prior art is low in intelligence, the fault identification effect for complex motor operating states is poor, and the reliability of motor fault identification is difficult to improve. The invention realizes the corresponding identification of faults existing in each operating state during the operation of the motor, and improves the fault identification effect and identification reliability of complex motor operating states.
[0040] The method provided in the embodiment of the present application also includes:
[0041] The monitoring data of the target motor in a previous operation window of the real-time operation window is retrieved to generate a previous window monitoring data set.
[0042] The first operation constraint is used to perform startup sequence constraint authentication on the previous window monitoring data set and the real-time monitoring data set. If the authentication passes, a fault identification instruction is obtained.
[0043] Retrieve the monitoring data of the target motor in the previous operation window of the real-time operation window, generate a previous window monitoring data set, and include various motor startup monitoring parameters in the previous window monitoring data set. Subsequently, use the first operation constraint to perform constraint authentication on the startup sequence of the previous window monitoring data set and the real-time monitoring data set to determine whether the parameter difference between the two meets the requirements. If the authentication is passed, obtain a fault identification instruction. By performing startup sequence constraint authentication on the motor, the preliminary judgment and screening of motor abnormalities are completed. When there is an obvious abnormality, no subsequent operation is required. Otherwise, a fault identification instruction is output for further fault identification, further improving the efficiency of fault identification processing.
[0044] The method provided in the embodiment of the present application also includes:
[0045] Parameter values of K operating indicators in the previous window monitoring data set and the real-time monitoring data set are extracted respectively to generate K previous window operating indicator parameter values and K real-time operating indicator parameter values.
[0046] Determine whether the differences between the K previous window operation index parameter values and the K real-time operation index parameter values meet the first operation constraint, and if so, add them into the qualified index set.
[0047] The ratio of the number of indicators in the qualified indicator set to the K operating indicators is calculated to generate a real-time indicator pass rate.
[0048] When the real-time indicator pass rate meets the preset indicator pass rate, the authentication is passed.
[0049] When performing constraint authentication, extract the parameter values of K operating indicators in the previous window monitoring data set and the real-time monitoring data set respectively, generate K previous window operating indicator parameter values and K real-time operating indicator parameter values, and the indicator parameters in the K previous window operating indicator parameter values and the K real-time operating indicator parameter values correspond to each other. Obtain the difference between the K previous window operating indicator parameter values and the corresponding K real-time operating indicator parameter values, and perform operating constraint judgment on the obtained K differences, wherein K is the indicator category of the monitoring data, and judge whether the difference between the K previous window operating indicator parameter values and the K real-time operating indicator parameter values meets the difference standard range value recorded in the first operating constraint. If so, add them to the qualified indicator set. Exemplary: extract current parameter value A1, voltage parameter value B1, and temperature parameter value C1 from the previous window monitoring data set. Extract the corresponding current parameter value A2, voltage parameter value B2, and temperature parameter value C2 from the real-time monitoring data set. The difference between the corresponding indicator parameter values in the two data sets, i.e., A=A2-A1, B=B2-B1, and C=C2-C1. According to the difference standard value recorded in the first operation constraint, for example, the difference standard of the current parameter value is 2A, the voltage difference standard is 10V, and the temperature difference standard is 5°C. If A, B, and C are respectively the same as the corresponding difference standards, the corresponding indicator is deemed to meet the operation constraint, and the corresponding indicator that meets the operation constraint is added to the qualified indicator set. Further, the proportion of the number of indicators in the qualified indicator set to the K operation indicators is calculated to generate a real-time indicator pass rate. Finally, when the real-time indicator pass rate meets the preset indicator pass rate, the certification is passed, wherein the preset indicator pass rate is the minimum required threshold of the preset indicator pass rate. When it is less than or equal to the preset indicator pass rate, the corresponding indicator pass ratio is small, and the possibility of abnormalities in the operation data of the motor in different operation windows is high, otherwise the probability of abnormalities is small. By performing startup sequence constraint authentication on the motor, preliminary judgment and screening of motor abnormalities can be completed. When obvious abnormalities exist, no subsequent operations are required, further improving the efficiency of fault identification and processing.
