Switch cabinet defect prediction method and system

Abstract

The invention relates to the technical field of switch cabinet defect prediction, and discloses a switch cabinet defect prediction system, which comprises a signal filtering module, a feature extraction module, a comparative analysis module, an alarm and feedback module, and a detection module a, a detection module b, a detection module c and a detection module d which are arranged in a switch cabinet, the detection module a, the detection module b, the detection module c and the detection module d are all used for detecting the sound variation of the switch cabinet during partial discharge and non-partial discharge under the power frequency of 50Hz; and the feature extraction module is used for extracting change features of sound loudness and frequency during partial discharge and non-partial discharge. According to the method, the abnormity of the switch cabinet is analyzed through the sound signals, the situation that multiple sensors are prone to being interfered by a high-frequency electric field or electromagnetic waves can be avoided, and due to the fact that the partial discharge sound is easy to collect and the needed sensor is single, the problem that the method highly depends on the quality and accuracy of sensor data can be avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of switch cabinet defect prediction, and in particular to a switch cabinet defect prediction method and a system thereof. Background Art

[0002] With the rapid development of social economy and the continuous expansion of production scale, the demand for electric energy in all walks of life has shown a rapid growth trend. In order to meet the growing demand for electric energy, the number of power equipment has increased year by year. The operating status of power equipment is directly related to the safety and reliability of the entire distribution network system. Once the power system fails, it will not only cause serious economic losses, but also have a serious impact on production and life.

[0003] The existing patent discloses a method, system and device for high-voltage switchgear defect prediction (publication number CN117094705A), which includes: collecting and storing sensor data; normalizing the sensor data and encoding it into distributed representation; constructing a mean square error loss function to train the Transformer model, and using the trained Transformer model to extract the state characteristics of the high-voltage switchgear; performing embedded expansion based on the distributed representation to obtain a preset number of sensor data; adjusting the Transformer model using the feature representation of the defect category to obtain a Transformer reconstruction model and training the model; inputting the real-time collected sensor data into the trained Transformer reconstruction model to obtain the defect prediction results of the high-voltage switchgear. The technology disclosed in this patent has the following defects: 1) This method involves multiple complex steps and techniques (such as Transformer model, multi-head self-attention mechanism, etc.), which may lead to high computational cost and resource consumption, especially when processing large-scale data.

[0004] 2) This method is highly dependent on the quality and accuracy of sensor data. If the sensor itself is faulty or the data collection is inaccurate, it may affect the final prediction results.

[0005] 3) Although the patent provides a detailed theoretical framework and technical details, the long-term stability and reliability in actual industrial environments still need further verification. Summary of the invention

[0006] The main technical problem solved by the present invention is to provide a switch cabinet defect prediction method and system thereof, which solves the problems in the above-mentioned background technology.

[0007] To solve the above technical problems, according to one aspect of the present invention, more specifically, a switch cabinet defect prediction system includes a signal filtering module, a feature extraction module, a comparison analysis module, an alarm and feedback module, and a detection module a, a detection module b, a detection module c, and a detection module d placed in the switch cabinet; The detection module a, detection module b, detection module c, and detection module d are all used to detect the sound changes of the switch cabinet during partial discharge and non-partial discharge at a power frequency of 50 Hz; The feature extraction module is used to extract the change characteristics of the sound loudness and frequency during partial discharge and non-partial discharge; The comparative analysis module estimates the energy coefficient of the sound source according to the change characteristics of the sound emitted by the sound source, and predicts whether the equipment at the sound source detected in the switch cabinet will fail according to whether the energy coefficient exceeds a set threshold.

[0008] Furthermore, the comparative analysis module estimates the energy coefficient of the sound emitted by the sound source according to the condition coefficient of the abnormal equipment at the sound source during partial discharge and the condition coefficient of the abnormal equipment at the sound source during non-partial discharge, which is: In the formula, It represents the energy coefficient at the sound source at 50Hz power frequency. The condition coefficient indicating the abnormality of the equipment at the sound source during partial discharge in the switch cabinet. The condition coefficient indicating the abnormality of equipment at the sound source when there is no partial discharge in the switch cabinet.

[0009] Furthermore, the condition coefficient of the abnormal equipment at the sound source during the partial discharge is obtained based on the average increment of the center frequency of the sound wave emitted by the detected sound source, the duration of the pulse sound wave emitted by the sound source, and the repetition rate of the pulse sound wave, which is: In the formula, The condition coefficient indicating the abnormality of the equipment at the sound source during partial discharge in the switch cabinet. represents the average increment of the center frequency of the sound wave, represents the duration of the pulse sound wave, Indicates the repetition rate of the pulse sound wave.

