Intelligent Maintenance System and Method for Electric Actuators Based on Fault Prediction

Abstract

The present invention discloses an intelligent maintenance system and method for an electric actuator based on fault prediction, specifically relating to the technical field of equipment maintenance; by collecting historical operation data within a preset operation cycle, generating a historical health data set, extracting characteristic parameters, obtaining a historical characteristic parameter set, and constructing a trend change characteristic matrix according to the change trend of the historical characteristic parameters; using the trend change characteristic matrix to construct a multi-dimensional state space, calculating the state aggregation degree of fault events in the space, and determining an early warning index set; at the same time, collecting real-time operation data to extract real-time characteristic parameters, forming a real-time characteristic state vector, and performing a correlation calculation with the early warning index in the space to obtain a real-time risk correlation factor; finally, generating a risk probability prediction model according to the real-time risk correlation factor, the historical characteristic parameter set, and the trend change characteristic matrix, predicting the fault probability and outputting a maintenance scheduling suggestion to realize the intelligent maintenance of the electric actuator.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment maintenance, and more specifically, to an intelligent maintenance system and method for electric actuators based on fault prediction. Background Art

[0002] The current operating state of electric actuators is directly related to equipment safety and production efficiency. However, traditional maintenance methods mainly rely on fixed-period inspections, making it difficult to capture potential fault signals of equipment in a timely manner, often resulting in a lag in maintenance response after a fault occurs, increasing the risk of downtime and maintenance costs. With the rapid development of sensor and data acquisition technologies, the operating data of electric actuators is becoming increasingly rich, providing reliable data support for equipment condition assessment. However, there are still major technical problems in how to extract effective features from the massive data and build an accurate fault prediction model. The existing technologies for the intelligent maintenance of electric actuators are relatively scattered, lacking a unified and systematic analysis method.

[0003] To solve the above problems, a technical solution is provided as follows. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an intelligent maintenance system and method for electric actuators based on fault prediction to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] An intelligent maintenance method for electric actuators based on fault prediction, comprising the following steps:

[0007] Obtain the historical operating data of the electric actuator within a preset operating cycle, and generate a historical health data set of the electric actuator according to the historical operating data;

[0008] Extract characteristic parameters from the historical health data set to obtain a historical characteristic parameter set of the electric actuator, and establish a trend change characteristic matrix according to the changing trend of the historical characteristic parameters over time;

[0009] Construct a multi-dimensional state space based on the trend change characteristic matrix, calculate the state aggregation degree of the historical fault events of the electric actuator in the multi-dimensional state space, and determine a characteristic early warning index set according to the state aggregation degree;

[0010] Extract real-time characteristic parameters according to the real-time operating data of the electric actuator, establish a real-time characteristic state vector, and calculate the spatial position correlation with the characteristic early warning index set in the multi-dimensional state space to determine a real-time risk correlation factor;

[0011] Generate a risk probability prediction model based on real-time risk correlation factors, historical feature parameter sets, and trend change feature matrices, predict the failure probability of the electric actuator, and generate maintenance scheduling suggestions.

[0012] In a preferred embodiment, obtain the historical operation data of the electric actuator within a preset operation cycle, and generate a historical health data set of the electric actuator according to the historical operation data. Specifically:

[0013] Collect the historical operation data of the electric actuator within a preset operation cycle;

[0014] Based on the change range, change trend, and fluctuation amplitude of each parameter data in the historical operation data, perform data validity screening to remove parameter data that exceeds the preset reasonable range;

[0015] Integrate the historical operation data after data validity screening according to the parameter correspondence relationship, and generate a historical health data set of the electric actuator including motor operation parameters, actuator motion parameters, and environmental vibration parameters.

[0016] In a preferred embodiment, extract feature parameters from the historical health data set to obtain a historical feature parameter set of the electric actuator, and establish a trend change feature matrix according to the change trend of the historical feature parameters over time. Specifically:

[0017] Perform historical feature parameter extraction processing on the historical health data set;

[0018] Establish a historical feature parameter set according to different dimensions of the historical feature parameters;

[0019] Calculate the change trend values of different historical feature parameters in the historical feature parameter set within a specified time period;

[0020] Arrange the change trend values of the historical feature parameters in chronological order to form a trend change feature matrix; the rows of the trend change feature matrix represent different historical feature parameters, and the columns represent the corresponding time series data.

