Multi-parameter fusion non-all-plastic hook type commutator fault diagnosis method and system

Abstract

The invention discloses a multi-parameter fusion non-all-plastic hook type commutator fault diagnosis method and system, and relates to the technical field of data processing, and the method comprises the steps: obtaining a design drawing of a non-all-plastic hook type commutator, and recognizing M key regions; a sensor is deployed, a monitoring network is established, and multi-source monitoring data streams of the M key areas are collected. And performing multi-parameter fusion on the data stream in combination with an application scene to form a fused monitoring data stream. And performing modal decomposition and fault identification on the fused data stream, extracting a fault feature set, and constructing a fault diagnosis rule space. And analyzing the fault feature set based on rule space matching, and generating a commutator fault diagnosis result. Therefore, the technical effects of fusing multiple parameters, improving the information utilization efficiency, and improving the accuracy and response speed of fault recognition are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a fault diagnosis method and system for a non-all-plastic hook-type commutator with multi-parameter fusion. Background Art

[0002] As a key component in motors and power generation equipment, the working state of a non-all-plastic hook-type commutator directly affects the overall performance and operation stability of the equipment. The design of the commutator not only focuses on structural optimization and performance improvement, but also requires efficient on-line monitoring and fault diagnosis capabilities to detect potential hidden dangers in the initial stage of equipment operation. At present, the fault diagnosis of the commutator mainly relies on local monitoring technologies and single-signal feature analysis methods. Traditional solutions usually only focus on monitoring and analyzing a certain type of parameters (such as temperature, vibration, sound, etc.), easily ignoring the interconnection characteristics between key parts and the complexity of the multi-factor interaction, and failing to make full use of the global multi-parameter information during the operation of the commutator, thus having deficiencies in the identification of fault modes and early warning. In addition, the existing fault diagnosis methods are also relatively single in extracting and matching diagnosis rules in complex application scenarios, and it is difficult to meet the detection requirements in the case of intertwined multi-working conditions and multi-fault modes. Summary of the Invention

[0003] The present invention provides a fault diagnosis method and system for a non-all-plastic hook-type commutator with multi-parameter fusion, so as to solve the technical problems of single information modality and poor detection efficiency in the case of intertwined multi-working conditions and multi-fault modes in the prior art, and achieve the technical effects of fusing multi-parameters, improving the information utilization efficiency, and improving the accuracy and response speed of fault identification.

[0004] In a first aspect, the present invention provides a fault diagnosis method for a non-all-plastic hook-type commutator with multi-parameter fusion, wherein the fault diagnosis method for a non-all-plastic hook-type commutator with multi-parameter fusion includes:

[0005] Obtain the design drawing information of the non-all-plastic hook-type commutator, and identify the key parts of the non-all-plastic hook-type commutator based on the design drawing information to obtain M pieces of commutator key area information.

[0006] Deploy sensor analysis respectively on the M pieces of commutator key area information, build a commutator monitoring sensor network, and collect and obtain M pieces of multi-source monitoring data streams of key areas through the commutator monitoring sensor network.

[0007] Perform multi-parameter fusion on the M pieces of multi-source monitoring data streams of key areas respectively according to the application scenario information of the non-all-plastic hook-type commutator to obtain M pieces of fusion monitoring data streams of key areas.

[0008] Modal decomposition and fault identification are performed on the M key area fusion monitoring data streams to obtain M key area fault feature sets, and a commutator fault diagnosis rule space is constructed.

[0009] Based on the commutator fault diagnosis rule space, an integrated matching analysis is performed on the M key area fault feature sets to generate a commutator fault diagnosis result.

[0010] In a feasible implementation, the obtaining of M commutator key area information includes:

[0011] The non-all-plastic hook-type commutator is disassembled into structural components based on the design drawing information to obtain a commutator structural component set.

[0012] Each structural component in the commutator structural component set is marked with an application function to determine a commutator component application function set.

[0013] Adjacent components of the commutator structural component set are integrated according to the commutator component application function set to obtain N commutator functional area sets.

[0014] The N commutator functional area sets are evaluated and screened to obtain M commutator key area information, where N≥M.

[0015] In a feasible implementation, the obtaining of M commutator key area information includes:

[0016] Historical fault data is crawled based on the N commutator functional area sets to obtain a non-all-plastic hook-type commutator fault database.

[0017] The non-all-plastic hook-type commutator fault database is mapped and diverted according to the N commutator functional area sets to obtain N functional area fault data sets.

[0018] Frequency statistics and probability calculation are performed based on the N functional area fault data sets to obtain the probability of fault occurrence in the N functional areas.

[0019] According to the probability of occurrence of faults in the N functional areas, a fault probability threshold is preset, and based on the fault probability threshold, the N commutator functional area sets are compared and screened to obtain the M commutator key area information.

[0020] In a feasible implementation, the step of building a commutator monitoring sensor network includes:

[0021] The monitoring demand analysis is performed on the M commutator key area information in turn to determine the M key area monitoring types and the M key area monitoring value ranges.

[0022] Select M key area monitoring sensor models according to the M key area monitoring types and the M key area monitoring value ranges.

[0023] Perform monitoring coverage analysis on the M commutator key area information based on the M key area monitoring sensor models to obtain M key area sensor deployment parameters.

[0024] Based on the M key area monitoring sensor models and the M key area sensor deployment parameters, deploy sensors on the M commutator key area information to build the commutator monitoring sensor network.

[0025] In a feasible implementation manner, the obtaining of the M key area fusion monitoring data streams includes:

[0026] Evaluate the influence degree of each area data monitoring source in the M key area multi-source monitoring data streams according to the application scenario information of the non-all-plastic hook-type commutator to obtain an M-area data monitoring source influence factor set.

[0027] Determine an M-area data monitoring source credibility coefficient set according to the M-area data monitoring source influence factor set.

[0028] Integrate the M key area multi-source monitoring data streams according to the timing information to obtain M key area multi-source timing data streams.

[0029] Perform multi-parameter fusion on the M key area multi-source timing data streams based on the M area data monitoring source credibility coefficient set to obtain the M key area fusion monitoring data streams.

