Fault detection method for special release motor

By acquiring and analyzing the multi-source detection information and environmental information of the motor, generating the truth value and deviation matrix of the multi-source detection indicators, and performing abnormal analysis, the problem of inaccurate fault detection in the existing technology is solved, and more accurate fault prediction and Efficient maintenance.

CN118967106AInactive Publication Date: 2024-11-15WOYI HEALTH TECH CO LTD

Patent Information

Application Number
CN202411148709.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology relies on single parameter detection and simple statistical analysis, and cannot fully capture complex fault signals during motor operation, resulting in inaccurate fault detection and untimely maintenance.

Method used

By obtaining the multi-source detection information of the special release motor in the health detection equipment, combining the motor control status and environment information, the multi-source detection index truth value is generated, the multi-source detection deviation matrix is ​​established, and abnormal analysis is carried out, and fault prediction is used to use intelligent algorithms and maintain.

Benefits of technology

It realizes more accurate fault detection and efficient maintenance, improving the accuracy and maintenance efficiency of fault detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967106A_ABST
    Figure CN118967106A_ABST
Patent Text Reader

Abstract

The invention provides a special release motor fault detection method, and relates to the technical field of fault detection, and the method comprises the steps: obtaining the multi-source detection information of a special release motor; acquiring a motor control state and motor environment information; determining a motor reference state according to the fault reason, the motor control state and the motor environment information, and generating a multi-source detection index true value; performing deviation analysis according to the multi-source detection index truth value and the multi-source detection information, and establishing a multi-source detection deviation matrix; performing anomaly analysis on the multi-source detection deviation matrix and the fault association relationship to obtain a motor fault prediction result; and if the motor fault operator is obtained, extracting a fault association relationship for maintenance. According to the method and the device, the technical problem that the accuracy of fault detection is affected due to dependence on single parameter detection and simple statistical analysis in the prior art can be solved, the technical targets of comprehensively analyzing multiple detection parameters and performing motor fault prediction by using an intelligent algorithm are achieved, and the technical effect of improving the accuracy of fault detection is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of fault detection, and in particular to a method for detecting a special release motor fault. Background Art

[0002] Motor fault prediction and maintenance is one of the key technologies to ensure equipment reliability and safety in modern industry. At present, traditional motor fault prediction methods mainly rely on the detection of single parameters and simple statistical analysis, but this method has great limitations and cannot fully capture the complex fault signals in motor operation, resulting in inaccurate fault detection and untimely maintenance. At the same time, the existing technology is also relatively weak in the extraction and application of fault correlation relationships, often relying on experience judgment, lacking systematic and data-driven analysis, affecting the effectiveness and scientificity of fault prediction. Therefore, there is an urgent need for a motor fault prediction method based on multi-source detection information and intelligent algorithms that can comprehensively analyze multiple parameters to achieve more accurate fault detection and efficient maintenance.

[0003] In summary, the prior art has technical problems in that it relies on single parameter detection and simple statistical analysis methods, which makes it impossible to fully capture complex fault signals in motor operation, further affecting the accuracy of fault detection and the timeliness and effectiveness of maintenance. Summary of the invention

[0004] The purpose of this application is to provide a special release motor fault detection method to solve the technical problem that the prior art relies on single parameter detection and simple statistical analysis methods, which leads to the inability to fully capture the complex fault signals in the operation of the motor, further affecting the accuracy of fault detection and the timeliness and effectiveness of maintenance.

[0005] In view of the above problems, the present application provides a special release motor fault detection method.

[0006] The present application provides a special release motor fault detection method, including: obtaining multi-source detection information of a special release motor in a health detection device, wherein the multi-source detection information has a fault correlation relationship with the fault cause; obtaining the motor control state and motor environment information of the special release motor; determining the motor reference state according to the fault cause, the motor control state and the motor environment information, and generating a multi-source detection index true value; performing deviation analysis on the multi-source detection index true value and the multi-source detection information, and establishing a multi-source detection deviation matrix; performing an abnormality analysis on the multi-source detection deviation matrix and the fault correlation relationship, and obtaining a motor fault prediction result; if a motor fault operator is obtained based on the motor fault prediction result, extracting the fault correlation relationship for maintenance.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: By acquiring multi-source detection information of a special release motor in a health detection device, wherein the multi-source detection information has a fault correlation relationship with the fault cause; acquiring the motor control state and motor environment information of the special release motor; determining the motor reference state according to the fault cause, the motor control state and the motor environment information, and generating a multi-source detection index true value; performing deviation analysis on the multi-source detection index true value and the multi-source detection information, and establishing a multi-source detection deviation matrix; performing an abnormality analysis on the multi-source detection deviation matrix and the fault correlation relationship, and obtaining a motor fault prediction result; if a motor fault operator is obtained based on the motor fault prediction result, the fault correlation relationship is extracted for maintenance, thereby achieving the technical goal of comprehensively analyzing multiple detection parameters and using an intelligent algorithm to predict motor faults, and achieving the technical effect of improving fault detection accuracy and maintenance efficiency.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 This is a flow chart of a special release motor fault detection method for this application.