[0050] The method provided in the embodiment of the present application also includes:
[0051] Consistency identification is performed on the M historical operation data sets respectively to generate M consistency factors.
[0052] Determine whether the M consistency factors are greater than or equal to a preset consistency factor, and if so, generate a data supplement instruction.
[0053] Based on the data supplementation instruction, data is retrieved from the cloud to obtain M sets of supplementary operating data.
[0054] The M historical operation data sets are updated using the M supplementary operation data sets.
[0055] Consistency identification is performed on the M historical operation data sets obtained respectively, and M consistency factors are generated. When performing consistency identification, the repetition ratio of the parameters in the historical operation data sets is obtained respectively, that is, the ratio of the parameters with the highest frequency to the total parameters, and the consistency factor is obtained based on the ratio. M consistency factors are obtained and it is determined whether they are greater than or equal to the preset consistency factor. The preset consistency factor is the maximum preset threshold of the consistency factor. When it is greater than or equal to the threshold, the consistency of the data is too high and the data is too simplistic, resulting in a decrease in the representativeness of the data, and it is difficult to represent the operating conditions of the motor in the operating window, then a data supplement instruction is generated. Otherwise, the corresponding historical operation data set meets the requirements and no data supplementation is required. Based on the data supplement instruction, data retrieval is performed from the cloud to obtain M supplementary operation data sets, that is, M supplementary operation data sets are obtained from the cloud based on the data supplement instruction. Finally, the M historical operation data sets are updated using the M supplementary operation data sets to ensure the representativeness of the historical operation data sets and improve the accuracy and universality of the analysis results.
[0056] The method provided in the embodiment of the present application also includes:
[0057] A first historical operation data set is extracted from the M historical operation data sets.
[0058] The first historical operation data set is searched using the K operation indicators as indexes to generate K first historical operation indicator parameter value sets.
[0059] Parameter situation identification is performed on each of the K first historical operation indicator parameter value sets to generate a first operation indicator parameter value set, wherein the first operation indicator parameter value set includes K first operation indicator parameter values.
[0060] Parameter situation identification is performed on the M historical operation data sets to generate M operation indicator parameter value sets.
[0061] A first historical operation data set is extracted from the M historical operation data sets, wherein the first historical operation data set is a historical operation data set of a random window in the M operation windows. The first historical operation data set is retrieved with K operation indicators as indexes to generate K first historical operation indicator parameter value sets, each of which corresponds to an indicator parameter in the first historical operation data set. Subsequently, parameter situation identification is performed on the K first historical operation indicator parameter value sets respectively to generate a first operation indicator parameter value set, wherein the first operation indicator parameter value set includes K first operation indicator parameter values. Finally, parameter situation identification is completed for the M historical operation data sets corresponding to all operation windows to generate M operation indicator parameter value sets.
[0062] The method provided in the embodiment of the present application also includes:
[0063] A first first historical operation indicator parameter set is extracted from the K first historical operation indicator parameter sets.
[0064] Calculate the parameter mean of the first first historical operation indicator parameter set to generate a first first historical parameter mean.
[0065] The mean value interval is constructed by taking the first historical parameter mean value as the interval center and combining it with the preset interval radius.
[0066] The density of the mean interval is counted to generate the mean interval density.
[0067] Obtaining a set of operating indicator parameter values, including: extracting a first first historical operating indicator parameter set from K first historical operating indicator parameter sets. Calculating the parameter mean of the corresponding indicator parameter based on the first first historical operating indicator parameter set. Further, taking the first first historical parameter mean as the center of the interval, and combining with a preset interval radius to construct a mean interval, wherein the preset interval radius is the preferred range of the parameter, and the specific setting can set different interval radii based on different indicator parameters. Taking the first historical parameter mean as the midpoint of the interval, and taking the preset interval radius as the left and right parameter traversal range, the construction of the mean interval is completed. Finally, the amount of data in the first first historical operating indicator parameter set that is in the mean interval is counted, the density of the mean interval is obtained based on the ratio of the amount of data to the interval length, and the mean interval density is generated.