[0030] In a feasible implementation manner, the obtaining of the M key area fault feature sets includes:

[0031] Perform empirical mode decomposition on the M key area fusion monitoring data streams respectively to obtain M key area IMF components.

[0032] Determine M key area normal working thresholds according to the performance working standards of the M commutator key area information.

[0033] Extract fault features from the M key area IMF components according to the M key area normal working thresholds to obtain the M key area fault feature sets.

[0034] In a feasible implementation manner, the constructing of the commutator fault diagnosis rule space includes:

[0035] Partition the non-all-plastic hook-type commutator fault database according to the M commutator key area information to obtain M key area fault data spaces.

[0036] Analyze and extract the characteristics of the M key area fault data spaces respectively to determine M key area fault feature spaces.

[0037] Mine fault diagnosis rules in the M key area fault feature spaces to construct a commutator fault diagnosis rule space.

[0038] In a feasible implementation manner, the construction of the commutator fault diagnosis rule space includes:

[0039] According to the M key area fault feature spaces, obtain M key area fault feature combinations and corresponding commutator diagnosis information.

[0040] Conduct fault correlation topology analysis on the M key area fault feature combinations to obtain a commutator fault cascade feature set.

[0041] Logically map and associate the commutator fault cascade feature set and the commutator diagnosis information to construct the commutator fault diagnosis rule space.

[0042] In a feasible implementation manner, the generation of the commutator fault diagnosis result includes:

[0043] Based on the commutator fault diagnosis rule space, perform matching analysis on the M key area fault feature sets to activate the matching commutator diagnosis rule sets.

[0044] Integrate and output the matching commutator diagnosis rule sets to generate the commutator fault diagnosis result.

[0045] In a second aspect, the present invention also provides a multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis system, wherein the multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis system includes:

[0046] An area recognition module, configured to obtain the design drawing information of the non-all-plastic hook-type commutator, and identify the key parts of the non-all-plastic hook-type commutator based on the design drawing information to obtain M commutator key area information.

[0047] An area monitoring module, configured to respectively deploy sensor analysis on the M commutator key area information, build a commutator monitoring sensor network, and collect and obtain M key area multi-source monitoring data streams through the commutator monitoring sensor network.

[0048] A data stream fusion module, configured to perform multi-parameter fusion on the multi-source monitoring data streams of the M key regions respectively according to the application scenario information of the non-all-plastic hook-type commutator, so as to obtain the fused monitoring data streams of the M key regions.

[0049] A fault feature acquisition module, configured to perform modal decomposition and fault identification on the fused monitoring data streams of the M key regions, so as to obtain the fault feature sets of the M key regions, and simultaneously construct a commutator fault diagnosis rule space.

[0050] A fault diagnosis execution module, configured to perform integrated matching analysis on the fault feature sets of the M key regions based on the commutator fault diagnosis rule space, and generate a commutator fault diagnosis result.

[0051] The present invention discloses a fault diagnosis method and system for a non-all-plastic hook-type commutator with multi-parameter fusion, including: obtaining the design drawing information of the non-all-plastic hook-type commutator, identifying the key parts of the non-all-plastic hook-type commutator based on the design drawing information, so as to obtain the information of M key regions of the commutator; respectively performing sensor analysis and deployment on the information of the M key regions of the commutator, building a commutator monitoring sensor network, and collecting the multi-source monitoring data streams of the M key regions through the sensor network; respectively performing multi-parameter fusion on the multi-source monitoring data streams of the M key regions according to the application scenario information of the non-all-plastic hook-type commutator, so as to obtain the fused monitoring data streams of the M key regions; performing modal decomposition and fault identification on the fused monitoring data streams of the M key regions, so as to obtain the fault feature sets of the M key regions, and constructing a commutator fault diagnosis rule space; based on the commutator fault diagnosis rule space, performing integrated matching analysis on the fault feature sets of the M key regions, and generating a commutator fault diagnosis result. The fault diagnosis method and system for a non-all-plastic hook-type commutator with multi-parameter fusion disclosed by the present invention solve the technical problem of poor detection efficiency in the case of single information modality and the interweaving of multiple working conditions and multiple fault modes, and achieve the technical effects of fusing multiple parameters, improving the information utilization efficiency, and improving the accuracy and response speed of fault identification. Description of the Drawings

[0052] Figure 1 It is a schematic flow chart of the fault diagnosis method for a non-all-plastic hook-type commutator with multi-parameter fusion of the present invention;

[0053] Figure 2 It is a schematic structural diagram of the fault diagnosis system for a non-all-plastic hook-type commutator with multi-parameter fusion of the present invention.

[0054] Description of the reference numerals: Region identification module 11, Region monitoring module 12, Data stream fusion module 13, Fault feature acquisition module 14, Fault diagnosis execution module 15. Detailed Embodiments

[0055] The above technical solution will be described in detail below in combination with the specification drawings and specific implementation manners to better understand the above technical solution. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments only used to explain the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts belong to the scope of protection of the present invention. In addition, it should be noted that, for the sake of convenience of description, only the parts related to the present invention rather than all are shown in the drawings.

[0056] Embodiment 1, as Figure 1 is a schematic flowchart of a multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method of the present invention. Among them, the multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method includes:

[0057] S100: Obtain the design drawing information of the non-all-plastic hook-type commutator, and identify the key parts of the non-all-plastic hook-type commutator based on the design drawing information to obtain M pieces of commutator key area information.

[0058] Specifically, first, interact with the production management end to obtain the design materials of the target non-all-plastic hook-type commutator, and extract the design drawing information therefrom. The design drawing information includes the detailed structure of the commutator and the position information of each component. By analyzing the design drawings, the key parts of the commutator can be accurately identified, such as the areas that are prone to failure or have a greater impact on the performance of the commutator. The accurate identification of the key parts provides a basis for the subsequent deployment of sensors and helps to ensure the quality of the monitoring data.

[0059] In some embodiments, the obtaining of M pieces of commutator key area information includes:

[0060] Based on the design drawing information, disassemble the structural components of the non-all-plastic hook-type commutator to obtain a set of commutator structural components; respectively mark the application functions of each structural component in the set of commutator structural components to determine a set of commutator component application functions; integrate adjacent components of the set of commutator structural components according to the set of commutator component application functions to obtain N sets of commutator functional area sets; evaluate and screen the N sets of commutator functional area sets to obtain M pieces of commutator key area information, where N≥M.