[0011] Figure 2 A schematic diagram of a process for generating true values ​​of multi-source detection indicators in a special release motor fault detection method of the present application. DETAILED DESCRIPTION

[0012] This application provides a special release motor fault detection method to solve the technical problem that the existing technology cannot fully capture the complex fault signals in the motor operation due to the reliance on single parameter detection and simple statistical analysis methods, which further affects the accuracy of fault detection and the timeliness and effectiveness of maintenance. The technical goal of comprehensively analyzing multiple detection parameters and using intelligent algorithms to predict motor faults is achieved, achieving the technical effect of improving fault detection accuracy and maintenance efficiency.

[0013] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0014] For examples, please see the attached Figure 1 , the present application provides a special release motor fault detection method, specifically including: Step 1: Acquire multi-source detection information of a special release motor in a health detection device, wherein the multi-source detection information has a fault correlation relationship with a fault cause.

[0015] Specifically, health detection equipment refers to equipment such as smart bracelets and health watches that have special release motors for automatic calibration of sensors. Multi-source detection information of special release motors in health detection equipment is obtained by integrating multiple detection methods. Acquiring multi-source detection information is not limited to the data collection of a single sensor, but uses multiple technical means such as current detection, vibration analysis, temperature monitoring, sound detection and insulation resistance testing to comprehensively capture abnormal phenomena that may occur during the operation of the motor. Through multi-source information collection, the operating status of the motor can be more accurately reflected, and the comprehensiveness and reliability of fault detection can be improved. There is a close correlation between multi-source detection information and the cause of the fault. The data obtained by each detection method corresponds to a specific fault that may occur in the motor. For example, current detection can detect whether the motor is overloaded or short-circuited, vibration analysis can identify wear or looseness of mechanical parts, temperature monitoring can detect motor overheating problems, sound detection can identify abnormal noise of internal parts, and insulation resistance testing can detect electrical insulation degradation. By analyzing multi-source detection data, the specific cause of the fault can be traced back to provide a basis for timely maintenance measures. In the process of obtaining multi-source detection information, data from different sensors can be comprehensively analyzed through data fusion technology. Data fusion technology can process heterogeneous data from different sensors and extract more representative feature information, thereby improving the accuracy of fault detection.

[0016] Step 2: Obtain the motor control state and motor environment information of the special release motor.

[0017] Specifically, the motor control state and motor environment information of the special release motor are obtained. The motor control state includes parameters such as the motor's operating mode, output power, voltage and current conditions, which can be obtained through real-time monitoring of the control system. The acquisition of the motor control state can help determine whether the motor is in normal working condition and under what conditions the fault occurs. By real-time detection and recording of control state data, historical data analysis can be performed to find abnormal operating modes, thereby predicting and preventing potential faults. Obtaining motor environment information ensures stable operation of the motor. Motor environment information includes external conditions such as temperature, humidity, vibration, and noise, which have a direct impact on the normal operation of the motor. For example, too high temperature may cause the motor to overheat, too high humidity may cause electrical insulation problems, and excessive vibration and noise may be a signal of wear or looseness of mechanical parts. By installing environmental sensors, environmental changes around the motor can be detected in real time to ensure that the motor works in a suitable environment and to promptly discover and deal with potential faults caused by environmental problems.

[0018] Step three: Determine the motor reference state according to the fault cause, motor control state and motor environment information, and generate a true value of a multi-source detection index.

[0019] Specifically, the motor reference state is determined according to the fault cause, motor control state and motor environmental information, and the true value of the multi-source detection index is generated. Comprehensively analyze various types of detection data to determine the reference state of the motor under normal working conditions, and use it as a comparison benchmark to evaluate the current state, thereby generating the true value of the multi-source detection index. The determination of the motor reference state requires comprehensive consideration of the motor's control state and environmental information. By long-term detection and recording of the motor's voltage, current, power and other parameters in different working modes, combined with environmental conditions such as temperature, humidity, vibration and other information, the typical parameter range of the motor under normal working conditions is established, which is the motor's reference state. The determination of the reference state requires a large amount of historical data support. Through statistical analysis and data mining, the normal fluctuation range and typical value of each parameter can be found. Generating the true value of the multi-source detection index requires a comprehensive analysis of the real-time detection data. The multi-source detection indicators include the motor's electrical parameters (such as voltage, current, power), mechanical parameters (such as vibration, noise), environmental parameters (such as temperature, humidity), etc., which represent the ideal working state of the motor under the current environment and control state. By comparing with the reference state, it is possible to evaluate whether the current motor operating state is normal, whether there are abnormal fluctuations or deviations from the reference range.

[0020] Step 4: Perform deviation analysis based on the true value of the multi-source detection index and the multi-source detection information to establish a multi-source detection deviation matrix.