[0021] In a preferred embodiment, construct a multi-dimensional state space based on the trend change feature matrix, calculate the state aggregation degree of the historical failure events of the electric actuator in the multi-dimensional state space, and determine a feature warning index set according to the state aggregation degree. Specifically:

[0022] Determine the dimension of the multi-dimensional state space according to the number of historical feature parameters in the trend change feature matrix; the coordinate axes of the multi-dimensional state space are different historical feature parameters of the trend change feature matrix, and the feature position coordinates in the multi-dimensional state space are the change trend values of the historical feature parameters;

[0023] Calculate the aggregation degree of the state points when the electric actuator has historical failures based on the coordinate positions of each feature in the multi-dimensional state space, and obtain the state aggregation degree value of the historical failure events of the electric actuator in the multi-dimensional state space;

[0024] Set a warning threshold according to the state aggregation degree value, and select the historical feature parameters corresponding to the coordinate positions of the features whose state aggregation degree values exceed the warning threshold as the feature warning index set.

[0025] In a preferred embodiment, extract real-time feature parameters from the real-time operation data of the electric actuator, establish a real-time feature state vector, and calculate the spatial position correlation with the feature warning index set in the multi-dimensional state space to determine the real-time risk correlation factor. Specifically:

[0026] Collect the operation data of the electric actuator under the current operation state in real time;

[0027] Extract real-time feature parameters based on the real-time operation data, and form a real-time feature state vector according to the parameter order of the historical feature parameter set;

[0028] Map the real-time feature state vector to the multi-dimensional state space, calculate the spatial distance between the real-time feature state vector and the coordinate positions of the features in the feature warning index set, calculate the spatial position correlation according to the spatial distance, and determine the real-time risk correlation factor of the current operation state of the electric actuator according to the spatial position correlation.

[0029] In a preferred embodiment, generate a risk probability prediction model according to the real-time risk correlation factor, the historical feature parameter set, and the trend change feature matrix, predict the failure probability of the electric actuator, and generate a maintenance scheduling recommendation. Specifically:

[0030] Use the real-time risk correlation factor and the historical feature parameter set as the input data of the risk probability prediction model;

[0031] Use the risk probability prediction model to calculate the change trend values of the historical feature parameters in the trend change feature matrix, and predict the probability value of the electric actuator failing under the current operation state;

[0032] Generate a maintenance scheduling recommendation for the electric actuator according to the predicted failure probability value and the preset maintenance threshold.

[0033] On the other hand, the present invention provides an intelligent maintenance system for an electric actuator based on failure prediction, including a data acquisition module, a feature parameter extraction module, a state space construction module, a correlation calculation module, and a risk probability prediction module;

[0034] The data acquisition module obtains the historical operation data of the electric actuator within a preset operation cycle, and generates a historical health data set of the electric actuator according to the historical operation data;

[0035] The feature parameter extraction module extracts feature parameters from the historical health data set to obtain a historical feature parameter set of the electric actuator, and establishes a trend change feature matrix according to the changing trend of the historical feature parameters over time;

[0036] The state space construction module constructs a multi-dimensional state space based on the trend change feature matrix, calculates the state aggregation degree of the historical fault events of the electric actuator in the multi-dimensional state space, and determines a feature warning index set according to the state aggregation degree;

[0037] The correlation calculation module extracts real-time feature parameters according to the real-time operation data of the electric actuator, establishes a real-time feature state vector, and calculates the spatial position correlation with the feature warning index set in the multi-dimensional state space to determine the real-time risk correlation factor;

[0038] The risk probability prediction module generates a risk probability prediction model according to the real-time risk correlation factor, the historical feature parameter set and the trend change feature matrix, predicts the fault probability of the electric actuator, and generates a maintenance scheduling suggestion.