[0061] Specifically, first, the structural components of the target non-all-plastic hook-type commutator are disassembled according to the design drawing information to form a set of commutator structural components containing all components. Among them, the design drawing information includes the component markings of the target non-all-plastic hook-type commutator, and the commutator is theoretically decomposed into individual independent structural components, such as commutator segments, hooks, insulating materials, etc.; then, each component is marked with its specific function during the actual operation of the commutator. For example, the main function of the commutator segment is to achieve current commutation, and the function of the hook is to fix the commutator, etc. Through this process, the role of each component in the overall function of the commutator can be clarified.

[0062] Furthermore, according to the set of application functions of the commutator components, the set of commutator structural components is integrated with adjacent components to obtain N sets of commutator functional areas. In other words, adjacent components with similar or related functions are integrated together to form larger sets of functional areas. For example, the commutator segment is integrated with its adjacent insulating sheet and hook into a functional area because they cooperate with each other during the commutation process to jointly complete specific functions.

[0063] Finally, the N sets of commutator functional areas are evaluated and screened to obtain M pieces of key area information of the commutator, where N≥M. Specifically, the evaluation and screening refer to selecting M of the most critical functional areas from the N sets of functional areas as the key areas for subsequent monitoring according to certain criteria, such as the functionality of the components during the operation of the commutator and the probability of failure occurrence. Preferably, components that are adjacent in position and have the same application function are integrated into the same functional area.

[0064] Through the above steps, the complex structure of the commutator can be decomposed into several key areas, providing clear targets for subsequent sensor deployment and fault monitoring, helping to improve the pertinence and accuracy of fault diagnosis, and at the same time optimizing the configuration of monitoring resources, avoiding unnecessary monitoring point settings, thereby reducing the complexity and cost of the system.

[0065] In some implementation manners, obtaining the M pieces of key area information of the commutator includes:

[0066] Crawling historical fault data based on the N sets of commutator functional areas to obtain a fault database of the non-all-plastic hook-type commutator; mapping and diverting the fault database of the non-all-plastic hook-type commutator according to the N sets of commutator functional areas to obtain N sets of functional area fault data; performing frequency statistics and probability calculation based on the N sets of functional area fault data to obtain the fault occurrence probabilities of the N sets of functional areas; presetting a fault probability threshold according to the fault occurrence probabilities of the N sets of functional areas, and comparing and screening the N sets of commutator functional areas based on the fault probability threshold to obtain the M pieces of key area information of the commutator.

[0067] Specifically, first, it refers to collecting fault data related to the commutator functional area from various data sources (such as equipment maintenance records, fault reports, online databases, etc.), crawling historical fault data, and the non-all-plastic hook-type commutator fault database obtained includes information such as past fault types, occurrence times, and occurrence locations.

[0068] Table 1 Exemplary historical fault data

[0069]

[0070]

[0071] Then, classify the fault data according to the commutator functional area, so that each functional area corresponds to an independent fault data set, forming a functional area fault data set; through this classification, the fault data can be corresponding to the specific functional area, facilitating subsequent targeted analysis.

[0072] Exemplarily, the functional area fault data set includes:

[0073] Contact fault data set: F001, F003, F005, F006.

[0074] Spring fault data set: F002.

[0075] Rotating shaft fault data set: F004.

[0076] Furthermore, based on the N functional area fault data sets, frequency statistics and probability calculations are performed to obtain the N functional area fault occurrence probabilities. This step clarifies the likelihood of faults occurring in each functional area through quantitative analysis.

[0077] Table 2 Exemplary fault occurrence probabilities of each functional area

[0078] Functional area Number of faults Probability of fault occurrence Contact point 4 66.67％ Spring 1 16.67％ Rotating shaft 1 16.67％

[0079] Next, according to the N functional area fault occurrence probabilities, a fault probability threshold is preset, and based on this fault probability threshold, the N commutator functional area sets are compared and screened to obtain M commutator key area information. For example, if the area with a fault occurrence probability greater than 30% is set as the key area, then: contact (66.67%) > 30%, the contact is the key area; spring (16.67%) < 30%, the spring is the non-key area; rotating shaft (16.67%) < 30%, the rotating shaft is the non-key area.

[0080] Through the analysis of historical data, the functional area set is further refined into key areas, providing a more targeted target area for subsequent monitoring and fault diagnosis. Among them, the data-driven screening method can better reflect the fault characteristics of the commutator during actual operation, improving the accuracy and efficiency of fault diagnosis.

[0081] S200: Deploy sensor parsing respectively on the M pieces of commutator key area information, build a commutator monitoring sensor network, and collect and obtain multi-source monitoring data streams of M key areas through the commutator monitoring sensor network.

[0082] Specifically, based on the obtained M pieces of commutator key area information, locate the M commutator key areas, and respectively arrange corresponding monitoring sensors to form a commutator monitoring sensor network; among them, sensor parsing refers to the process of determining the index categories that need to be monitored for each according to the M pieces of commutator key area information, and then selecting appropriate sensor types and models, and determining the specific installation positions, quantities and layout methods of the sensors, ensuring that the sensors can effectively monitor the operating status of the key areas while avoiding mutual interference between sensors.

[0083] Exemplarily, for the commutator segment group area, parameters such as temperature, current and vibration may need to be monitored; for the hook area, parameters such as stress and displacement may need to be monitored. Furthermore, select appropriate sensor models according to the monitoring requirements.

[0084] Specifically, the commutator monitoring sensor network is a network system composed of multiple sensors. These sensors are distributed in the key areas of the commutator and are used to collect various physical quantity data during the operation of the commutator in real time, such as temperature, vibration, current, voltage, etc. The multi-source monitoring data stream is a collection of monitoring data strings from different types of sensors and different monitoring areas. Generally, these data have different physical dimensions, sampling frequencies and data formats, and need to be integrated and processed before being used for subsequent fault diagnosis.

