Generator fault diagnosis method based on carbon brush wear signal

Abstract

The invention provides a generator fault diagnosis method based on a carbon brush wear signal, and belongs to the technical field of motor fault diagnosis, and the method comprises the steps: 1, deploying a preset sensor group at a key position of a carbon brush, collecting the related data of the carbon brush, and installing a preset monitoring system in a power collection chamber of a generator, data related to operation of the carbon powder and the generator are collected; 2, processing the collected related data of the carbon brush, and extracting key features of a preset type; 3, determining the wear coefficient of the carbon brush based on the extracted key features; 4, determining the risk coefficient of the generator based on the wear coefficient of the carbon brush, the carbon powder and the related data of the generator operation; and 5, performing multi-level early warning based on the risk coefficient of the generator, and performing corresponding early warning processing measures. The accuracy of fault diagnosis and the timeliness of early warning are improved, and potential faults are effectively prevented.

Description

Technical Field

[0001] The invention relates to the technical field of motor fault diagnosis, and in particular to a generator fault diagnosis method based on carbon brush wear signals. Background Art

[0002] Currently, the carbon brush wear monitoring technology during generator operation is relatively limited and cannot fully and accurately reflect the carbon brush wear condition and its impact on the safe operation of the generator.

[0003] Some existing diagnostic solutions usually rely on single carbon brush wear data and fail to combine multiple related parameters for comprehensive analysis, resulting in insufficient diagnostic accuracy and difficulty in timely warning.

[0004] Therefore, the present invention provides a generator fault diagnosis method based on carbon brush wear signals. Summary of the invention

[0005] The present invention provides a generator fault diagnosis method based on carbon brush wear signals, which is used to determine the risk factor of the generator and realize multi-level early warning and multi-dimensional data comprehensive analysis by combining carbon brush wear signals, carbon powder and generator operation data, thereby improving the accuracy of fault diagnosis and the timeliness of early warning, effectively preventing potential faults, and solving the defects of low diagnostic accuracy and delayed early warning in the prior art.

[0006] The present invention provides a generator fault diagnosis method based on carbon brush wear signals, comprising:

[0007] Step 1: Deploy the preset sensor group at the key position of the carbon brush and collect relevant data of the carbon brush. At the same time, install the preset monitoring system in the generator collector and collect data related to carbon powder and generator operation;

[0008] Step 2: Process the collected data related to the carbon brush and extract the key features of the preset type;

[0009] Step 3: Determine the wear coefficient of the carbon brush based on the extracted key features;

[0010] Step 4: Determine the risk factor of the generator based on the wear coefficient of the carbon brush, the carbon powder and the data related to the operation of the generator;

[0011] Step 5: Conduct multi-level warnings based on the risk factor of the generator and take corresponding warning treatment measures.

[0012] The present invention provides a generator fault diagnosis method based on carbon brush wear signals, wherein a sensor group is preset, including: a pressure sensor, a temperature sensor, a vibration sensor, a current sensor and a voltage sensor;

[0013] Key positions include: carbon brush contact surface, generator collector ring, carbon brush conductive arm, brush holder, commutator, carbon brush wire connection terminal, near the generator bearing and collector ring cooling system.

[0014] The present invention provides a generator fault diagnosis method based on carbon brush wear signals, wherein a preset monitoring system is installed in the generator collector chamber, and data related to carbon powder and generator operation are collected, including:

[0015] A particle sensor is installed in the generator collector chamber to monitor the concentration and accumulation of carbon powder. At the same time, a dynamic electrical isolation system is installed in the generator collector chamber to monitor the carbon powder concentration and voltage changes in real time.

[0016] The present invention provides a generator fault diagnosis method based on carbon brush wear signals, which processes the collected carbon brush related data and extracts key features of preset types, including:

[0017] Filter, normalize and denoise the collected carbon brush data to eliminate environmental interference and measurement errors;

[0018] The preprocessed data is segmented according to a preset time window by a preset algorithm, and a plurality of key features of a first preset type are extracted therefrom;

[0019] A plurality of key features of a second preset type are extracted from the plurality of key features of the first preset type.