[0021] Specifically, the true value of the multi-source detection index is obtained. The true value of the multi-source detection index is obtained by statistically analyzing the historical data and the normal operation data. The true value of the multi-source detection index is the standard value of each detection index under an ideal or normal working state. The true value of the multi-source detection index may include key parameters such as the working current, vibration spectrum, working temperature, working sound and insulation resistance of the motor. The current multi-source detection information is obtained. The multi-source detection information is the motor operation data collected by the real-time detection equipment, which reflects the actual operation of the motor under the current environment and working state. The multi-source detection information corresponds to the true value of the multi-source detection index one by one and covers the same detection parameters. The real-time detection information is compared with the true value of the detection index, and a deviation analysis is performed to calculate the deviation value of each detection parameter. According to the calculation results, the deviation modulus (i.e., absolute value) and deviation direction (i.e., positive and negative value) of each parameter are obtained. The deviation modulus indicates the size of the gap between the current value and the true value, and the deviation direction indicates whether the gap is a positive deviation or a negative deviation. Construct a deviation vector. The deviation modulus and deviation direction of each detection parameter can be combined into a deviation vector. The deviation vectors of all detection parameters are collected to form a complete deviation vector set. All deviation vectors are arranged in chronological order to form a multi-dimensional deviation matrix. Each row of the multi-source detection deviation matrix represents a deviation vector at a time point, and each column represents the deviation of a detection parameter. By analyzing the multi-source detection deviation matrix, it is possible to systematically identify abnormal conditions in motor operation, understand the degree and direction of deviation of each detection parameter at different time points, and provide data support for early warning of motor failures and maintenance decisions.

[0022] Step 5: Perform an abnormal analysis on the multi-source detection deviation matrix and the fault association relationship to obtain a motor fault prediction result.

[0023] Specifically, anomaly analysis of the multi-source detection deviation matrix can detect abnormal conditions in motor operation by identifying and evaluating the deviation values ​​of each detection parameter. The multi-source detection deviation matrix records the deviation values ​​of various parameters of the motor at different time points, and the deviation values ​​reflect the difference between the current state and the normal state. Through statistical methods, such as calculating the mean and standard deviation of the deviation values, abnormal data points that are beyond the normal range can be identified. In addition, using machine learning algorithms, such as cluster analysis and anomaly detection algorithms, abnormal patterns can be automatically detected and classified. Anomaly analysis methods help identify abnormal phenomena in the operation of the motor and provide a basis for further fault analysis. Fault association relationships are used for analysis to link abnormal data in the deviation matrix with specific fault types. Fault association relationships describe the relationship between different detection parameters and various fault types, and are models established based on historical data and experience. By comparing the abnormal data and fault association relationships in the deviation matrix, it can be determined whether these anomalies correspond to a specific fault type. Comparative analysis helps us understand the fault causes behind the abnormal data and associate the detection data with the actual fault phenomenon, thereby improving the accuracy and pertinence of fault diagnosis. Combined with the abnormal analysis of the multi-source detection deviation matrix and the comparison results of the fault association relationship, the running state of the motor can be comprehensively evaluated to obtain the motor fault prediction results. The identified abnormal data is combined with the fault model for comprehensive analysis. By analyzing these data, the predicted probability of each fault type is calculated, and the possible fault type and probability of occurrence are determined by combining the historical fault data and the current abnormal data using Bayesian reasoning or other statistical methods. Comprehensive analysis not only helps us identify current faults, but also predicts possible future faults, providing a basis for early warning and prevention. According to the results of the comprehensive analysis, the motor fault prediction results are output. Including possible fault types, predicted probabilities and recommended treatment measures, providing specific operation guidelines for maintenance personnel. The accuracy and reliability of the prediction results depend on the detailed analysis and comparison of the previous steps to ensure that the fault prediction has a scientific basis and practical reference value. Through the detailed fault prediction results, preventive and maintenance measures can be taken in time to avoid the further development of motor faults and ensure the safe and stable operation of the motor.

[0024] Step six: If a motor fault operator is obtained based on the motor fault prediction result, the fault association relationship is extracted for maintenance.

[0025] Specifically, a motor fault operator is obtained based on the motor fault prediction result. When the motor fault prediction result shows that the running state of the motor has reached or exceeded the preset fault threshold, it indicates that the motor may be about to fail or has already failed. Generate a motor fault operator. The motor fault operator contains detailed fault information, including fault type, prediction probability, relevant detection parameters and possible fault causes. Through the fault operator, the nature, scope and impact of the fault can be clarified to provide precise guidance for subsequent maintenance. Extract fault associations for analysis. Using the fault information in the motor fault operator, relevant fault associations are extracted from historical data and fault models. By analyzing the associations, the cause of the fault can be further confirmed, and a basis can be provided for formulating specific maintenance measures. After extracting the fault associations, maintenance operations are performed. According to the analysis results of the fault operator and the associations, a specific maintenance plan is formulated.