[0039] Technical effects and advantages of the intelligent maintenance system and method for electric actuators based on fault prediction of the present invention:

[0040] By analyzing the historical operation data and real-time operation data of the electric actuator, the key feature parameters affecting the fault of the electric actuator can be accurately identified, and a trend change feature matrix is constructed based on the parameter change trend, realizing the full-cycle monitoring of the operation state of the electric actuator. Using the multi-dimensional state space to quantitatively analyze the state aggregation degree of historical fault events can effectively screen out the feature warning index set that has a significant impact on the abnormality of the electric actuator, improving the accuracy of fault prediction. At the same time, the spatial position correlation calculation method is adopted in real-time monitoring, which can dynamically evaluate the matching degree between the current state of the electric actuator and the fault warning index, improving the real-time performance and accuracy of risk identification. Combining with the risk probability prediction model, it can predict potential fault risks in advance according to the changing trend of the current operation state of the electric actuator and generate corresponding maintenance scheduling suggestions, which can adapt to the maintenance needs of different types of electric actuators, significantly improve the reliability and safety of the operation of the electric actuator, reduce the risk of unplanned shutdown, and effectively optimize the allocation of maintenance resources. Description of the Drawings

[0041] Figure 1 It is a schematic diagram of the intelligent maintenance method for electric actuators based on fault prediction of the present invention;

[0042] Figure 2 This is a schematic structural diagram of the intelligent maintenance system for electric actuators based on fault prediction of the present invention. Specific embodiments

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

[0044] Embodiment 1:

[0045] Figure 1 The intelligent maintenance method for electric actuators based on fault prediction of the present invention is given, which includes the following steps:

[0046] Collect historical operation data of the electric actuator within a preset operation cycle, and generate a historical health data set of the electric actuator according to the historical operation data;

[0047] Extract characteristic parameters from the historical health data set to obtain a historical characteristic parameter set of the electric actuator, and establish a trend change characteristic matrix according to the change trend of the historical characteristic parameters over time;

[0048] Construct a multi-dimensional state space based on the trend change characteristic matrix, calculate the state aggregation degree of the historical fault events of the electric actuator in the multi-dimensional state space, and determine a characteristic warning index set according to the state aggregation degree;

[0049] Extract real-time characteristic parameters according to the real-time operation data of the electric actuator, establish a real-time characteristic state vector, and perform spatial position correlation calculation with the characteristic warning index set in the multi-dimensional state space to determine a real-time risk correlation factor;

[0050] Generate a risk probability prediction model according to the real-time risk correlation factor, the historical characteristic parameter set and the trend change characteristic matrix, predict the fault probability of the electric actuator, and generate a maintenance scheduling suggestion.

[0051] Specifically, collecting historical operation data of the electric actuator within a preset operation cycle, and generating a historical health data set of the electric actuator according to the historical operation data includes:

[0052] Collect historical operation data of the electric actuator within a preset operation cycle;

[0053] Specifically, within a fixed time period (such as daily, weekly or monthly), collect data of various parameters during the operation of the electric actuator. For example, the collected data includes:

[0054] Motor operating parameters: motor current, motor voltage, motor temperature, etc.;

[0055] Motion parameters of the electric actuator: displacement, speed of the actuator, etc.;

[0056] Environmental vibration parameters: vibration signal, vibration frequency, etc.

[0057] Exemplarily, the preset operating cycle set for the electric actuator is 24 hours. Within 24 hours, data is recorded every 1 minute, so 1440 data points will be obtained in a day, and each data point records the values of the above parameters.

[0058] Based on the change range, change trend, and fluctuation amplitude of the parameter data in the historical operating data, perform data validity screening to remove parameter data that exceeds the preset reasonable range;

[0059] Specifically, the collected data may have abnormal values due to sensor failures or external interference. It is necessary to set reasonable value ranges, allowable change trends, and fluctuation amplitudes to screen the data.

[0060] For example, under normal conditions, the motor current should be between 5 A and 15 A; if a certain record is 20 A or 3 A, it is judged as abnormal data and excluded.

[0061] Classify and integrate the historical operating data after data validity screening according to the parameter correspondence relationship to generate a historical health data set of the electric actuator including motor operating parameters, actuator motion parameters, and environmental vibration parameters;

[0062] Specifically, classify and integrate the historical operating data after data validity screening according to the parameter attributes to form a structured data set.