[0068] The method provided in the embodiment of the present application also includes:
[0069] Taking the two endpoints of the mean interval as the interval centers, a left interval and a right interval are constructed, wherein the left interval has a left interval density, and the right interval has a right interval density.
[0070] Density optimization is performed according to the mean interval density, the left interval density and the right interval density, and the interval corresponding to the maximum interval density is used as the stage interval.
[0071] According to the stage interval, iterative optimization is performed in the first first historical operation indicator parameter set until a preset number of iterations is met, and the stage interval corresponding to the maximum interval density is used as the target interval.
[0072] Calculate the average of multiple first historical operating indicator parameters within the target interval to obtain a first first operating indicator parameter value.
[0073] The K first historical operation indicator parameter sets are analyzed to generate K first operation indicator parameter values.
[0074] Generate the mean interval density, and then the method also includes: taking the two endpoints of the mean interval as the interval center, constructing the left interval and the right interval, that is, taking the left and right endpoints of the original mean interval as the midpoint and taking the preset interval radius as the interval length to obtain the left interval of the left endpoint and the right interval of the right endpoint, wherein the left interval has a left interval density, and the right interval has a right interval density, and the left interval density is the ratio of the left interval data volume to the left interval length, and the right interval density is the ratio of the right interval data volume to the right interval length. According to the mean interval density, the left interval density and the right interval density, density optimization is performed to obtain the interval with the largest data density, and the interval corresponding to the maximum interval density is used as the stage interval. Subsequently, taking the current stage interval as the starting point, the left interval and the right interval are obtained, and the corresponding interval density is obtained for comparison, and the interval with the largest data density is obtained as the stage interval, and the above steps are continuously looped, and iterative optimization is performed in the first first historical operation indicator parameter set until the preset number of iterations is met, and the stage interval corresponding to the maximum interval density is obtained. The preset number of iterations is the preset number of loop iterations. When the number of iterations is met, the final output stage interval is obtained, the final optimized stage interval is used as the final target interval, the mean of multiple first historical operation index parameters in the target interval is calculated, and the first first operation index parameter value is obtained. The K first historical operation index parameter sets are analyzed in the same way to generate K first operation index parameter values.
[0075] In the above, refer to Figure 1 The present invention describes in detail a method for identifying faults of a self-starting permanent magnet synchronous motor based on timing analysis according to an embodiment of the present invention. Figure 2 A self-starting permanent magnet synchronous motor fault identification system based on timing analysis according to an embodiment of the present invention is described.
[0076] According to the self-starting permanent magnet synchronous motor fault identification system based on timing analysis of the embodiment of the present invention, the technical problem that the permanent magnet synchronous motor fault identification method in the prior art is low in intelligence, the fault identification effect for complex motor operating states is poor, and the reliability of motor fault identification is difficult to improve is solved. The corresponding identification of faults existing in each operating state during the operation of the motor is achieved, and the fault identification effect and identification reliability of complex motor operating states are improved. The self-starting permanent magnet synchronous motor fault identification system based on timing analysis includes: an operation window acquisition module 11, a historical data acquisition module 12, an indicator parameter value acquisition module 13, a parameter value sequence acquisition module 14, an operation constraint acquisition module 15, a real-time operation monitoring module 16, a constraint authentication module 17, and a fault identification module 18.
[0077] The operation window acquisition module 11 is used to divide the nodes of the start-stop cycle based on the model of the target motor and generate M operation windows.
[0078] The historical data acquisition module 12 is used to extract historical operation data of M operation windows based on the target motor to obtain M historical operation data sets.
[0079] The indicator parameter value acquisition module 13 is used to perform parameter situation identification on the M historical operation data sets respectively to generate M operation indicator parameter value sets.
[0080] The parameter value sequence acquisition module 14 is used to perform time series retrieval on the M operating indicator parameter value sets with the operating indicator as an index to obtain K operating indicator parameter value sequences.