[0085] By deploying a variety of sensors specifically in the key areas, various physical quantity changes during the operation of the commutator can be comprehensively monitored, more accurately reflecting the actual operating status of the commutator, and providing reliable data support for subsequent steps such as multi-parameter fusion, modal decomposition and fault identification.

[0086] In some embodiments, the building of the commutator monitoring sensor network includes:

[0087] Perform a monitoring requirement analysis on the information of the M key regions of the commutator in sequence to determine the monitoring types of the M key regions and the monitoring value ranges of the M key regions; select the model numbers of the M key region monitoring sensors according to the monitoring types of the M key regions and the monitoring value ranges of the M key regions; perform a monitoring coverage analysis on the information of the M key regions of the commutator based on the model numbers of the M key region monitoring sensors to obtain the sensor deployment parameters of the M key regions; deploy sensors on the information of the M key regions of the commutator based on the model numbers of the M key region monitoring sensors and the sensor deployment parameters of the M key regions, and build the commutator monitoring sensor network.

[0088] Specifically, first, perform a monitoring requirement analysis on the M key regions of the commutator to determine: the monitoring types (such as temperature, vibration, current, etc.) and the monitoring value ranges (the normal / abnormal thresholds of each physical quantity).

[0089] Table 3 Exemplary Monitoring Requirements

[0090]

[0091]

[0092] Then, according to the determined monitoring types and monitoring value ranges above, traverse the sensor selection library or the corresponding equipment selection manual to select the matching sensor model numbers. For example, for temperature monitoring, a thermocouple or an infrared temperature sensor can be selected; for vibration monitoring, an acceleration sensor or a velocity sensor can be selected.

[0093] Table 4 Exemplary Sensor Selection Results

[0094] Monitoring type Sensor model Measurement range Accuracy Temperature sensor Type K thermocouple 0℃-400℃ ±1℃ Current sensor Hall current sensor 0A - 100A ±0.5A Vibration sensor MEMS accelerometer 0g - 10g ±0.1g Torque sensor Torque measurement module 0 Nm - 50 Nm ±0.2 Nm

[0095] Next, perform a monitoring coverage analysis to determine the deployment positions and quantities of the sensors. The purpose of the monitoring coverage analysis is to ensure that the sensors can fully cover the key regions and avoid monitoring blind spots. For example, in the commutator segment group region, multiple temperature sensors can be evenly arranged along the arrangement direction of the commutator segments to monitor the temperature changes at different positions.

[0096] Exemplarily, in the commutator segment group region, assuming the length of the commutator segment is 10 cm and the width is 2 cm, a temperature sensor can be arranged every 2 cm along the length direction to ensure that the temperature changes of each commutator segment and its adjacent regions can be monitored; and then the sensor deployment parameters of this region are obtained, the number of sensors is 6, and the spacing is 2 cm.

[0097] Further, based on the monitoring sensor model and sensor deployment parameters, sensors are deployed in key areas to build a commutator monitoring sensor network. For example, in the commutator segment group area, thermocouple sensors, current transformers, and acceleration sensors are installed according to the above deployment parameters and connected to the data acquisition system to ensure that the sensors can work properly and transmit the collected data to the system.

[0098] Through the above process, reasonable sensor selection and deployment are achieved, so that the collected data can comprehensively and accurately reflect the operating status of the key areas of the commutator, providing reliable data support for subsequent steps such as multi-parameter fusion, modal decomposition, and fault identification.

[0099] S300: Perform multi-parameter fusion on the multi-source monitoring data streams of the M key areas respectively according to the application scenario information of the non-all-plastic hook-type commutator to obtain M key area fusion monitoring data streams.

[0100] Specifically, the application scenario information refers to the specific environment and working conditions information of the commutator in actual use, such as the speed range of the motor, load type, working environment temperature, humidity, etc. These information will affect the operating characteristics and fault modes of the commutator.

[0101] Specifically, multi-parameter fusion is performed on the multi-source monitoring data streams of each key area according to the application scenario information. For example, according to the speed range and load type of the motor, the weights of parameters such as temperature, vibration, and current are determined under different speeds and loads. In the case of low speed and high load, the weight of the vibration parameter may be higher; in the case of high speed and low load, the weight of the temperature parameter may be higher.

[0102] The above steps combine the collected multi-source monitoring data with the actual application scenario of the commutator for parameter fusion, forming a data stream that can comprehensively reflect the operating status of the commutator, providing a more adaptable and accurate information basis for subsequent modal decomposition and fault identification.

[0103] In some embodiments, obtaining the M key area fusion monitoring data streams includes:

[0104] Evaluate the influence degree of each area data monitoring source in the multi-source monitoring data streams of the M key areas respectively according to the application scenario information of the non-all-plastic hook-type commutator to obtain M area data monitoring source influence factor sets; determine M area data monitoring source credibility coefficient sets according to the M area data monitoring source influence factor sets; integrate the multi-source monitoring data streams of the M key areas according to the time sequence information to obtain M key area multi-source time sequence data streams; perform multi-parameter fusion on the M key area multi-source time sequence data streams based on the M area data monitoring source credibility coefficient sets to obtain the M key area fusion monitoring data streams.

[0105] Specifically, first, according to the application scenario information of the commutator, the influence degree of each regional data monitoring source on the operating state of the commutator is evaluated. In other words, under different motor speeds and load conditions, the influence degrees of monitoring sources such as temperature, vibration, and current on the operating state of the commutator are different. Exemplarily, methods such as correlation coefficient and principal component analysis can be used to analyze the contribution degrees of each monitoring source to the abnormality of the commutator in the historical monitoring data associated with the application scenario information, and M regional data monitoring source influence factor sets are generated. For example, for the commutator segment group area, under the conditions of high motor speed and low load, the influence degree of the temperature monitoring source may be relatively high, and the influence factor is set to 0.6; the influence degree of the vibration monitoring source is the second, and the influence factor is set to 0.3; the influence degree of the current monitoring source is relatively low, and the influence factor is set to 0.1.