[0026] The special release motor fault detection method can achieve the technical goal of comprehensively analyzing multiple detection parameters and using intelligent algorithms to predict motor faults, thereby achieving the technical effect of improving fault detection accuracy and maintenance efficiency.

[0027] Furthermore, the present application also includes: The multi-source detection information includes motor operating current, motor vibration spectrum, motor operating temperature, motor operating sound, and motor insulation resistance.

[0028] Specifically, multi-source detection information includes motor operating current, motor vibration spectrum, motor operating temperature, motor operating sound and motor insulation resistance. By obtaining multi-source detection information, the operating status of the motor can be fully monitored and possible motor faults can be analyzed from multiple angles. Motor operating current is an important detection indicator. By detecting the current changes of the motor under different loads and operating conditions, it can be determined whether the motor is in normal working condition. Abnormal current fluctuations may indicate motor overload, short circuit or electrical system failure. Real-time current detection can provide timely fault warning to avoid further damage. Motor vibration spectrum provides important information about the mechanical state. By analyzing the vibration spectrum of the motor, bearing wear, gear failure or other mechanical problems can be detected. Vibration spectrum analysis usually uses sensors such as accelerometers. The spectrum diagram can identify abnormal vibrations at specific frequencies to help locate the location and nature of mechanical faults. Motor operating temperature is another key detection parameter. Excessive operating temperature may cause overheating and damage to the internal components of the motor, or even cause a fire. The temperature sensor can detect the temperature changes of the motor in real time, and issue an alarm in time when the temperature rises abnormally, ensuring that the motor operates within a safe temperature range. Temperature data can also be used to analyze the heat dissipation performance of the motor and the impact of the working environment. Motor working sound detection is mainly used to identify abnormal noise. The motor will produce normal operating sounds during operation, but if there is abnormal noise, such as sharp or irregular sounds, it may mean wear, looseness or other faults of internal parts. Sound detection through a microphone or acoustic sensor can capture these abnormal sound signals to assist in diagnosing motor faults. Motor insulation resistance is an important indicator for judging the insulation performance of the motor. Too low insulation resistance may cause motor leakage or short circuit, affecting the normal operation and safety of the motor. By regularly testing the insulation resistance of the motor, the deterioration of the insulation layer can be discovered, and maintenance and replacement can be carried out in time to ensure the electrical safety of the motor.

[0029] By acquiring and analyzing multi-source detection information such as motor operating current, motor vibration spectrum, motor operating temperature, motor operating sound, and motor insulation resistance, the operating status of the motor can be fully monitored. Determining the motor baseline status and generating the true value of multi-source detection indicators can help to timely detect and prevent motor failures and improve the operating reliability and safety of the equipment.

[0030] Further, if Figure 2 As shown, this application also includes: Based on the fault cause, motor control state and motor environment information as constraints, a multi-step association search is performed to obtain a multi-source detection information record value set; the multi-source detection information record value set is traversed to perform confidence analysis to generate the true value of the multi-source detection indicator.

[0031] Specifically, based on the fault cause, motor control state and motor environmental information as constraints, a multi-step correlation search is performed to obtain a multi-source detection information record value set. The fault cause, motor control state and environmental information are used as search constraints to systematically retrieve and collect relevant detection data. The specific operation includes extracting detection data records that meet these constraints from the historical database, and gradually filtering and aggregating relevant data through a multi-step correlation search algorithm to form a complete multi-source detection information record value set. The multi-source detection information record value set contains specific values ​​of multiple indicators such as motor working current, motor vibration spectrum, motor working temperature, motor working sound and motor insulation resistance. Then, the fault cause, motor control state and motor environmental information are used as constraints to define the scope and direction of data search. Through the correlation search, data that is irrelevant to the current fault or has a lot of noise can be effectively filtered out to ensure the relevance and accuracy of the search results. The multi-step correlation search algorithm uses these constraints to filter and aggregate relevant data layer by layer to form a detailed multi-source detection information record value set.

[0032] Then, the multi-source detection information record value set is accessed in sequence for confidence analysis to generate the true value of the multi-source detection index. Confidence analysis refers to the analysis of the data in the multi-source detection information record value set by statistical methods to evaluate the credibility and consistency of the data. The specific steps include calculating the mean, variance and confidence interval of each detection index, and identifying and eliminating outliers. By accessing the entire data set in sequence, analyzing the confidence of each record one by one, and performing weighted averaging on the data with higher credibility, the true value of each detection index is finally generated. The true value represents the most reliable and true value of each detection parameter of the motor under specific control states and environmental conditions.

[0033] By taking the fault cause, motor control status and motor environment information as constraints, a multi-step correlation search is performed to obtain a multi-source detection information record value set. Confidence analysis is performed by traversing the record value set to finally generate the true value of the multi-source detection indicator, thereby achieving high efficiency and accuracy in motor fault detection and providing solid data support for the stable operation and maintenance of the motor.