[0020] The present invention provides a generator fault diagnosis method based on a carbon brush wear signal, wherein a first preset type includes: a pressure characteristic, a temperature characteristic, a wear characteristic and an electrical characteristic.

[0021] The present invention provides a generator fault diagnosis method based on carbon brush wear signals, which determines the wear coefficient of the carbon brush based on the extracted key features, including:

[0022] A quantification method for each type of key feature is obtained based on a preset feature type-quantification method, and the extracted key features are quantified to obtain a number of original feature values;

[0023] Standardize all original eigenvalues ​​to obtain several eigenvalues;

[0024] Determine the wear coefficient of the carbon brush based on all characteristic values:

[0025]

[0026] Where W is the wear coefficient of the carbon brush, X i is the eigenvalue of the i-th key feature, α iis the weight coefficient of the i-th key feature, n is the number of key features, X j is the eigenvalue of the jth key feature, β ij is the interaction weight coefficient between the eigenvalue of the i-th key feature and the eigenvalue of the j-th key feature.

[0027] The present invention provides a generator fault diagnosis method based on a carbon brush wear signal, which determines the risk factor of the generator based on the wear coefficient of the carbon brush, carbon powder and data related to the operation of the generator, including:

[0028]

[0029] Among them, R is the risk factor of the generator, W is the carbon brush wear coefficient, and W ref is the preset standard reference value of the carbon brush wear coefficient, C is the carbon powder concentration, P is the operating load of the generator, and C ref Preset standard reference value for toner density, P ref The preset standard reference value is for the generator operating load, Δt is the continuous operating time of the generator, T is the generator operating temperature, T ref Preset standard reference values ​​for generator operating temperature, is the risk factor adjustment value, τ is the time attenuation coefficient, is the preset weight of temperature change, and ρ is the temperature adjustment coefficient.

[0030] The present invention provides a generator fault diagnosis method based on carbon brush wear signals, which performs multi-level early warning based on risk factors and performs corresponding early warning processing measures, including:

[0031] Match the risk factor with the preset risk level threshold, and then determine the warning level corresponding to the risk factor;

[0032] Generate corresponding warning signals based on the warning level and the preset warning level-signal data table;

[0033] Based on the early warning signal, corresponding processing suggestions are obtained in the preset database, and then processing suggestions are generated.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] By combining the carbon brush wear signal, carbon powder and generator operation data, the risk factor of the generator is determined and multi-level early warning is achieved. Comprehensive analysis of multi-dimensional data improves the accuracy of fault diagnosis and the timeliness of early warning, effectively prevents potential faults, and solves the defects of low diagnostic accuracy and delayed early warning in existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a flow chart of a generator fault diagnosis method based on carbon brush wear signals provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] Example 1

[0040] The embodiment of the present invention provides a generator fault diagnosis method based on carbon brush wear signal, such as Figure 1 As shown, including:

[0041] Step 1: Deploy the preset sensor group at the key position of the carbon brush and collect relevant data of the carbon brush. At the same time, install the preset monitoring system in the generator collector and collect data related to carbon powder and generator operation;

[0042] Step 2: Process the collected data related to the carbon brush and extract the key features of the preset type;

[0043] Step 3: Determine the wear coefficient of the carbon brush based on the extracted key features;

[0044] Step 4: Determine the risk factor of the generator based on the wear coefficient of the carbon brush, the carbon powder and the data related to the operation of the generator;

[0045] Step 5: Conduct multi-level warnings based on the risk factor of the generator and take corresponding warning treatment measures.

[0046] In this embodiment, the carbon brush related data refers to the carbon brush working status information collected by the sensor, including temperature, vibration, wear rate, current, voltage and other data. These data reflect the operating status of the carbon brush and help to determine its wear degree and failure risk. For example, the sensor detects that the temperature of the carbon brush is abnormally high. This data indicates that there may be wear or overheating problems, which requires further inspection.