[0063] Exemplarily, classify the motor parameters (such as current, voltage, temperature) into one category;

[0064] Classify the motion parameters of the electric actuator (such as displacement, speed) into another category;

[0065] Classify the environmental vibration parameters into a third category.

[0066] Specifically, extract characteristic parameters from the historical health data set to obtain a historical characteristic parameter set of the electric actuator. Establish a trend change characteristic matrix according to the change trend of the historical characteristic parameters over time, including:

[0067] Perform historical characteristic parameter extraction processing on the historical health data set;

[0068] Specifically, key indicators that can reflect the state changes of the electric actuator are extracted from the historical health dataset, such as the average value, peak value, standard deviation, change rate, etc.

[0069] Exemplarily, the average current value and volatility (standard deviation) are extracted from the motor current data; the temperature rise rate is extracted from the temperature data; the maximum displacement change amount is extracted from the displacement data, etc.

[0070] A historical feature parameter set is established according to different dimensions of the historical feature parameters;

[0071] Specifically, the extracted feature parameters are sorted respectively according to three dimensions: motor parameters, actuator motion parameters, and environmental parameters to form a historical feature parameter set. For example:

[0072] Motor parameters: average current, current variance, average voltage, voltage volatility, average motor temperature, motor temperature rise rate;

[0073] Electric actuator motion parameters: average valve displacement, average valve speed;

[0074] Environmental vibration parameters: vibration frequency, average vibration frequency.

[0075] The change trend values of different historical feature parameters in the historical feature parameter set are calculated respectively within a specified time period;

[0076] Specifically, the degree of change of each feature parameter within a certain time interval is calculated to quantify its change trend.

[0077] For example, calculate the trend change of the average current:

[0078] Current change trend = (average current of the next day - average current of the previous day) / average current of the previous day

[0079] Similarly, the same calculation method is performed on other historical feature parameters to obtain their respective change trend values.

[0080] The change trend values of the historical feature parameters are arranged in chronological order to form a trend change feature matrix; the rows of the trend change feature matrix represent different historical feature parameters, and the columns represent the corresponding time series data.

[0081] Specifically, a multi-dimensional state space is constructed based on the trend change feature matrix, the state aggregation degree of the historical fault events of the electric actuator in the multi-dimensional state space is calculated, and a feature warning index set is determined according to the state aggregation degree, including:

[0082] Determine the dimension of the multi-dimensional state space according to the number of historical feature parameters in the trend change feature matrix; the coordinate axes of the multi-dimensional state space are different historical feature parameters of the trend change feature matrix, and the characteristic position coordinates in the multi-dimensional state space are the change trend values of the historical feature parameters.

[0083] Specifically, regard each row in the trend change feature matrix as an independent dimension to form a multi-dimensional space. For example, if three historical feature parameters, namely "change trend of average motor current", "change trend of voltage volatility", and "change trend of motor temperature gradient", are extracted from the trend change feature matrix, then the dimension of the multi-dimensional state space is 3. The three coordinate axes of this state space respectively represent the above three parameters.

[0084] For a certain historical fault event, by calculating the change trends of each parameter during this event (for example, the change trend of motor current is 2.5%, the change trend of voltage volatility is -0.4%, and the change trend of motor temperature gradient is 1.2%), the corresponding space coordinates can be obtained as (2.5, -0.4, 1.2).

[0085] Based on the characteristic position coordinates in the multi-dimensional state space, calculate the aggregation degree of the state points when the electric actuator has historical faults, and obtain the state aggregation degree value of the historical fault events of the electric actuator in the multi-dimensional state space.

[0086] Specifically, after mapping the historical fault events to the multi-dimensional state space, each fault event has a coordinate point in this space. To calculate the aggregation degree of these fault event points, methods such as local density or distance statistics can be used.

[0087] For example, set a radius (the unit is the same as that of each coordinate axis), and count the number of other fault event points within the radius , and at the same time calculate the spatial volume corresponding to this radius (for example, in a three-dimensional space ). Then the calculation formula for the state aggregation degree can be defined as: ; where represents the state aggregation degree, which is used to reflect the distribution concentration of historical fault points in the multi-dimensional space.