[0034] Furthermore, the present application also includes: Taking the fault cause as a search constraint and multi-source sensor detection as a search target, a search is performed to obtain a first-step sensor record value set; taking the motor control state as a search constraint and multi-source sensor detection as a search target, a search is performed to obtain a second-step sensor record value set; taking the motor environment information as a search constraint and multi-source sensor detection as a search target, a search is performed to obtain a third-step sensor record value set; the first-step sensor record value set, the second-step sensor record value set and the third-step sensor record value set are searched based on the multi-source detection information record value set with the multi-source sensor detection as a connection constraint.

[0035] Specifically, the search is performed with the cause of the fault as the search constraint and multi-source sensor detection as the search target to obtain the first step sensor record value set. First, by analyzing historical data, the multi-source sensor detection data related to the specific fault cause is determined, such as motor overload fault, electrical short circuit or mechanical wear. With the fault type as the constraint, data retrieval and screening are performed to collect multi-source sensor detection data corresponding to the occurrence of the fault, such as motor operating current, vibration spectrum, operating temperature, operating sound and insulation resistance. The first step sensor record value set formed contains the sensor data under the specific fault scenario.

[0036] Next, the motor control state is used as the search constraint and multi-source sensor detection is used as the search target to obtain the second-step sensor record value set. The data search scope is further refined by the motor control state, such as the motor operation mode (startup, normal operation, shutdown, etc.), output power, voltage and current conditions, etc. With the control state as the search constraint, the multi-source sensor detection data in these states in the historical database is retrieved to form the second-step sensor record value set. This helps to understand the operating characteristics and sensor readings of the motor under different control states.

[0037] Next, the motor environment information is used as the search constraint and multi-source sensor detection is used as the search target to obtain the third step sensor record value set. The data search scope is further limited by the motor environment information (such as temperature, humidity, vibration, and noise). With the environmental conditions as the search constraint, multi-source sensor detection data in a specific environment is collected to form the third step sensor record value set. This reflects the impact of environmental factors on the motor operation status.

[0038] Then, the first step sensor record value set, the second step sensor record value set and the third step sensor record value set are subjected to multi-source detection information record value set based on multi-source sensor detection as the connection constraint. By comprehensively analyzing and integrating the sensor record value sets obtained in the first step, the second step and the third step, a complete multi-source detection information record value set is formed. Using multi-source sensor detection data as a connection constraint, the sensor data under different conditions are associated and compared to ensure the integrity and consistency of the data. Comprehensive detection information reflecting the motor under different faults, control states and environmental conditions can be obtained, providing reliable data support for subsequent fault detection and diagnosis.

[0039] By taking the fault cause, motor control status and motor environment information as constraints, multi-source sensor detection data is searched and collected step by step, and finally a comprehensive multi-source detection information record value set is formed. By comprehensively analyzing and integrating the three record value sets, a comprehensive multi-source detection information record value set is generated, providing accurate and reliable data support for motor fault detection and diagnosis.

[0040] Furthermore, the present application also includes: Extract the first detection constraint record value set of the multi-source detection information record value set; extract the reference detection constraint record value whose record value span with the first detection constraint record value is less than the record value span threshold based on the first detection constraint record value set; generate the first record discreteness according to the mean of the distance between the first detection constraint record value and the reference detection constraint record value; traverse the first detection constraint record value set to obtain the record discreteness mean, and generate the first discreteness according to the ratio of the first record discreteness to the record discreteness mean; if the first discreteness exceeds the discreteness threshold, eliminate the first detection constraint record value, traverse the first detection constraint record value set, obtain the first detection constraint true value, and add it to the multi-source detection indicator true value.

[0041] Specifically, extract the first detection constraint record value set of the multi-source detection information record value set. First, through the predefined detection constraint conditions, extract the record values ​​that meet the conditions from the multi-source detection information record value set to form the first detection constraint record value set. Ensure that the detection data related to the specific constraint conditions are obtained, thereby improving the pertinence and accuracy of the analysis.

[0042] Next, based on the first detection constraint record value set, reference detection constraint record values ​​whose record value span with the first detection constraint record value is less than the record value span threshold are extracted. Reference detection constraint record values ​​whose record value span with the current record value is less than a preset threshold are selected from the first detection constraint record value set. The reference detection value is numerically close to the current detection value and can provide a relatively reliable reference benchmark.

[0043] Next, the first record dispersion is generated according to the mean of the distance between the first detection constraint record value and the reference detection constraint record value. The mean of the distance between the first detection constraint record value and its reference detection constraint record value is calculated. This reflects the degree of dispersion of the first detection constraint record value relative to the reference value, and the concentration and reliability of the data can be evaluated through this dispersion.