[0047] In this embodiment, the generator collector chamber refers to an area inside the generator that is specifically used to collect and transmit current, including components such as collector rings and carbon brushes. This area is usually a high-temperature, high-friction environment, and is crucial for monitoring the status of carbon brushes and current collection conditions. For example, monitoring equipment is installed in the generator collector chamber to capture information such as the temperature near the carbon brushes and the amount of carbon powder generated by friction;

[0048] In this embodiment, the preset monitoring system refers to monitoring equipment installed in the collector chamber, including temperature sensors, pressure sensors, image monitoring equipment, etc., which can collect and monitor the status data of carbon brushes and other key components in real time. For example, temperature and vibration sensors are arranged in the collector chamber to continuously monitor the working status of carbon brushes and other components to ensure timely warning;

[0049] In this embodiment, the data related to the carbon powder and the operation of the generator include parameters such as the concentration of carbon powder generated by the friction of the carbon brush, voltage, current, and rotation speed. These data can be combined to reflect the workload and wear of the carbon brush. For example, an increase in the amount of carbon powder and a high operating temperature may indicate that the carbon brush is aggravated in wear and maintenance is required.

[0050] In this embodiment, the key features of the type refer to key parameters extracted from the carbon brush operation data, such as temperature, vibration amplitude, wear rate, etc., which can reflect the state of the carbon brush and be used to judge its wear condition. For example, the temperature exceeds the preset value and the vibration amplitude increases significantly. These are key features of abnormal wear.

[0051] In this embodiment, the wear coefficient of the carbon brush is a value calculated based on the extracted key features, which is used to measure the degree of wear of the carbon brush and reflect its service life and current health status. Example: For example, when the wear coefficient exceeds the critical value, it means that the carbon brush may be close to the end of its service life and needs to be replaced in time;

[0052] In this embodiment, the risk factor of the generator is obtained based on a comprehensive analysis of the wear coefficient of the carbon brush, the amount of carbon powder and operating data, and is used to quantify the possibility of failure of the generator. If the carbon powder concentration is high and the wear coefficient is high, the risk factor will also increase, indicating that the generator has a higher risk of failure.

[0053] In this embodiment, multi-level warning refers to different warning levels according to the risk coefficient, such as low, medium and high risks, so as to take corresponding preventive measures. For example, a reminder notification is issued when the risk is low, an inspection is required when the risk is medium, and an emergency warning is issued and shutdown maintenance is arranged when the risk is high.

[0054] In this embodiment, the early warning handling measures refer to the response operations taken at different early warning levels, such as adjusting operating parameters, arranging maintenance, replacing carbon brushes, etc., to reduce the risk of failure. For example, when a high-risk early warning is issued, an immediate shutdown inspection is arranged to avoid further wear or damage.

[0055] The beneficial effects of the above technical solution are: by combining the carbon brush wear signal, carbon powder and generator operation data, the risk factor of the generator is determined and multi-level early warning is achieved, and multi-dimensional data is comprehensively analyzed, which improves the accuracy of fault diagnosis and the timeliness of early warning, effectively prevents potential faults, and solves the defects of low diagnostic accuracy and delayed early warning in the existing technology.

[0056] Example 2

[0057] The embodiment of the present invention provides a generator fault diagnosis method based on carbon brush wear signals, wherein a sensor group is preset, including: a pressure sensor, a temperature sensor, a vibration sensor, a current sensor, and a voltage sensor;

[0058] Key positions include: carbon brush contact surface, generator collector ring, carbon brush conductive arm, brush holder, commutator, carbon brush wire connection terminal, near the generator bearing and collector ring cooling system.

[0059] In this embodiment, the pressure sensor is installed near the brush holder and the generator bearing to detect the pressure on the contact surface between the carbon brush and the collector ring to monitor the stability of the contact state of the carbon brush. If the pressure is abnormal, it may mean that the carbon brush is worn, the contact is poor, or the brush holder is abnormal.