[0088] Set an early warning threshold based on the state aggregation degree value, and select the historical feature parameters corresponding to the characteristic position coordinates whose state aggregation degree values exceed the early warning threshold as the characteristic early warning index set.

[0089] Specifically, according to the calculated state aggregation degrees of each fault event, determine an early warning threshold. For example, through statistical analysis, it is obtained that when the state aggregation degree is greater than 8, the fault risk increases significantly.

[0090] At this time, for all fault events with an aggregation degree exceeding 8, the respective historical characteristic parameters (such as current change trend, temperature gradient, etc.) corresponding to their coordinates in the multi-dimensional space are considered to have warning significance.

[0091] For example, if the coordinates of a certain fault event in space are (2.5, -0.4, 1.2) and the calculated result of its aggregation degree is 9.62, then the parameters represented by these three values can be used as warning indicators and constitute a characteristic warning indicator set.

[0092] Specifically, extract real-time characteristic parameters according to the real-time operation data of the electric actuator, establish a real-time characteristic state vector, and perform spatial position correlation calculation with the characteristic warning indicator set in the multi-dimensional state space to determine the real-time risk correlation factor, including:

[0093] Collect the operation data of the electric actuator under the current operation state in real time;

[0094] Specifically, during the normal operation of the electric actuator, collect the operation data at the current moment in real time. These data are consistent with the parameters during historical data collection, such as motor current, motor temperature, actuator displacement, speed, and vibration signal, etc.

[0095] Exemplarily, during the operation of the electric actuator, collect data once every minute, which is consistent with the historical data collection period to ensure the comparability of data comparison.

[0096] Extract real-time characteristic parameters based on the real-time operation data, and form a real-time characteristic state vector according to the parameter order of the historical characteristic parameter set;

[0097] Specifically, adopt the same processing method as the historical operation data processing to extract characteristic parameters from the real-time collected data. If the order of the historical characteristic parameter set is [motor current change trend, voltage volatility, temperature gradient, displacement change rate, speed change rate, vibration frequency change], then the real-time characteristic state vector is also arranged in this order to obtain the real-time characteristic state vector.

[0098] Map the real-time characteristic state vector to the multi-dimensional state space, calculate the spatial distance between the real-time characteristic state vector and the characteristic position coordinates in the characteristic warning indicator set, calculate the spatial position correlation according to the spatial distance, and determine the real-time risk correlation factor of the current operation state of the electric actuator based on the spatial position correlation;

[0099] Specifically, map the real-time characteristic state vector into the multi-dimensional state space to obtain the coordinates of the current state in this space. Then, compare this coordinate with the coordinates of each warning point in the characteristic warning indicator set, and use the Euclidean distance formula to calculate the spatial distance. Determine the spatial position correlation based on the inverse relationship of the distance.

[0100] Take this spatial position correlation as a real-time risk correlation factor to reflect the matching degree between the current operating state and the historical fault warning characteristics. The larger the value of the spatial position correlation, the closer the current state is to the warning index, and the higher the fault risk.

[0101] Specifically, according to the real-time risk correlation factor, the historical feature parameter set, and the trend change feature matrix, generate a risk probability prediction model to predict the fault probability of the electric actuator, and generate maintenance scheduling suggestions, including:

[0102] Take the real-time risk correlation factor and the historical feature parameter set as the input data of the risk probability prediction model;

[0103] Specifically, when constructing the risk probability prediction model, first select each parameter in the real-time risk correlation factor and the historical feature parameter set (such as the motor current change trend, temperature gradient, etc.) as the input of the risk probability prediction model.

[0104] Use the risk probability prediction model to calculate the change trend values of the historical feature parameters in the trend change feature matrix, and predict the probability value of the electric actuator failing under the current operating state;

[0105] Specifically, the risk probability prediction model can adopt methods such as weighted average or multiple regression to fuse and calculate the real-time risk correlation factor and the historical trend data. The output of the risk probability prediction model is a fault probability value between 0 and 1, reflecting the probability of the electric actuator failing under the current operating state.