[0081] The operation constraint acquisition module 15 is used to construct a first operation constraint, wherein the first operation constraint is that the parameter value difference between two adjacent operation windows of the K operation indicators must satisfy the parameter value difference between two adjacent operation windows in the K operation indicator parameter value sequence.
[0082] The real-time operation monitoring module 16 is used to monitor the startup process of the target motor in real time and generate a real-time monitoring data set, wherein the real-time monitoring data set includes a real-time operation window.
[0083] The constraint authentication module 17 is used to perform a startup sequence constraint authentication based on the first operation constraint and the real-time monitoring data set, and obtain a fault identification instruction when the authentication passes.
[0084] The fault identification module 18 is used to retrieve the real-time operation index parameter set corresponding to the real-time operation window according to the fault identification instruction, perform fault identification on the real-time monitoring data set, and generate a fault identification result.
[0085] The specific configuration of the constraint authentication module 17 will be described in detail below. The constraint authentication module 17 may further include: retrieving the monitoring data of the target motor in the previous operation window of the real-time operation window to generate a previous window monitoring data set. Using the first operation constraint, the previous window monitoring data set and the real-time monitoring data set are subjected to startup sequence constraint authentication, and if the authentication is passed, a fault identification instruction is obtained.
[0086] The specific configuration of the constraint authentication module 17 will be described in detail below. The constraint authentication module 17 further includes: extracting the parameter values of K operating indicators in the previous window monitoring data set and the real-time monitoring data set respectively, and generating K previous window operating indicator parameter values and K real-time operating indicator parameter values. Determine respectively whether the difference between the K previous window operating indicator parameter values and the K real-time operating indicator parameter values meets the first operating constraint, and if so, add them to the qualified indicator set. Calculate the ratio of the number of indicators in the qualified indicator set to the K operating indicators, and generate the real-time indicator pass rate. When the real-time indicator pass rate meets the preset indicator pass rate, the authentication is passed.
[0087] The specific configuration of the historical data acquisition module 12 will be described in detail below. The historical data acquisition module 12 may further include: performing consistency identification on the M historical operation data sets respectively to generate M consistency factors. Determine whether the M consistency factors are greater than or equal to the preset consistency factors, and if so, generate data supplement instructions. Based on the data supplement instructions, retrieve data from the cloud to obtain M supplementary operation data sets. Use the M supplementary operation data sets to update the M historical operation data sets.
[0088] The specific configuration of the indicator parameter value acquisition module 13 will be described in detail below. The indicator parameter value acquisition module 13 further includes: extracting a first historical operation data set from the M historical operation data sets. Using K operation indicators as indexes, the first historical operation data set is searched to generate K first historical operation indicator parameter value sets. Parameter situation identification is performed on the K first historical operation indicator parameter value sets respectively to generate a first operation indicator parameter value set, wherein the first operation indicator parameter value set includes K first operation indicator parameter values. Parameter situation identification is performed on the M historical operation data sets to generate M operation indicator parameter value sets.
[0089] The specific configuration of the indicator parameter value acquisition module 13 will be described in detail below. The indicator parameter value acquisition module 13 further includes: extracting a first first historical operation indicator parameter set from the K first historical operation indicator parameter sets. Calculating the parameter mean of the first first historical operation indicator parameter set to generate a first first historical parameter mean. Taking the first first historical parameter mean as the interval center and combining with a preset interval radius to construct a mean interval. Counting the density of the mean interval to generate a mean interval density.
[0090] Below, the specific configuration of the indicator parameter value acquisition module 13 will be described in detail. The indicator parameter value acquisition module 13 further includes: generating a mean interval density, and then the method further includes: taking the two endpoints of the mean interval as the interval center, constructing a left interval and a right interval, wherein the left interval has a left interval density, and the right interval has a right interval density. Density optimization is performed according to the mean interval density, the left interval density, and the right interval density, and the interval corresponding to the maximum interval density is used as the stage interval. According to the stage interval, iterative optimization is performed in the first first historical operation indicator parameter set until the preset number of iterations is met, and the stage interval corresponding to the maximum interval density is used as the target interval. Calculate the mean of multiple first historical operation indicator parameters within the target interval to obtain the first first operation indicator parameter value. Analyze the K first historical operation indicator parameter sets to generate K first operation indicator parameter values.