[0060] In this embodiment, the temperature sensor is usually installed in key locations such as the carbon brush contact surface, the generator collector ring, and the collector ring cooling system to monitor the temperature changes at these locations. If the temperature rises quickly or exceeds the set threshold, it may indicate that the carbon brush is worn, the heat dissipation is poor, or the cooling system is abnormal;

[0061] In this embodiment, the vibration sensor is usually installed near the generator bearing and the collector ring area, and the monitoring data: detecting the vibration intensity and frequency of the key components of the generator. Abnormal vibration may indicate bearing wear, rotor imbalance, abnormal carbon brush beating, etc.;

[0062] In this embodiment, the current sensor is installed at the carbon brush conductive arm and the carbon brush wire connection terminal to detect the current passing through the carbon brush. Abnormal current fluctuations may indicate poor contact of the carbon brush, aging of the conductive arm, or unstable wire connection;

[0063] In this embodiment, voltage sensors are installed at the commutator and the collector ring to monitor the voltage changes of the generator. Abnormal voltage may indicate problems such as poor contact of carbon brushes, commutator failure or power transmission interruption.

[0064] The beneficial effects of the above technical solution are: by deploying a variety of sensors on the carbon brushes and key parts of the generator, real-time monitoring of parameters such as pressure, temperature, vibration, current and voltage can be carried out, the wear signal can be fully captured and the source of the fault can be accurately located, providing multi-dimensional data support for generator fault diagnosis, thereby improving the accuracy and reliability of diagnosis.

[0065] Example 3

[0066] The embodiment of the present invention provides a generator fault diagnosis method based on carbon brush wear signals, wherein a preset monitoring system is installed in the generator collector chamber, and data related to carbon powder and generator operation are collected, including:

[0067] A particle sensor is installed in the generator collector chamber to monitor the concentration and accumulation of carbon powder. At the same time, a dynamic electrical isolation system is installed in the generator collector chamber to monitor the carbon powder concentration and voltage changes in real time.

[0068] In this embodiment, the particle sensor is a sensor for monitoring the concentration of tiny particles (such as carbon powder) in the air. In the generator collector chamber, the particle sensor can detect and analyze the concentration and accumulation of carbon powder in real time to reflect the degree of wear of the carbon brush. For example, the particle sensor detects a rapid increase in the concentration of carbon powder, which may indicate that the carbon brush is being worn more and more, and timely maintenance or replacement of the carbon brush is required.

[0069] In this embodiment, the dynamic electrical isolation system is a system that can automatically adjust the isolation state, and is intended to monitor and protect the stable operation of electrical equipment. By real-time monitoring of carbon powder concentration and voltage changes, the system can isolate the circuit when the carbon powder concentration is too high or the voltage fluctuates abnormally to prevent short circuits or other electrical faults. For example, when the dynamic electrical isolation system detects that the carbon powder concentration is too high and the voltage fluctuates abnormally, the system automatically isolates the relevant circuits to avoid the risk of short circuits caused by carbon powder.

[0070] The beneficial effects of the above technical solution are: by installing a particle sensor and a dynamic electrical isolation system in the collector chamber, the carbon powder concentration and voltage changes can be monitored in real time, which can accurately reflect the wear status of the carbon brushes, avoid the interference of dust on the electrical system, and achieve more stable and effective fault diagnosis, greatly improving the safety of generator operation and the real-time nature of diagnosis.

[0071] Example 4

[0072] The embodiment of the present invention provides a generator fault diagnosis method based on carbon brush wear signals, which processes the collected carbon brush related data and extracts key features of preset types, including:

[0073] Filter, normalize and denoise the collected carbon brush data to eliminate environmental interference and measurement errors;

[0074] The preprocessed data is segmented according to a preset time window by a preset algorithm, and a plurality of key features of a first preset type are extracted therefrom;

[0075] A plurality of key features of a second preset type are extracted from the plurality of key features of the first preset type.

[0076] In this embodiment, the preset algorithm refers to an algorithm predefined during system design, which is used for data processing and analysis to extract features related to carbon brush wear, including a specific mathematical model or machine learning algorithm, such as Fourier transform, support vector machine (SVM) or neural network, to identify regularities or abnormal patterns in data. For example, the preset algorithm using Fourier transform can convert the carbon powder concentration signal from the time domain to the frequency domain, thereby making it easier to identify the periodic changes in carbon brush wear;