[0106] Then, according to the M regional data monitoring source influence factor sets, M regional data monitoring source credibility coefficient sets are determined. Among them, the determination of the credibility coefficient can be based on the normalization processing of the influence factor. For example, each influence factor is divided by the sum of the same type of influence factors in the M regional data monitoring sources to obtain the credibility coefficient of each monitoring source. Exemplarily, the calculation formula can be expressed as:

[0107]

[0108] where C i is the credibility coefficient of the index item I in the i-th monitoring source, I i is the influence factor of the index item I in the i-th monitoring source, N is the number of monitoring sources with the index item I among the M regional data monitoring sources; I j is the influence factor of the index item I in the j-th monitoring source.

[0109] Furthermore, the multi-source monitoring data streams of the M key regions are integrated according to the time sequence information, that is, the multi-source monitoring data streams of the M key regions are aligned according to the time stamp to obtain the multi-source time sequence data streams of the M key regions. For example, for the commutator segment group area, the monitoring data such as temperature, vibration, and current are arranged in chronological order, and the values of these three types of data are available at each time point, forming a multi-source time sequence data stream. Optionally, compensation for missing values (difference method) is involved in the integration process.

[0110] Furthermore, finally, multi-parameter fusion is performed on the multi-source time sequence data streams of the M key regions based on the M regional data monitoring source credibility coefficient sets, such as weighted fusion according to the credibility coefficient, to obtain the fusion monitoring data streams of the M key regions. The fusion monitoring data streams obtained in this way can comprehensively reflect the operating state of the commutator at different time points, enabling the fault diagnosis method to adapt to a variety of different working conditions and environments, and having stronger versatility and flexibility.

[0111] S400: Perform modal decomposition and fault identification on the fused monitoring data stream of the M key regions to obtain M key region fault feature sets, and simultaneously construct a commutator fault diagnosis rule space.

[0112] Specifically, modal decomposition is used to decompose complex data into multiple intrinsic mode functions (IMFs), so as to better reflect the local characteristics of the signal. The hierarchical methods include empirical mode decomposition (EMD), ensemble empirical mode decomposition (EEMD), etc.; the fault feature set is a set composed of fault-related feature information extracted by analyzing the decomposed IMF components. Exemplarily, it can be the energy, kurtosis value, root mean square value, etc. in a specific frequency band; the fault diagnosis rule space is a rule base constructed based on historical fault data and expert knowledge, which contains the mapping relationships between various fault modes and corresponding features, and is used to guide the identification and diagnosis of faults.

[0113] In some embodiments, obtaining the M key region fault feature sets includes:

[0114] Perform empirical mode decomposition on the fused monitoring data stream of the M key regions respectively to obtain M key region IMF components; determine the normal working thresholds of the M key regions according to the performance working standards of the M commutator key region information; perform fault feature extraction on the M key region IMF components according to the normal working thresholds of the M key regions to obtain the M key region fault feature sets.

[0115] Specifically, first, perform modal decomposition on the fused monitoring data stream of the M key regions respectively. For example, for the fused data stream of the commutator segment group region, use the empirical mode decomposition (EMD) method to decompose it into multiple IMF components, and the multiple IMF components are arranged in order from high frequency to low frequency, corresponding to different modal characteristics in the signal respectively; then, determine the normal working thresholds of the M key regions according to the performance working standards of the non-all-plastic hook-type commutator, that is, the signal change range of the device in the healthy state. For example, calculate the mean (μ) and standard deviation (σ) of the IMF components, and set the normal working threshold: [μ - nσ, μ + nσ], where n can take 2 or 3, corresponding to the 95% or 99% confidence interval.

[0116] Next, fault feature extraction is performed on the decomposed IMF components according to these normal operating thresholds to obtain M key area fault feature sets. Exemplarily, the fault feature extraction methods include: statistical analysis method, calculating eigenvalue such as peak value, mean value, variance, skewness, kurtosis, etc. of each IMF component, and comparing with the normal operating threshold. If it exceeds the threshold, it is marked as an abnormal feature; time-frequency analysis method, performing Hilbert transform on the IMF component, calculating the instantaneous frequency and amplitude, and detecting whether there is an abnormal frequency shift; energy distribution analysis, calculating the energy proportion of the IMF component. If the energy of the low-frequency component suddenly increases, it may mean component wear or aging.

[0117] Through the above modal decomposition, complex fusion monitoring data can be decomposed into more easily analyzable IMF components, so as to more accurately identify fault features; among them, the construction of the fault feature set provides a direct basis for subsequent fault diagnosis.

[0118] In some embodiments, the construction of the commutator fault diagnosis rule space includes:

[0119] Dividing the non-all-plastic hook-type commutator fault database according to the M commutator key area information to obtain M key area fault data spaces; respectively performing feature analysis and extraction on the M key area fault data spaces to determine M key area fault feature spaces; performing fault diagnosis rule mining in the M key area fault feature spaces to construct the commutator fault diagnosis rule space.

[0120] Specifically, first, divide the non-all-plastic hook-type commutator fault database according to the M commutator key area information to obtain M key area fault data spaces, and each space contains the historical fault data of this key area; then, respectively perform feature analysis and extraction on the M key area fault data spaces (such as calculating statistical features such as mean value, standard deviation, peak value, skewness, kurtosis, etc.; using (principal component analysis) for dimensionality reduction, etc.) to determine M key area fault feature spaces. For example, for the fault data space of the commutator segment group area, it is found through analysis that the energy values in certain specific frequency bands will change significantly when a fault occurs, and the energy values of these frequency bands are extracted as fault features to form the fault feature space of this area; finally, perform fault diagnosis rule mining in the M key area fault feature spaces. For each key area, extract the parameter set that can best represent the fault features to form the M key area fault feature spaces.

[0121] Exemplarily, it is found that when a certain fault feature in the commutator segment group area exceeds a certain threshold, it often corresponds to a specific fault type, such as poor contact fault. Therefore, in the fault feature space of M key areas, data mining methods are used to extract fault diagnosis rules and construct a commutator fault diagnosis rule space, including a commutator fault diagnosis rule space based on decision trees, Apriori algorithms, fuzzy logic, or deep learning (LSTM or CNN models).

[0122] Exemplarily, the formed fault diagnosis rule space includes:

[0123] Contact fault rule space: Rule 1: Temperature > 85°C and current > 22A → Contact overheating fault; Rule 2: Current fluctuation > 5A and temperature rise rate > 2°C / s → Contact aging.