[0044] Then, the first detection constraint record value set is accessed in sequence to obtain the record dispersion mean. The entire first detection constraint record value set is accessed in sequence to calculate the dispersion mean of all record values. The dispersion mean provides an overall discreteness evaluation benchmark, which enables the dispersion of a single record value to be compared with the whole.

[0045] Next, if the first discreteness exceeds the discreteness threshold, the first detection constraint record value is eliminated. The first detection constraint record value set is accessed in sequence to obtain the first detection constraint true value, and added to the multi-source detection indicator true value. If the first discreteness exceeds the preset discreteness threshold, it indicates that the record value has a large relative deviation and should be eliminated to ensure the accuracy and consistency of the data. The remaining first detection constraint record value sets are traversed and analyzed to finally obtain a reliable first detection constraint true value and add it to the multi-source detection indicator true value.

[0046] By extracting the first detection constraint record value set from the multi-source detection information record value set, filtering out the reference detection constraint record values ​​whose record value span is less than the threshold, and calculating the discreteness of the record value, reliable detection data is evaluated and filtered out. After eliminating the record values ​​that exceed the discreteness threshold, the first detection constraint true value obtained can more accurately reflect the actual operating state of the motor.

[0047] Furthermore, the present application also includes: Obtain a first detection indicator true value according to the multi-source detection indicator true value; obtain first detection information according to the multi-source detection information, wherein the first detection indicator true value corresponds to the first detection information; compare the first detection indicator true value with the first detection information to obtain a first deviation modulus value and a first deviation direction; construct a first deviation vector according to the first deviation modulus value and the first deviation direction, and add it to the multi-source detection deviation matrix.

[0048] Specifically, the true value of the first detection indicator is obtained according to the true value of the multi-source detection indicator. First, a specific first detection indicator true value is extracted from the multi-source detection indicator true value set. The first detection indicator true value represents the ideal state of the first detection indicator under normal working conditions, such as the standard values ​​of specific parameters such as motor operating current, vibration spectrum, and operating temperature. By analyzing and processing multi-source detection data, the true value of the indicator can be accurately determined, providing a benchmark for subsequent deviation analysis.

[0049] Then, first detection information is obtained according to the multi-source detection information, wherein the first detection index true value corresponds to the first detection information. The first detection information is obtained by real-time acquisition and detection of the current operation data of the motor by multi-source sensors. The first detection information includes parameters such as the real-time working current, vibration spectrum, and working temperature of the motor, and each item of detection information corresponds to a specific detection index true value.

[0050] Next, the true value of the first detection index is compared with the first detection information to obtain the first deviation modulus and the first deviation direction. The true value of the first detection index is compared with the first detection information collected in real time one by one to calculate the degree of deviation of each parameter. The deviation modulus indicates the absolute difference between the current detection value and the true value, while the deviation direction indicates the direction of the deviation, whether it is a positive deviation or a negative deviation, which can quantify the difference between the current state of the motor and the ideal state.

[0051] Then, a first deviation vector is constructed according to the first deviation modulus and the first deviation direction, and added to the multi-source detection deviation matrix. The deviation modulus and direction of each detection parameter are combined into a deviation vector, which contains all deviation information of the current state relative to the reference state. By constructing the deviation vector, the abnormality of the motor operation state can be described. Then, the deviation vector is added to the multi-source detection deviation matrix to form a comprehensive deviation data set. The multi-source detection deviation matrix is ​​a multidimensional data structure that contains the deviation of each detection parameter at different time points. By analyzing the matrix, the operating state of the motor can be fully understood and potential faults and anomalies can be identified. The deviation matrix not only helps to monitor the health status of the motor in real time, but can also be used for historical data analysis to discover long-term trends and hidden dangers.

[0052] By extracting the first detection indicator true value from the multi-source detection indicator true value and comparing it with the first detection information collected in real time, the first deviation modulus and direction can be calculated. The constructed first deviation vector can fully describe the abnormal situation of the current state and add it to the multi-source detection deviation matrix, which helps to systematize and quantify the fault detection and prediction of the motor. It not only improves the accuracy of fault detection, but also provides important reference data for the preventive maintenance of the motor.

[0053] Furthermore, the present application also includes: According to the fault association relationship, the execution times of the multi-source detection information are collected to obtain the motor fault prediction weight; according to the multi-source detection deviation matrix, the motor fault prediction weight coefficient is obtained; according to the motor fault prediction coefficient and the motor fault prediction weight coefficient, the motor fault prediction result is obtained.

[0054] Specifically, according to the fault association, the execution times of multi-source detection information are collected to obtain the motor fault prediction weight. First, based on historical data and motor fault association, the execution times of multi-source detection information are analyzed and counted. For example, for each specific fault type, the frequency of abnormalities of related detection indicators (such as current, vibration, temperature, etc.) is calculated. The execution times reflect the strength of association between each detection indicator and a specific fault, and are used to determine the fault prediction weight. The higher the weight, the greater the importance of the detection indicator in predicting a specific fault.