[0077] In this embodiment, the preset time window segmentation refers to dividing the continuous data stream into multiple small time periods at fixed time intervals for feature extraction. This segmentation helps to capture the changes in the carbon brush wear signal in different time periods. For example, the system segments the collected signal in units of 1 second, so that the changes in carbon powder concentration within each second can be analyzed, thereby capturing short-term abnormalities or sudden wear phenomena;

[0078] In this embodiment, several key features of the second preset type refer to specific features further screened out from the initially extracted key features of the first preset type, which are used for deeper analysis. They are data that have undergone secondary processing or aggregation and can more accurately reflect the wear condition of the carbon brushes. For example, the features of the first preset type may include the mean and variance of the carbon powder concentration, while the key features of the second preset type may be the trend or rate of change of these values ​​in different time periods, so as to more accurately predict the degree of wear and failure risk of the carbon brushes.

[0079] The beneficial effects of the above technical solution are: eliminating environmental interference and errors through filtering, normalization and denoising, and introducing time window segmentation and multi-level feature extraction strategies to deeply explore the key information in the carbon brush wear signal, significantly improving the accuracy of feature extraction, and providing more comprehensive and reliable data support for generator fault diagnosis.

[0080] Example 5

[0081] An embodiment of the present invention provides a generator fault diagnosis method based on a carbon brush wear signal, wherein the first preset type includes: pressure characteristics, temperature characteristics, wear characteristics, and electrical characteristics.

[0082] In this embodiment, the pressure and temperature characteristics include: pressure mean, temperature change rate, peak temperature, etc.;

[0083] In this embodiment, the wear characteristics include: a wear rate, a rate of change in the thickness of the carbon brush calculated based on the sensor signal;

[0084] In this embodiment, the electrical characteristics include: harmonic components and fluctuation amplitudes of voltage and current;

[0085] The beneficial effects of the above technical solution are: by introducing multi-dimensional features such as pressure, temperature, wear and electrical, the carbon brush status can be comprehensively evaluated. Compared with the diagnosis method of a single feature, the accuracy and reliability of the diagnosis are greatly improved. Through the fusion of multiple features, potential faults can be more effectively identified, and more timely fault warnings and maintenance guidance can be achieved.

[0086] Example 6

[0087] The embodiment of the present invention provides a generator fault diagnosis method based on a carbon brush wear signal, which determines the wear coefficient of the carbon brush based on the extracted key features, including:

[0088] A quantification method for each type of key feature is obtained based on a preset feature type-quantification method, and the extracted key features are quantified to obtain a number of original feature values;

[0089] Standardize all original eigenvalues ​​to obtain several eigenvalues;

[0090] Determine the wear coefficient of the carbon brush based on all characteristic values:

[0091]

[0092] Where W is the wear coefficient of the carbon brush, X i is the eigenvalue of the i-th key feature, α i is the weight coefficient of the i-th key feature, n is the number of key features, X j is the eigenvalue of the jth key feature, β ij is the interaction weight coefficient between the eigenvalue of the i-th key feature and the eigenvalue of the j-th key feature.

[0093] In this embodiment, the preset feature type-quantization method refers to pre-defining a quantization method according to the characteristics of different features, and converting these features into numerical forms. This method can help to uniformly process various types of feature data for subsequent analysis. Commonly used quantization methods include scaling, normalization, and standardization, which depend on the attributes of the feature and the required quantization accuracy. For example, for the vibration characteristics of the carbon brush, the mean is selected as the quantization method to represent the overall vibration intensity. For temperature characteristics, the maximum value is quantified to capture possible overheating of the carbon brush during use. In this way, each type of feature data can be expressed in numerical form after quantization, which is convenient for analysis;

[0094] In this embodiment, the interaction weight coefficient refers to the coefficient used to describe the interaction between different features when calculating the carbon brush wear coefficient. The interaction weight coefficient helps the model capture the complex relationship between different features to more accurately reflect the wear of the carbon brush. This coefficient is usually used in multivariate analysis to express the strength of association between two types of characteristic values, and has a direct impact on the calculation of the wear coefficient. Assuming that the interaction between the temperature characteristics and current characteristics of the carbon brush is important, the interaction weight coefficient between the two can be set to a higher value. This setting shows that the combined effect of temperature and current has a greater impact on the wear of the carbon brush. If the interaction between the temperature and vibration characteristics is not significant, their interaction weight coefficients can be set smaller. This approach enables the model to effectively integrate the independent and interactive effects of the features, thereby more accurately calculating the carbon brush wear coefficient.