[0124] Spring fault rule space: Rule 1: Vibration > 1.8g → Spring mechanical fatigue; Rule 2: Abnormal rise in the energy of the high-frequency component of IMF → Risk of spring fracture.

[0125] Rotating shaft fault rule space: Rule 1: Torque change rate > 2 Nm / s and vibration > 1.5g → Bearing wear; Rule 2: Operating time > 5000 h and frequent speed fluctuations → Rotating shaft fatigue.

[0126] By systematically sorting and analyzing historical fault data, the association rules between fault features and fault types are mined. The formed fault diagnosis rule space provides intelligent decision support for subsequent fault diagnosis, helps to quickly match the real-time monitored fault features with known fault modes, and thus realizes accurate and efficient fault diagnosis.

[0127] In some embodiments, the construction of the commutator fault diagnosis rule space includes:

[0128] According to the fault feature space of the M key areas, M key area fault feature combinations and corresponding commutator diagnosis information are obtained; fault correlation topology analysis is performed on the M key area fault feature combinations to obtain a commutator fault cascade feature set; the commutator fault cascade feature set and the commutator diagnosis information are logically mapped and associated to construct the commutator fault diagnosis rule space.

[0129] Specifically, first, based on the fault feature spaces of M key regions, interactive historical diagnostic records or logs are used to obtain M combinations of fault features of key regions and corresponding commutator diagnostic information, which serve as samples for subsequent analysis. Then, a fault correlation topology analysis is performed on the M combinations of fault features of key regions to obtain a set of cascade features of commutator faults. In other words, during the operation of the commutator, faults in different key regions may affect each other and propagate cascadingly. For example: overheating of contacts → increase in contact resistance → abnormal current → ablation risk; spring fatigue → unstable commutation of the commutator → increase in shaft load; shaft wear → increase in vibration → further acceleration of spring aging. Organizing these cascade features into a set can more comprehensively reflect the occurrence and development process of faults.

[0130] Furthermore, a logical mapping association is made between the set of cascade features of commutator faults and the commutator diagnostic information to construct a commutator fault diagnosis rule space. When a certain combination of fault features is detected, based on its position and association relationship in the set of cascade features, combined with the corresponding diagnostic information, the type, severity, and possible impact range of the fault can be quickly determined.

[0131] S500: Based on the commutator fault diagnosis rule space, an integrated matching analysis is performed on the M sets of fault features of key regions to generate a commutator fault diagnosis result.

[0132] In some embodiments, the generation of the commutator fault diagnosis result includes:

[0133] Based on the commutator fault diagnosis rule space, a matching analysis is performed on the M sets of fault features of key regions to activate a set of matching commutator diagnosis rules; the set of matching commutator diagnosis rules is integrated and output to generate the commutator fault diagnosis result.

[0134] Specifically, first, according to the key region categories and positions of the M sets of fault features of key regions, a set of rules of the corresponding category is matched in the commutator fault diagnosis rule space, and the M sets of fault features of key regions are scanned, and each item is matched with the diagnostic rule library to obtain a set of matching commutator diagnosis rules. Then, an integration of the set of matching commutator diagnosis rules is carried out: if multiple rules involve the same key region, feature fusion is performed to reduce redundant information; if multiple faults belong to a cascade fault relationship, fault propagation analysis is carried out to determine the dominant fault; finally, the commutator fault diagnosis result is integrated and output. Exemplarily, a structured commutator fault diagnosis result is generated, including the device name, diagnosis time, and fault type.

[0135] In summary, the multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method provided by the present invention has the following technical effects:

[0136] By obtaining the design drawing information of the non-fully plastic hook-type commutator, identifying the key parts of the non-fully plastic hook-type commutator based on the design drawing information, and obtaining the information of M key regions of the commutator; deploying sensor analysis respectively on the information of M key regions of the commutator, building a commutator monitoring sensor network, and collecting multi-source monitoring data streams of M key regions through the sensor network; according to the application scenario information of the non-fully plastic hook-type commutator, performing multi-parameter fusion on the multi-source monitoring data streams of M key regions respectively to obtain the fused monitoring data streams of M key regions; performing modal decomposition and fault identification on the fused monitoring data streams of M key regions to obtain the fault feature sets of M key regions, and constructing a commutator fault diagnosis rule space; based on the commutator fault diagnosis rule space, performing integrated matching analysis on the fault feature sets of M key regions to generate the commutator fault diagnosis result, thereby achieving the technical effects of fusing multi-parameters, improving the information utilization efficiency, and improving the accuracy and response speed of fault identification.

[0137] Embodiment 2, such as Figure 2 is a schematic structural diagram of the multi-parameter fusion non-fully plastic hook-type commutator fault diagnosis system of the present invention. For example, Figure 1 In the present invention, the flow schematic diagram of the multi-parameter fusion non-fully plastic hook-type commutator fault diagnosis method can be implemented by a structure such as Figure 2 shown.

[0138] Based on the same concept as the multi-parameter fusion non-fully plastic hook-type commutator fault diagnosis method in the above embodiment, the multi-parameter fusion non-fully plastic hook-type commutator fault diagnosis system provided by the present invention further includes:

[0139] A region identification module 11, configured to obtain the design drawing information of the non-fully plastic hook-type commutator, and identify the key parts of the non-fully plastic hook-type commutator based on the design drawing information to obtain the information of M key regions of the commutator.

[0140] A region monitoring module 12, configured to deploy sensor analysis respectively on the information of M key regions of the commutator, build a commutator monitoring sensor network, and collect and obtain multi-source monitoring data streams of M key regions through the commutator monitoring sensor network.

[0141] A data stream fusion module 13, configured to perform multi-parameter fusion on the multi-source monitoring data streams of M key regions respectively according to the application scenario information of the non-fully plastic hook-type commutator to obtain the fused monitoring data streams of M key regions.

[0142] A fault feature acquisition module 14, configured to perform modal decomposition and fault identification on the fused monitoring data streams of M key regions to obtain the fault feature sets of M key regions, and simultaneously construct a commutator fault diagnosis rule space.