[0091] The self-starting permanent magnet synchronous motor fault identification system based on timing analysis provided in an embodiment of the present invention can execute the self-starting permanent magnet synchronous motor fault identification method based on timing analysis provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0092] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.
[0093] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.
Claims
1. A self-starting permanent magnet synchronous motor fault identification method based on timing analysis, characterized in that: The method comprises: The nodes of the start-stop cycle are divided based on the model of the target motor to generate M operation windows; Extract historical operation data of M operation windows based on the target motor to obtain M historical operation data sets; Performing parameter situation identification on the M historical operation data sets respectively to generate M operation indicator parameter value sets; Using the operation index as an index, performing a time series search on the M operation index parameter value sets to obtain K operation index parameter value sequences; Constructing a first operation constraint, wherein the first operation constraint is that a parameter value difference between two adjacent operation windows of the K operation indicators must satisfy a parameter value difference between two adjacent operation windows in the K operation indicator parameter value sequence; Performing real-time monitoring on the startup process of the target motor to generate a real-time monitoring data set, wherein the real-time monitoring data set includes a real-time operation window; Performing startup sequence constraint authentication based on the first operation constraint and the real-time monitoring data set, and obtaining a fault identification instruction when the authentication passes; Retrieving the real-time monitoring data set corresponding to the real-time operation window according to the fault identification instruction, performing fault identification on the real-time monitoring data set and the real-time operation window using a fault identification model, and generating a fault identification result; Wherein, the method comprises: Extracting a first historical operation data set from the M historical operation data sets; Using the K operating indicators as indexes, searching the first historical operating data set to generate K first historical operating indicator parameter value sets; Performing parameter situation identification on the K first historical operation indicator parameter value sets respectively to generate a first operation indicator parameter value set, wherein the first operation indicator parameter value set includes K first operation indicator parameter values; Performing parameter situation identification on the M historical operation data sets to generate M operation indicator parameter value sets; Wherein, the method comprises: Extracting a first first historical operation indicator parameter set from the K first historical operation indicator parameter sets; Calculating the parameter mean of the first first historical operation indicator parameter set to generate a first first historical parameter mean; Taking the first historical parameter mean as the interval center and combining it with the preset interval radius to construct the mean interval; Counting the density of the mean interval to generate the mean interval density; Wherein, the mean interval density is generated, and then the method further comprises: Taking the two endpoints of the mean interval as the interval center, constructing a left interval and a right interval, wherein the left interval has a left interval density, and the right interval has a right interval density; Density optimization is performed according to the mean interval density, the left interval density and the right interval density, and the interval corresponding to the maximum interval density is used as the stage interval; Iteratively search for the best among the first historical operation indicator parameter set according to the stage interval until a preset number of iterations is met, and take the stage interval corresponding to the maximum interval density as the target interval; Calculating the average of multiple first historical operating indicator parameters within the target interval to obtain a first first operating indicator parameter value; The K first historical operation indicator parameter sets are analyzed to generate K first operation indicator parameter values.
2. The method for identifying faults of a self-starting permanent magnet synchronous motor based on timing analysis according to claim 1, characterized in that: The method comprises: Retrieving monitoring data of the target motor in a previous operation window of the real-time operation window to generate a previous window monitoring data set; The first operation constraint is used to perform startup sequence constraint authentication on the previous window monitoring data set and the real-time monitoring data set. If the authentication passes, a fault identification instruction is obtained.