[0095] The beneficial effects of the above technical solution are: by quantifying key features and performing standardization to determine the wear coefficient of the carbon brush, introducing the interaction weight coefficient between features, improving the accuracy of diagnosis, utilizing the complexity of multiple features and their mutual influence, more accurately reflecting the wear state of the carbon brush, and achieving more efficient fault prediction and maintenance decisions.

[0096] Example 7

[0097] The embodiment of the present invention provides a generator fault diagnosis method based on a carbon brush wear signal, which determines the risk factor of the generator based on the wear coefficient of the carbon brush, carbon powder and data related to the operation of the generator, including:

[0098]

[0099] Among them, R is the risk factor of the generator, W is the carbon brush wear coefficient, and W ref is the preset standard reference value of the carbon brush wear coefficient, C is the carbon powder concentration, P is the operating load of the generator, and C ref Preset standard reference value for toner density, P ref The preset standard reference value is for the generator operating load, Δt is the continuous operating time of the generator, T is the generator operating temperature, T ref is the preset standard reference value for the generator operating temperature, φ1 is the risk factor adjustment value, τ is the time attenuation coefficient, φ2 is the preset weight of the temperature change, and ρ is the temperature adjustment coefficient.

[0100] In this embodiment, the preset standard reference value of the carbon brush wear coefficient refers to a reference value set according to the normal wear state of the carbon brush, which is used to determine whether the actual wear coefficient exceeds a reasonable range. By comparing the actual wear coefficient with the reference value, the system can identify whether the wear of the carbon brush is abnormal. Assume that the preset standard reference value of the carbon brush wear coefficient is 0.8. When the actual wear coefficient is greater than 0.8, the system will consider that the carbon brush is wearing too fast and there may be a risk of failure, which requires further inspection;

[0101] In this embodiment, the preset standard reference value of carbon powder concentration refers to a reasonable range of carbon powder concentration when the generator is operating normally, which is usually set based on historical data or experimental data to measure whether the carbon powder concentration generated by carbon brush wear is within a safe range. For example, the standard reference value of carbon powder concentration can be set to 100 ppm (parts per million). If the actual concentration exceeds this value, it indicates that the carbon brush wear may be abnormal or the carbon brush needs to be replaced;

[0102] In this embodiment, the preset standard reference value of the generator operating load refers to the load level of the generator under normal working conditions. Setting this value helps to determine whether the actual load is within the normal range, thereby evaluating the health status of the generator. For example, the normal load of the generator can be set to 70% of the rated load. When the actual load exceeds 70%, the generator may be running under high load, which will aggravate wear and tear, and its risk factor needs to be paid attention to;

[0103] In this embodiment, the preset standard reference value of the generator operating temperature refers to the standard temperature range of the generator during normal operation, so as to timely detect temperature abnormalities and prevent failures caused by overheating. For example, the normal operating temperature reference value of the generator is set to 80°C. When the temperature exceeds this value, it indicates that the device may be in an overheated state, and it is necessary to increase cooling or reduce the load;

[0104] In this embodiment, the risk factor adjustment value is used to increase or decrease the risk under specific conditions when calculating the risk factor of the generator. For example, in a harsh environment, such as a high temperature or high humidity environment, the risk factor adjustment value can be set to 0.2 to appropriately increase the overall risk factor to reflect the additional risk brought by the special environment;

[0105] In this embodiment, the temperature adjustment coefficient is used to perform weighted control on the influence of temperature on the risk factor to adapt to the different influences of temperature changes on the health of the generator under different working conditions. For example, when the generator is operating at high load, the temperature adjustment coefficient can be set to 1.5 to increase the influence of temperature changes on the risk factor, while at low load it can be reduced to 1.0 to reduce the influence of temperature changes on the risk.