[0010] Furthermore, the condition coefficient of the abnormal equipment at the sound source during non-partial discharge is obtained based on the change in the frequency range of the sound wave of the detected sound source and the change in the peak amplitude of the sound wave, which is: In the formula, Indicates the condition coefficient of abnormal equipment at the sound source when there is no partial discharge in the switch cabinet, Indicates the change in the frequency range of the sound wave. Indicates the change in the peak amplitude of the sound wave.

[0011] Furthermore, the switch cabinet defect prediction system further includes a threshold setting module, and the threshold setting module is used to determine whether a fault will occur at the detected sound source according to a set threshold; when When , it means that there is an abnormality in the equipment at the sound source of the switch cabinet under the 50Hz power frequency, and the equipment at the detected sound source will fail.

[0012] Furthermore, the switch cabinet defect prediction system also includes a sample analysis module, which analyzes the size of the threshold used for judgment based on a large amount of sample data.

[0013] Furthermore, the alarm and feedback module sends early warning feedback to the remote monitoring terminal according to the analysis result of the comparison and analysis module.

[0014] Furthermore, the signal filtering module uses a Kalman filtering method to reduce noise to filter the external noise detected by the combination of the detection module a, the detection module b, the detection module c, and the detection module d.

[0015] Furthermore, the switch cabinet defect prediction system also includes a position analysis module, which analyzes the position of the equipment at the detected sound source according to the different times when the abnormal sounds are collected by the detection modules a, b, c and d.

[0016] A switch cabinet defect prediction method comprises the following steps: S1, detection module a, detection module b, detection module c, detection module d collect the sound of each unit module in the switch cabinet at a power frequency of 50Hz, and after the external noise is removed by the signal filtering module, it is fed back to the feature extraction module; S2, the feature extraction module extracts the change characteristics of the loudness and frequency in the collected sound, and feeds the change characteristics back to the comparison and analysis module for analyzing whether the unit module emitting the sound wave is abnormal; S3, the position analysis module analyzes the position of the device at the detected sound source according to the different times when the abnormal sound is collected by the detection module a, the detection module b, the detection module c, and the detection module d; S4. Finally, the alarm and analysis module will feedback the analyzed anomaly and the location of the anomaly.

[0017] The present invention provides a switch cabinet defect prediction method and system thereof. Compared with the prior art, the present method has the following effects: 1. The method of the present invention determines whether there is an abnormality in the equipment where the switch cabinet emits the sound by analyzing the sound emitted by the switch cabinet under 50HZ high voltage conditions. This can greatly reduce the requirements for hardware and, due to the relatively low cost, can realize large-scale switch cabinet fault prediction.

[0018] 2. The present invention analyzes switch cabinet anomalies through sound signals. This method can avoid the interference of multiple sensors that are easily affected by high-frequency electric fields or electromagnetic waves. In addition, since the sound of partial discharge is also easy to collect, the required sensor is single, thus avoiding the problem of being highly dependent on the quality and accuracy of sensor data.

[0019] 3. The present invention analyzes the position of the equipment detected at the sound source by using the different times at which the four detection modules receive the sound waves at the source. It can accurately predict the location of the abnormality in the switch cabinet, and then specifically know the unit module where the abnormality occurs. This can reduce the high computing cost and resource consumption to analyze which unit module has the abnormality. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the distribution of the detection modules of the present invention; Figure 2 It is a schematic diagram of the process of the present invention; Figure 3 is the conditional coefficient of the energy coefficient w and the partial discharge in the present invention , non-partial discharge condition coefficient relationship diagram. DETAILED DESCRIPTION

[0021] In order to make the technical solution of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] Example 1 like Figure 1 - As shown in Figure 2, according to one aspect of the present invention, a switch cabinet defect prediction system is provided, including a signal filtering module, a feature extraction module, a comparison analysis module, an alarm and feedback module, and a detection module a, a detection module b, a detection module c, and a detection module d placed in the switch cabinet.

[0023] The alarm and feedback module sends early warning feedback to the remote monitoring end according to the analysis results of the comparison and analysis module. The signal filtering module uses the Kalman filter method to filter the external noise detected by the combination of detection module a, detection module b, detection module c, and detection module d. Detection module a, detection module b, detection module c, and detection module d are all used to detect the sound changes of the switch cabinet during partial discharge and non-partial discharge at a power frequency of 50Hz; the feature extraction module is used to extract the change characteristics of the sound loudness and frequency during partial discharge and non-partial discharge; the comparison and analysis module estimates the energy coefficient of the sound source according to the change characteristics of the sound emitted by the sound source, and predicts whether the equipment at the sound source detected in the switch cabinet will fail according to whether the energy coefficient exceeds the set threshold. The method of analyzing the sound emitted by the switch cabinet under 50HZ high voltage conditions to determine whether there is an abnormality in the equipment where the switch cabinet emits the sound can greatly reduce the requirements for hardware, and due to the relatively low cost, it can realize large-scale switch cabinet fault prediction.