3. The method for identifying faults of a self-starting permanent magnet synchronous motor based on timing analysis according to claim 2, characterized in that: The method comprises: Respectively extracting parameter values of K operating indicators in the previous window monitoring data set and the real-time monitoring data set to generate K previous window operating indicator parameter values and K real-time operating indicator parameter values; Determine respectively whether the differences between the K previous window operation index parameter values and the K real-time operation index parameter values meet the first operation constraint, and if so, add them into the qualified index set; Calculate the ratio of the number of indicators in the qualified indicator set to the K operating indicators to generate a real-time indicator pass rate; When the real-time indicator pass rate meets the preset indicator pass rate, the authentication is passed.
4. The method for identifying faults of a self-starting permanent magnet synchronous motor based on timing analysis according to claim 1, characterized in that: The method comprises: Perform consistency identification on the M historical operation data sets respectively to generate M consistency factors; Determine whether the M consistency factors are greater than or equal to a preset consistency factor, and if so, generate a data supplement instruction; Based on the data supplementation instruction, retrieve data from the cloud to obtain M sets of supplementary operation data; The M historical operation data sets are updated using the M supplementary operation data sets.
5. A self-starting permanent magnet synchronous motor fault identification system based on timing analysis, characterized in that: The system comprises: An operation window acquisition module is used to divide the nodes of the start-stop cycle based on the model of the target motor and generate M operation windows; A historical data acquisition module, used for extracting historical operation data of M operation windows based on the target motor to obtain M historical operation data sets; An indicator parameter value acquisition module is used to perform parameter situation recognition on the M historical operation data sets respectively to generate M operation indicator parameter value sets; A parameter value sequence acquisition module is used to perform time series retrieval on the M operating indicator parameter value sets with the operating indicator as an index to obtain K operating indicator parameter value sequences; An operation constraint acquisition module, used to construct a first operation constraint, wherein the first operation constraint is that a parameter value difference between two adjacent operation windows of K operation indicators must satisfy a parameter value difference between two adjacent operation windows in the K operation indicator parameter value sequence; A real-time operation monitoring module, used for monitoring the startup process of the target motor in real time and generating a real-time monitoring data set, wherein the real-time monitoring data set includes a real-time operation window; A constraint authentication module, configured to perform a startup sequence constraint authentication based on the first operation constraint and the real-time monitoring data set, and obtain a fault identification instruction when the authentication passes; A fault identification module, configured to retrieve the real-time monitoring data set corresponding to the real-time operation window according to the fault identification instruction, perform fault identification on the real-time monitoring data set and the real-time operation window using a fault identification model, and generate a fault identification result; Wherein, the indicator parameter value acquisition module also includes: Extracting a first historical operation data set from the M historical operation data sets; Using the K operating indicators as indexes, searching the first historical operating data set to generate K first historical operating indicator parameter value sets; Performing parameter situation identification on the K first historical operation indicator parameter value sets respectively to generate a first operation indicator parameter value set, wherein the first operation indicator parameter value set includes K first operation indicator parameter values; Performing parameter situation identification on the M historical operation data sets to generate M operation indicator parameter value sets; Wherein, the indicator parameter value acquisition module also includes: Extracting a first first historical operation indicator parameter set from the K first historical operation indicator parameter sets; Calculating the parameter mean of the first first historical operation indicator parameter set to generate a first first historical parameter mean; Taking the first historical parameter mean as the interval center and combining it with the preset interval radius to construct the mean interval; Counting the density of the mean interval to generate the mean interval density; Among them, the indicator parameter value acquisition module also includes: Taking the two endpoints of the mean interval as the interval center, constructing a left interval and a right interval, wherein the left interval has a left interval density, and the right interval has a right interval density; Density optimization is performed according to the mean interval density, the left interval density and the right interval density, and the interval corresponding to the maximum interval density is used as the stage interval; Iteratively search for the best among the first historical operation indicator parameter set according to the stage interval until a preset number of iterations is met, and take the stage interval corresponding to the maximum interval density as the target interval; Calculating the average of multiple first historical operating indicator parameters within the target interval to obtain a first first operating indicator parameter value; The K first historical operation indicator parameter sets are analyzed to generate K first operation indicator parameter values.