[0055] Next, the motor fault prediction weight coefficient is obtained according to the multi-source detection deviation matrix. Using the data in the multi-source detection deviation matrix, the deviation value of each detection parameter is calculated, and compared with the historical fault data to determine the fault prediction weight coefficient of each detection parameter. The weight coefficient can be calculated using machine learning algorithms (such as linear regression, decision trees, etc.) or statistical analysis methods to evaluate the influence of each parameter in fault prediction. The size of the weight coefficient reflects the contribution of each detection parameter to fault prediction.

[0056] Next, the motor fault prediction result is obtained according to the motor fault prediction coefficient and the motor fault prediction weight coefficient. The motor fault prediction coefficient (i.e., the weight of each detection parameter) is combined with the fault prediction weight coefficient to calculate the comprehensive fault prediction result. The overall fault prediction result of the motor is obtained, which indicates the possibility of the motor failing in the current state.

[0057] The motor fault prediction weight is determined by collecting the execution times of multi-source detection information through fault association relationships; the fault prediction weight coefficient is calculated using the multi-source detection deviation matrix; and a comprehensive fault prediction result is obtained by combining the fault prediction coefficient and the weight coefficient. This not only improves the accuracy of fault prediction, but also provides a scientific basis for preventive maintenance of motors, helps to timely discover and handle potential faults, and ensures the safe and stable operation of motors.

[0058] Furthermore, the present application also includes: If the motor fault prediction result meets the motor fault threshold, the motor fault operator is obtained; based on the motor fault operator, the fault association relationship is extracted, and sensor repair is performed according to the multi-source detection information; based on the motor fault operator, the fault association relationship is extracted, and detection and repair are performed according to the fault cause.

[0059] Specifically, if the motor fault prediction result meets the motor fault threshold, a motor fault operator is obtained. First, when the motor fault prediction result exceeds the preset fault threshold, the motor's operating state has shown significant abnormalities and has a high risk of failure. In this case, a motor fault operator needs to be generated. The motor fault operator is an operator used to describe and handle faults, which contains detailed fault types, prediction probabilities, and related detection parameters. The nature and scope of the fault are clarified through the fault operator to provide guidance for subsequent repairs and processing.

[0060] Next, based on the motor fault operator, the fault association is extracted, and sensor repair is performed based on the multi-source detection information. The fault information contained in the motor fault operator is used to extract relevant associated data from the fault association. The associated data includes historical data and repair measures related to the fault type and detection parameters. Sensor repair is performed, that is, the sensors and detection equipment involved in the fault are calibrated, adjusted or replaced to ensure the accuracy and reliability of the detection data. For example, if the temperature sensor of the motor fails, the sensor can be recalibrated or replaced through sensor repair to restore its normal function.

[0061] Next, based on the motor fault operator, the fault association is extracted, and detection and repair are performed based on the cause of the fault. After the sensor repair is completed, the actual motor operating status still needs to be detected and repaired. Using the fault cause information in the motor fault operator, the corresponding repair measures and operation guidelines are extracted from the fault association. Measures may include adjusting the operating parameters of the motor, replacing damaged parts, improving the cooling system, etc. By performing repair operations, the root cause of the fault can be directly resolved and the normal operating state of the motor can be restored. For example, if the motor overheats, it may be necessary to check and improve the cooling system, or replace problematic parts.

[0062] Through comprehensive sensor repair and detection repair, motor fault problems can be fully processed and solved. Sensor repair ensures the accuracy of the detection equipment and avoids data misjudgment due to sensor failure; detection repair directly adjusts and repairs the actual operating status of the motor to ensure that the motor can operate in the best condition. By forming a complete fault handling and repair process, the operating reliability and safety of the motor can be maximized.

[0063] By generating motor fault operators, describing the fault type and detection parameters in detail, and then performing sensor repair and detection repair respectively, sensor repair ensures the accuracy of the detection equipment, while detection repair directly solves the root cause of the fault, comprehensively solves the motor fault problem, ensures the safe and stable operation of the motor, and provides maintenance personnel with clear operation guidelines and repair measures.

[0064] In summary, the special release motor fault detection method provided by the present application has the following technical effects: By acquiring multi-source detection information of a special release motor in a health detection device, wherein the multi-source detection information has a fault correlation relationship with the fault cause; acquiring the motor control state and motor environment information of the special release motor; determining the motor reference state according to the fault cause, the motor control state and the motor environment information, and generating a multi-source detection index true value; performing deviation analysis on the multi-source detection index true value and the multi-source detection information, and establishing a multi-source detection deviation matrix; performing an abnormality analysis on the multi-source detection deviation matrix and the fault correlation relationship, and obtaining a motor fault prediction result; if a motor fault operator is obtained based on the motor fault prediction result, the fault correlation relationship is extracted for maintenance, thereby achieving the technical goal of comprehensively analyzing multiple detection parameters and using an intelligent algorithm to predict motor faults, and achieving the technical effect of improving fault detection accuracy and maintenance efficiency.