[0106] The beneficial effects of the above technical solution are: through multiple operating data such as carbon brush wear coefficient, carbon powder concentration, generator operating load and temperature, combined with preset reference values ​​and weight factors, the risk factor of the generator is calculated, so as to diagnose faults more accurately, and integrate multiple real-time data and adjustment coefficients to achieve more accurate fault warning and maintenance guidance, thereby improving the reliability of diagnosis and warning effect.

[0107] Example 8

[0108] The embodiment of the present invention provides a generator fault diagnosis method based on carbon brush wear signals, which performs multi-level warnings based on risk factors and performs corresponding warning processing measures, including:

[0109] Match the risk factor with the preset risk level threshold, and then determine the warning level corresponding to the risk factor;

[0110] Generate corresponding warning signals based on the warning level and the preset warning level-signal data table;

[0111] Based on the early warning signal, corresponding processing suggestions are obtained in the preset database, and then processing suggestions are generated.

[0112] In this embodiment, the preset risk level threshold refers to one or more numerical intervals set according to different risk levels, which are used to determine the risk level to which the risk coefficient belongs. These thresholds help the system classify potential faults caused by carbon brush wear, so as to facilitate the adoption of different degrees of countermeasures. It is assumed that there are three preset risk level thresholds: low risk (0-0.4), medium risk (0.4-0.7) and high risk (0.7-1.0). When the risk coefficient is 0.5, the system classifies it as a medium risk level.

[0113] In this embodiment, the preset warning level-signal data table is a data table that predefines the warning signal type corresponding to each warning level (such as low, medium, and high risk). Each warning signal will indicate different levels of fault severity, so that the system can generate different levels of alarm notifications or display signals. The table may set low risk to correspond to a green signal, medium risk to correspond to a yellow signal, and high risk to correspond to a red signal. If the system detects a medium risk level, a yellow signal will be displayed to prompt relevant personnel to pay attention but no emergency treatment is required.

[0114] In this embodiment, the warning levels include first, second, third and fourth warnings. First warning (low risk, normal operation): When the risk factor is within the safe range, it means that the carbon brush wear, carbon powder accumulation and the operating status of the generator are all in normal state and will not pose a direct threat to the system. No measures are required at this time, but long-term monitoring can be performed by recording status information. Second warning (medium risk, pay attention to monitoring): When the risk factor exceeds a certain threshold but has not yet reached a dangerous level, it indicates that there is a certain wear or potential problem with the carbon brush or generator components. At this time, it is necessary to strengthen monitoring, and it is recommended to check the relevant components regularly and perform maintenance or replacement when necessary. Third warning (high risk, timely maintenance): When the risk factor is close to the critical value, it indicates that the carbon brush is seriously worn, the carbon powder accumulates too much, or the operating status of the generator is significantly abnormal, which may cause equipment damage or system shutdown. Maintenance work should be arranged in time, and carbon brush replacement or other equipment maintenance should be performed when necessary to avoid further deterioration of potential faults. Fourth warning (serious risk, immediate shutdown): When the risk factor exceeds the dangerous threshold, it means that the equipment has a serious failure risk, which may cause the generator to shut down or be damaged at any time. At this time, emergency shutdown measures should be taken immediately to prevent the fault from expanding, and emergency repairs and comprehensive inspections should be organized immediately to ensure that the system returns to normal operation.

[0115] In this embodiment, the preset database is a data set that stores various treatment suggestions and warning level information. The database stores corresponding treatment measures according to different warning levels, providing specific response strategies for fault diagnosis. For example, the database may contain the following treatment suggestions: for low risk, only routine inspection is required; for medium risk, regular monitoring or replacement of carbon brushes is required; for high risk, immediate shutdown and maintenance is required to avoid damage.

[0116] The beneficial effects of the above technical solution are as follows: by converting the carbon brush wear signal into a risk coefficient and matching it with the preset risk level threshold, a multi-level fault warning is achieved, the corresponding warning signal is generated through the warning level, and specific processing suggestions are obtained based on the database, which effectively improves the accuracy and timeliness of generator fault diagnosis, reduces the error caused by human intervention, and thus improves the safety and reliability of the equipment.