[0106] Generate maintenance scheduling suggestions for the electric actuator according to the predicted fault probability value and the preset maintenance threshold;

[0107] Specifically, compare the fault probability value with the preset maintenance threshold:

[0108] When the fault probability value is greater than or equal to the preset maintenance threshold, it is recommended to immediately start maintenance measures, such as arranging maintenance in advance, adjusting working parameters, or replacing components;

[0109] When the fault probability value is less than the preset maintenance threshold, it is recommended to continue monitoring or arrange regular maintenance.

[0110] Embodiment 2:

[0111] The difference between Embodiment 2 and Embodiment 1 of the present invention is that this embodiment introduces an intelligent maintenance system for electric actuators based on fault prediction.

[0112] Figure 2The structural schematic diagram of the intelligent maintenance system for an electric actuator based on fault prediction according to the present invention is given. The intelligent maintenance system for an electric actuator based on fault prediction includes a data acquisition module, a feature parameter extraction module, a state space construction module, a correlation calculation module, and a risk probability prediction module;

[0113] The data acquisition module obtains the historical operation data of the electric actuator within a preset operation cycle, and generates a historical health data set of the electric actuator according to the historical operation data;

[0114] The feature parameter extraction module extracts feature parameters from the historical health data set to obtain a historical feature parameter set of the electric actuator, and establishes a trend change feature matrix according to the changing trend of the historical feature parameters over time;

[0115] The state space construction module constructs a multi-dimensional state space based on the trend change feature matrix, calculates the state aggregation degree of the historical fault events of the electric actuator in the multi-dimensional state space, and determines a feature warning index set according to the state aggregation degree;

[0116] The correlation calculation module extracts real-time feature parameters according to the real-time operation data of the electric actuator, establishes a real-time feature state vector, and calculates the spatial position correlation with the feature warning index set in the multi-dimensional state space to determine a real-time risk correlation factor;

[0117] The risk probability prediction module generates a risk probability prediction model according to the real-time risk correlation factor, the historical feature parameter set, and the trend change feature matrix, predicts the fault probability of the electric actuator, and generates a maintenance scheduling suggestion.

[0118] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula closest to the real situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0119] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0120] Those of ordinary skill in the art will realize that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0121] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here.

[0122] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces, and the indirect coupling or communication connection of the devices or modules can be in an electrical, mechanical, or other form.

[0123] The module described as a separation component may or may not be physically separated. The component shown as a module may or may not be a physical module, and it may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0124] In addition, in each embodiment of this application, each functional module can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0125] If the above function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0126] As described above, this is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claimed rights.

[0127] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent maintenance method for an electric actuator based on fault prediction, characterized in that, It includes the following steps: Obtain the historical operation data of the electric actuator within a preset operation cycle, and generate a historical health data set of the electric actuator according to the historical operation data; Extract characteristic parameters from the historical health data set to obtain a historical characteristic parameter set of the electric actuator, and establish a trend change characteristic matrix according to the change trend of the historical characteristic parameters over time; Construct a multi-dimensional state space based on the trend change characteristic matrix, calculate the state aggregation degree of the historical fault events of the electric actuator in the multi-dimensional state space, and determine a characteristic early warning index set according to the state aggregation degree; Determine the dimension of the multi-dimensional state space according to the number of historical characteristic parameters in the trend change characteristic matrix; The coordinate axes of the multi-dimensional state space are different historical characteristic parameters of the trend change characteristic matrix, and the characteristic position coordinates in the multi-dimensional state space are the change trend values of the historical characteristic parameters; Calculate the aggregation degree of the state points when the electric actuator has historical faults based on the characteristic position coordinates in the multi-dimensional state space, and obtain the state aggregation degree value of the historical fault events of the electric actuator in the multi-dimensional state space; Set an early warning threshold according to the state aggregation degree value, and select the historical characteristic parameters corresponding to the characteristic position coordinates whose state aggregation degree values exceed the early warning threshold as the characteristic early warning index set; Extract real-time characteristic parameters according to the real-time operation data of the electric actuator, establish a real-time characteristic state vector, and perform spatial position correlation calculation with the characteristic early warning index set in the multi-dimensional state space to determine the real-time risk correlation factor; Generate a risk probability prediction model according to the real-time risk correlation factor, the historical characteristic parameter set and the trend change characteristic matrix, predict the fault probability of the electric actuator, and generate a maintenance scheduling suggestion; Use the real-time risk correlation factor and the historical characteristic parameter set as the input data of the risk probability prediction model; Use the risk probability prediction model to calculate the change trend values of the historical characteristic parameters in the trend change characteristic matrix, and predict the probability value of the electric actuator failing under the current operating state; Generate a maintenance scheduling suggestion for the electric actuator according to the predicted fault probability value and the preset maintenance threshold.