[0143] A fault diagnosis execution module 15, configured to perform integrated matching analysis on the M key area fault feature sets based on the commutator fault diagnosis rule space, and generate a commutator fault diagnosis result.

[0144] In some embodiments, the area identification module 11 includes:

[0145] A structural component splitting unit, configured to split the non-all-plastic hook-type commutator into structural components based on the design drawing information, and obtain a commutator structural component set.

[0146] A component application function marking unit, configured to respectively mark the application functions of each structural component in the commutator structural component set, and determine a commutator component application function set.

[0147] A functional area set integration unit, configured to perform adjacent component integration on the commutator structural component set according to the commutator component application function set, and obtain N commutator functional area sets.

[0148] A key area information screening unit, configured to evaluate and screen the N commutator functional area sets, and obtain M commutator key area information, where N≥M.

[0149] In some implementation manners, the key area information screening unit in the area identification module 11 includes:

[0150] A historical fault data crawling unit, configured to crawl historical fault data based on the N commutator functional area sets, and obtain a non-all-plastic hook-type commutator fault database.

[0151] A functional area fault data set generation unit, configured to perform mapping and splitting on the non-all-plastic hook-type commutator fault database according to the N commutator functional area sets, and obtain N functional area fault data sets.

[0152] A functional area fault occurrence probability calculation unit, configured to perform frequency statistics and probability calculation based on the N functional area fault data sets, and obtain N functional area fault occurrence probabilities.

[0153] The key area information screening unit, configured to preset a fault probability threshold according to the N functional area fault occurrence probabilities, and perform comparison and screening on the N commutator functional area sets based on the fault probability threshold, so as to obtain the M commutator key area information.

[0154] In some embodiments, the area monitoring module 12 includes:

[0155] The monitoring requirement analysis unit is used to sequentially analyze the monitoring requirements for the M key area information of the commutator, and determine the M key area monitoring types and the M key area monitoring value ranges.

[0156] The monitoring sensor model selection unit is used to select the M key area monitoring sensor models according to the M key area monitoring types and the M key area monitoring value ranges.

[0157] The sensor deployment parameter acquisition unit is used to perform monitoring coverage analysis on the M key area information of the commutator based on the M key area monitoring sensor models, and obtain the M key area sensor deployment parameters.

[0158] The commutator monitoring sensor network construction unit is used to deploy sensors on the M key area information of the commutator based on the M key area monitoring sensor models and the M key area sensor deployment parameters, and construct the commutator monitoring sensor network.

[0159] In some embodiments, the data stream fusion module 13 includes:

[0160] The regional data monitoring source impact factor evaluation unit is used to evaluate the impact degree of each regional data monitoring source in the M key area multi-source monitoring data streams according to the application scenario information of the non-all-plastic hook-type commutator, and obtain the M regional data monitoring source impact factor sets.

[0161] The regional data monitoring source credibility coefficient determination unit is used to determine the M regional data monitoring source credibility coefficient sets according to the M regional data monitoring source impact factor sets.

[0162] The key area multi-source time-series data stream integration unit is used to integrate the M key area multi-source monitoring data streams according to the time-series information, and obtain the M key area multi-source time-series data streams.

[0163] The key area fusion monitoring data stream generation unit is used to perform multi-parameter fusion on the M key area multi-source time-series data streams based on the M regional data monitoring source credibility coefficient sets, and obtain the M key area fusion monitoring data streams.

[0164] In some embodiments, the fault feature acquisition module 14 includes:

[0165] The key area empirical mode decomposition unit is used to perform empirical mode decomposition on the M key area fusion monitoring data streams respectively, and obtain the M key area IMF components.

[0166] The key area normal working threshold determination unit is used to determine the M key area normal working thresholds according to the performance working standards of the M key area information of the commutator.

[0167] A key area fault feature extraction unit, configured to extract fault features from the IMF components of the M key areas according to the normal operation thresholds of the M key areas, so as to obtain the M key area fault feature sets.

[0168] In some embodiments, the fault feature acquisition module 14 includes:

[0169] A key area fault data space partitioning unit, configured to partition the non-all-plastic hook-type commutator fault database according to the M commutator key area information to obtain M key area fault data spaces.

[0170] A key area fault feature space determination unit, configured to perform feature analysis and extraction on the M key area fault data spaces respectively to determine M key area fault feature spaces.

[0171] A commutator fault diagnosis rule space construction unit, configured to perform fault diagnosis rule mining in the M key area fault feature spaces to construct a commutator fault diagnosis rule space.

[0172] In some embodiments, the fault feature acquisition module 14 includes:

[0173] A key area fault feature combination and diagnosis information acquisition unit, configured to obtain M key area fault feature combinations and corresponding commutator diagnosis information according to the M key area fault feature spaces.

[0174] A fault correlation topology analysis unit, configured to perform fault correlation topology analysis on the M key area fault feature combinations to obtain a commutator fault cascade feature set.

[0175] A commutator fault diagnosis rule space construction unit, configured to perform logical mapping association on the commutator fault cascade feature set and the commutator diagnosis information to construct the commutator fault diagnosis rule space.

[0176] In some embodiments, the fault diagnosis execution module 15 includes:

[0177] A commutator fault diagnosis rule matching unit, configured to perform matching analysis on the M key area fault feature sets based on the commutator fault diagnosis rule space to activate a matching commutator diagnosis rule set.

[0178] A commutator fault diagnosis result generation unit, configured to integrate and output the matching commutator diagnosis rule set to generate the commutator fault diagnosis result.

[0179] It should be understood that the embodiments mentioned in this specification focus on their differences from other embodiments. The specific embodiments in the first embodiment mentioned above are equally applicable to the multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis system described in the second embodiment. For the sake of brevity of the specification, no further elaboration will be made here.