[0117] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0118] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A generator fault diagnosis method based on carbon brush wear signal, characterized in that: include: Step 1: Deploy the preset sensor group at the key position of the carbon brush and collect relevant data of the carbon brush. At the same time, install the preset monitoring system in the generator collector and collect data related to carbon powder and generator operation; Step 2: Process the collected data related to the carbon brush and extract the key features of the preset type; Step 3: Determine the wear coefficient of the carbon brush based on the extracted key features; Step 4: Determine the risk factor of the generator based on the wear coefficient of the carbon brush, the carbon powder and the data related to the operation of the generator; Step 5: Conduct multi-level warnings based on the risk factor of the generator and take corresponding warning treatment measures.

2. A generator fault diagnosis method based on carbon brush wear signal according to claim 1, characterized in that: A preset sensor group includes: a pressure sensor, a temperature sensor, a vibration sensor, a current sensor, and a voltage sensor; Key positions include: carbon brush contact surface, generator collector ring, carbon brush conductive arm, brush holder, commutator, carbon brush wire connection terminal, near the generator bearing and collector ring cooling system.

3. A generator fault diagnosis method based on carbon brush wear signal according to claim 1, characterized in that: A preset monitoring system is installed in the generator collector room to collect data related to carbon powder and generator operation, including: A particle sensor is installed in the generator collector chamber to monitor the concentration and accumulation of carbon powder. At the same time, a dynamic electrical isolation system is installed in the generator collector chamber to monitor the carbon powder concentration and voltage changes in real time.

4. A generator fault diagnosis method based on carbon brush wear signal according to claim 1, characterized in that: The collected data about carbon brushes are processed and key features of preset types are extracted, including: Filter, normalize and denoise the collected carbon brush data to eliminate environmental interference and measurement errors; The preprocessed data is segmented according to a preset time window by a preset algorithm, and a plurality of key features of a first preset type are extracted therefrom; A plurality of key features of a second preset type are extracted from the plurality of key features of the first preset type.

5. A generator fault diagnosis method based on carbon brush wear signal according to claim 4, characterized in that: The first preset type includes: pressure characteristics, temperature characteristics, wear characteristics and electrical characteristics.

6. The generator fault diagnosis method based on carbon brush wear signal according to claim 1, characterized in that: The wear coefficient of the carbon brush is determined based on the extracted key features, including: A quantification method for each type of key feature is obtained based on a preset feature type-quantification method, and the extracted key features are quantified to obtain a number of original feature values; Standardize all original eigenvalues ​​to obtain several eigenvalues; Determine the wear coefficient of the carbon brush based on all characteristic values: Where W is the wear coefficient of the carbon brush, X i is the eigenvalue of the i-th key feature, α i is the weight coefficient of the i-th key feature, n is the number of key features, X j is the eigenvalue of the jth key feature, β ij is the interaction weight coefficient between the eigenvalue of the i-th key feature and the eigenvalue of the j-th key feature.

7. A generator fault diagnosis method based on carbon brush wear signal according to claim 1, characterized in that: The risk factor of the generator is determined based on the wear coefficient of the carbon brush, carbon powder and data related to the operation of the generator, including: Among them, R is the risk factor of the generator, W is the carbon brush wear coefficient, and W ref is the preset standard reference value of the carbon brush wear coefficient, C is the carbon powder concentration, P is the operating load of the generator, and C ref Preset standard reference value for toner density, P ref The preset standard reference value is for the generator operating load, Δt is the continuous operating time of the generator, T is the generator operating temperature, T ref Preset standard reference values ​​for generator operating temperature, is the risk factor adjustment value, τ is the time attenuation coefficient, is the preset weight of temperature change, and ρ is the temperature adjustment coefficient.

8. The generator fault diagnosis method based on carbon brush wear signal according to claim 1, characterized in that: Multi-level early warning is carried out based on risk factors and corresponding early warning treatment measures are taken, including: Match the risk factor with the preset risk level threshold, and then determine the warning level corresponding to the risk factor; Generate corresponding warning signals based on the warning level and the preset warning level-signal data table; Based on the early warning signal, corresponding processing suggestions are obtained in the preset database, and then processing suggestions are generated.

Images