2. The intelligent maintenance method for an electric actuator based on fault prediction according to claim 1, wherein Obtain the historical operation data of the electric actuator within a preset operation cycle, and generate a historical health data set of the electric actuator according to the historical operation data. Specifically: Collect the historical operation data of the electric actuator within a preset operation cycle; According to the change range, change trend and fluctuation range of each parameter data in the historical operation data, perform data validity screening, and remove the parameter data that exceeds the preset reasonable range; Classify and integrate the historical operation data after data validity screening according to the parameter correspondence relationship, and generate a historical health data set of the electric actuator including motor operation parameters, actuator movement parameters and environmental vibration parameters.

3. The intelligent maintenance method for an electric actuator based on fault prediction according to claim 2, wherein Extract characteristic parameters from the historical health data set to obtain a historical characteristic parameter set of the electric actuator, and establish a trend change characteristic matrix according to the change trend of the historical characteristic parameters over time. Specifically: Perform historical characteristic parameter extraction processing on the historical health data set; Establish a historical characteristic parameter set according to different dimensions of the historical characteristic parameters; Calculate the change trend values of different historical feature parameters in the historical feature parameter set within a specified time period respectively; Arrange the change trend values of the historical feature parameters in chronological order to form a trend change feature matrix; the rows of the trend change feature matrix represent different historical feature parameters, and the columns represent the corresponding time series data.

4. The intelligent maintenance method for an electric actuator based on fault prediction according to claim 3, wherein Extract real-time feature parameters from the real-time operation data of the electric actuator, establish a real-time feature state vector, and calculate the spatial position correlation with the feature warning index set in the multi-dimensional state space to determine the real-time risk correlation factor. Specifically: Collect the operation data of the electric actuator under the current operation state in real time; Extract real-time feature parameters based on the real-time operation data, and form a real-time feature state vector according to the parameter order of the historical feature parameter set; Map the real-time feature state vector to the multi-dimensional state space, calculate the spatial distance between the real-time feature state vector and the feature position coordinates in the feature warning index set, calculate the spatial position correlation according to the spatial distance, and determine the real-time risk correlation factor of the current operation state of the electric actuator based on the spatial position correlation.

5. An intelligent maintenance system for an electric actuator based on fault prediction, which is used to implement the intelligent maintenance method for an electric actuator based on fault prediction according to any one of claims 1-4, characterized in that It includes a data acquisition module, a feature parameter extraction module, a state space construction module, a correlation calculation module, and a risk probability prediction module; The data acquisition module obtains the historical operation data of the electric actuator within a preset operation cycle, and generates a historical health data set of the electric actuator according to the historical operation data; The feature parameter extraction module extracts feature parameters from the historical health data set to obtain a historical feature parameter set of the electric actuator, and establishes a trend change feature matrix according to the change trend of the historical feature parameters over time; The state space construction module constructs a multi-dimensional state space based on the trend change feature matrix, calculates the state aggregation degree of the historical fault events of the electric actuator in the multi-dimensional state space, and determines the feature warning index set according to the state aggregation degree; The correlation calculation module extracts real-time feature parameters from the real-time operation data of the electric actuator, establishes a real-time feature state vector, and calculates the spatial position correlation with the feature warning index set in the multi-dimensional state space to determine the real-time risk correlation factor; The risk probability prediction module generates a risk probability prediction model according to the real-time risk correlation factor, the historical feature parameter set and the trend change feature matrix, predicts the fault probability of the electric actuator, and generates a maintenance scheduling suggestion.

Images