[0180] It should be understood that the embodiments and the above descriptions disclosed in the present invention enable those skilled in the art to implement the present invention by using the present invention. At the same time, the present invention is not limited to the above-mentioned part of the embodiments. It should be understood that those of ordinary skill in the art can still modify the technical solutions recorded in the foregoing embodiments or make equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A multi-parameter fusion fault diagnosis method for a non-all-plastic hook-type commutator, characterized in that Including: Obtain the design drawing information of the non-all-plastic hook-type commutator, identify the key parts of the non-all-plastic hook-type commutator based on the design drawing information, and obtain M commutator key area information; Perform sensor analysis and deployment on the M commutator key area information respectively, build a commutator monitoring sensor network, and collect and obtain multi-source monitoring data streams of M key areas through the commutator monitoring sensor network; Perform multi-parameter fusion on the multi-source monitoring data streams of the M key areas respectively according to the application scenario information of the non-all-plastic hook-type commutator to obtain M key area fusion monitoring data streams; Perform modal decomposition and fault identification on the M key area fusion monitoring data streams to obtain M key area fault feature sets, and at the same time build a commutator fault diagnosis rule space; Perform integrated matching analysis on the M key area fault feature sets based on the commutator fault diagnosis rule space to generate a commutator fault diagnosis result.

2. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 1, characterized in that, The obtaining of the M commutator key area information includes: Split the structural components of the non-all-plastic hook-type commutator based on the design drawing information to obtain a set of commutator structural components; Mark the application functions of each structural component in the set of commutator structural components respectively to determine a set of commutator component application functions; Integrate adjacent components of the set of commutator structural components according to the set of commutator component application functions to obtain N sets of commutator functional areas; Evaluate and screen the N sets of commutator functional areas to obtain M commutator key area information, where N≥M.

3. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 2, wherein The obtaining of the M commutator key area information includes: Crawl historical fault data based on the N sets of commutator functional areas to obtain a non-all-plastic hook-type commutator fault database; Map and split the non-all-plastic hook-type commutator fault database according to the N sets of commutator functional areas to obtain N functional area fault data sets; Perform frequency statistics and probability calculation based on the N functional area fault data sets to obtain the fault occurrence probabilities of N functional areas; Preset a fault probability threshold according to the fault occurrence probabilities of the N functional areas, and compare and screen the N sets of commutator functional areas based on the fault probability threshold to obtain the M commutator key area information.

4. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 1, characterized in that, The building of the commutator monitoring sensor network includes: Analyze the monitoring requirements of the M commutator key area information in sequence to determine the monitoring types of M key areas and the monitoring value ranges of M key areas; Select the monitoring sensor models of M key areas according to the monitoring types of M key areas and the monitoring value ranges of M key areas; Perform monitoring coverage analysis on the M commutator key area information based on the monitoring sensor models of M key areas to obtain the sensor deployment parameters of M key areas; Deploy sensors on the M commutator key area information based on the monitoring sensor models of M key areas and the sensor deployment parameters of M key areas to build the commutator monitoring sensor network.

5. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 1, characterized in that, The obtaining of the M key area fusion monitoring data streams includes: Evaluate the influence degree of each regional data monitoring source in the multi-source monitoring data streams of the M key regions according to the application scenario information of the non-all-plastic hook-type commutator, and obtain an influence factor set of M regional data monitoring sources; Determine a credibility coefficient set of M regional data monitoring sources according to the influence factor set of M regional data monitoring sources; Integrate the multi-source monitoring data streams of the M key regions according to the time sequence information to obtain multi-source time-sequence data streams of the M key regions; Perform multi-parameter fusion on the multi-source time-sequence data streams of the M key regions based on the credibility coefficient set of the M regional data monitoring sources to obtain the fused monitoring data streams of the M key regions.

6. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 1, characterized in that, The obtaining of the fault feature sets of the M key regions includes: Perform empirical mode decomposition on the fused monitoring data streams of the M key regions respectively to obtain IMF components of the M key regions; Determine normal working thresholds of the M key regions according to the performance working standards of the M key region information of the commutator; Extract fault features from the IMF components of the M key regions according to the normal working thresholds of the M key regions to obtain the fault feature sets of the M key regions.

7. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 3, characterized in that, The construction of the commutator fault diagnosis rule space includes: Divide the non-all-plastic hook-type commutator fault database according to the M key region information of the commutator to obtain M key region fault data spaces; Perform feature analysis and extraction on the M key region fault data spaces respectively to determine M key region fault feature spaces; Mine fault diagnosis rules within the M key region fault feature spaces to construct a commutator fault diagnosis rule space.

8. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 7, wherein The construction of the commutator fault diagnosis rule space includes: Obtain fault feature combinations of the M key regions and corresponding commutator diagnosis information according to the M key region fault feature spaces; Perform fault-associated topology analysis on the fault feature combinations of the M key regions to obtain a commutator fault cascade feature set; Perform logical mapping association on the commutator fault cascade feature set and the commutator diagnosis information to construct the commutator fault diagnosis rule space.

9. The multi-parameter fusion non-all-plastic hook-type commutator fault diagnosis method according to claim 1, characterized in that, The generation of the commutator fault diagnosis result includes: Perform matching analysis on the fault feature sets of the M key regions based on the commutator fault diagnosis rule space to activate a matching commutator diagnosis rule set; Integrate and output the matching commutator diagnosis rule set to generate the commutator fault diagnosis result.

10. A non-all-plastic hook-type commutator fault diagnosis system with multi-parameter fusion, characterized in that, A non-all-plastic hook-type commutator fault diagnosis method for performing the multi-parameter fusion according to any one of claims 1-9 includes: A region recognition module, configured to obtain design drawing information of a non-all-plastic hook-type commutator, and identify key parts of the non-all-plastic hook-type commutator based on the design drawing information to obtain M key region information of the commutator; A region monitoring module, configured to respectively perform sensor analysis and deployment on the M key region information of the commutator, build a commutator monitoring sensor network, and collect and obtain multi-source monitoring data streams of the M key regions through the commutator monitoring sensor network; A data flow fusion module, configured to perform multi-parameter fusion on the multi-source monitoring data flows of the M key regions respectively according to the application scenario information of the non-all-plastic hook-type commutator, so as to obtain M key region fusion monitoring data flows; A fault feature acquisition module, configured to perform modal decomposition and fault identification on the M key region fusion monitoring data flows, so as to obtain M key region fault feature sets, and simultaneously construct a commutator fault diagnosis rule space; A fault diagnosis execution module, configured to perform integrated matching analysis on the M key region fault feature sets based on the commutator fault diagnosis rule space, and generate a commutator fault diagnosis result.

Images