[0065] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A special release motor fault detection method, characterized in that: include: Acquire multi-source detection information of a special release motor in a health detection device, wherein the multi-source detection information has a fault correlation relationship with a fault cause; Acquiring the motor control state and motor environment information of the special release motor; Determine the motor reference state according to the fault cause, motor control state and motor environment information, and generate a true value of a multi-source detection index; Performing deviation analysis based on the true value of the multi-source detection indicator and the multi-source detection information to establish a multi-source detection deviation matrix; Performing an abnormal analysis on the multi-source detection deviation matrix and the fault association relationship to obtain a motor fault prediction result; If a motor fault operator is obtained based on the motor fault prediction result, the fault association relationship is extracted for maintenance.

2. A special release motor fault detection method as claimed in claim 1, characterized in that: The multi-source detection information includes motor operating current, motor vibration spectrum, motor operating temperature, motor operating sound, and motor insulation resistance.

3. A special release motor fault detection method as claimed in claim 1, characterized in that: The generating of the true value of the multi-source detection indicator includes: Based on the fault cause, motor control state and motor environment information as constraints, a multi-step association search is performed to obtain a multi-source detection information record value set; The multi-source detection information record value set is traversed to perform confidence analysis to generate the true value of the multi-source detection indicator.

4. A special release motor fault detection method as claimed in claim 3, characterized in that: The obtaining of a multi-source detection information record value set includes: Taking the fault cause as a search constraint and multi-source sensor detection as a search target, a search is performed to obtain a first step sensor record value set; Taking the motor control state as a search constraint and the multi-source sensor detection as a search target, a search is performed to obtain a second step sensor record value set; Taking the motor environment information as a search constraint and the multi-source sensor detection as a search target, a search is performed to obtain a sensor record value set of the third step; The first step sensing record value set, the second step sensing record value set and the third step sensing record value set are processed based on the multi-source sensing detection as the multi-source detection information record value set with connection constraints.

5. A special release motor fault detection method as claimed in claim 3, characterized in that: The generating the true value of the multi-source detection indicator comprises: Extracting a first detection constraint record value set of the multi-source detection information record value set; Extracting, based on the first detection constraint record value set, a reference detection constraint record value whose record value span with the first detection constraint record value is less than a record value span threshold; generating a first record dispersion according to a mean value of the distance between the first detection constraint record value and the reference detection constraint record value; Traversing the first detection constraint record value set to obtain a record discreteness mean, and generating a first discreteness according to a ratio of the first record discreteness to the record discreteness mean; If the first discreteness exceeds the discreteness threshold, the first detection constraint record value is eliminated, the first detection constraint record value set is traversed, the first detection constraint true value is obtained, and added to the multi-source detection indicator true value.

6. A special release motor fault detection method as claimed in claim 1, characterized in that: The step of establishing a multi-source detection deviation matrix includes: Acquire a first detection indicator true value according to the multi-source detection indicator true value; Obtaining first detection information according to the multi-source detection information, wherein the first detection indicator true value corresponds to the first detection information; Comparing the first detection indicator true value with the first detection information to obtain a first deviation modulus value and a first deviation direction; A first deviation vector is constructed according to the first deviation modulus and the first deviation direction, and is added to the multi-source detection deviation matrix.

7. A special release motor fault detection method as claimed in claim 1, characterized in that: The obtaining of the motor fault prediction result comprises: According to the fault association relationship, the number of executions of the multi-source detection information is collected to obtain a motor fault prediction weight; Obtaining a motor fault prediction weight coefficient according to the multi-source detection deviation matrix; The motor fault prediction result is obtained according to the motor fault prediction coefficient and the motor fault prediction weight coefficient.

8. A special release motor fault detection method as claimed in claim 1, characterized in that: The extracting the fault association relationship for maintenance includes: If the motor fault prediction result meets the motor fault threshold, obtaining the motor fault operator; Extracting the fault association relationship according to the motor fault operator, and performing sensor repair according to the multi-source detection information; The fault association relationship is extracted according to the motor fault operator, and detection and repair are performed according to the fault cause.

Citation Information

Patent Citations

  • Equipment fault early warning method and device

    CN112130487A

  • Fault analysis method and device for power supply equipment

    CN113627629A

  • Equipment fault detection and analysis method and device based on artificial intelligence

    CN117909784A

  • Circuit breaker fault detection method and system based on multi-source sensor

    CN118091401A

Cited By

  • Motor operation abnormity monitoring method and system for electro-tricycle

    CN119511077A

  • Method and system for quickly checking compatibility big data of electrical equipment

    CN119513143A

  • Motor abnormal state early warning system and method

    CN120254610A

  • Motor abnormal state early warning system and method

    CN120254610B

  • Method and system for detecting open-circuit fault and long-time working fault of motor

    CN120891374A