[0024] Example 2 like Figure 2 , 3 As shown, the comparative analysis module estimates the energy coefficient of the sound emitted by the sound source based on the condition coefficient of the abnormal equipment at the sound source during partial discharge (partial discharge refers to the source generated by high-frequency electricity flowing through the equipment) and the condition coefficient of the abnormal equipment at the sound source during non-partial discharge (non-partial discharge refers to the source generated by the equipment itself during operation), which is: In the formula, It represents the energy coefficient at the sound source at 50Hz power frequency. The condition coefficient indicating the abnormality of the equipment at the sound source during partial discharge in the switch cabinet. The condition coefficient indicating the abnormality of equipment at the sound source when there is no partial discharge in the switch cabinet.

[0025] The condition coefficient of equipment abnormality at the sound source during partial discharge is obtained based on the average increment of the center frequency of the sound wave emitted by the detected sound source, the duration of the pulse sound wave emitted by the sound source, and the repetition rate of the pulse sound wave, which are: In the formula, The condition coefficient indicating the abnormality of the equipment at the sound source during partial discharge in the switch cabinet. represents the average increment of the center frequency of the sound wave, represents the duration of the pulse sound wave, Indicates the repetition rate of the pulse sound wave.

[0026] The condition coefficient of equipment abnormality at the sound source during non-partial discharge is obtained based on the change in the frequency range of the sound wave of the detected sound source and the change in the peak amplitude of the sound wave, which is: In the formula, Indicates the condition coefficient of abnormal equipment at the sound source when there is no partial discharge in the switch cabinet, Indicates the change in the frequency range of the sound wave. Indicates the change in the peak amplitude of the sound wave.

[0027] Among them, when calculating the condition coefficient of abnormal equipment at the sound source during partial discharge in any switch cabinet, the average increment of the sound wave center frequency is taken as (Average increment = the average value of the center frequency exceeding the set threshold ÷ the value of the set threshold × 100%). It should be noted that when the center frequency of the sound wave does not exceed the set threshold, it indicates that the condition coefficient of the equipment abnormality at the sound source during partial discharge in the switch cabinet is .

[0028] The duration of the pulse sound wave is (Unit: ms, duration = time when the pulse sound wave ends – time when the pulse sound wave starts).

[0029] The repetition rate of the pulse sound wave is (Repetition rate = the sum of pulse sound waves collected per unit time / unit time). Since the switch cabinet operates at a high voltage frequency of 50Hz, there must be pulse sound waves in the switch cabinet. Then we can have: From the above calculation, we can know that the condition coefficient of abnormal equipment at the sound source during partial discharge in the switch cabinet is The larger the value of the equipment abnormal condition coefficient at the sound source during partial discharge in the switch cabinet, the greater the probability of equipment abnormality at the sound source.

[0030] Then, when calculating the condition coefficient of equipment abnormality at the sound source during non-partial discharge in any switch cabinet, the change in the sound wave frequency range is taken as (The change in the sound wave frequency range = the sound wave frequency exceeding the set threshold ÷ the set threshold size × 100%).

[0031] The change in the peak amplitude of the sound wave is taken as (The change in the peak amplitude of the sound wave = the peak amplitude exceeding the set threshold ÷ the set threshold value × 100%). When the sound wave frequency or the sound wave peak value does not exceed the set threshold, the condition coefficient of the abnormal equipment at the sound source in the switch cabinet during non-partial discharge is . And there are: From the above calculation, we can know that the condition coefficient of the abnormal equipment at the sound source during non-partial discharge in the switch cabinet is Then the energy coefficient of the switch cabinet at the sound source under the 50Hz power frequency is: From the above calculation, we can know that the energy coefficient of the switch cabinet at the sound source under the power frequency of 50Hz is . This means that there is an abnormality in the equipment at the sound source of the switch cabinet at the 50Hz power frequency, and the equipment at the detected sound source will fail.

[0032] Example 3 like Figure 2 As shown, the switch cabinet defect prediction system also includes a threshold setting module, which is used to determine whether a fault will occur at the detected sound source according to the set threshold. The switch cabinet defect prediction system also includes a sample analysis module, which analyzes the size of the threshold used for judgment based on a large amount of sample data. By analyzing a large amount of data samples, there are: Table 1 Analysis parameters and device status at the sound source in some embodiments From the data in Table 1, we can know that when the sample tends to infinity, we can analyze whether the source equipment is abnormal or not by using the energy coefficient w as the dividing line. When , it means that there is an abnormality in the equipment at the sound source of the switch cabinet under the 50Hz power frequency, and the equipment at the detected sound source will fail.

[0033] Example 4 like Figure 1 , 2 As shown, the switch cabinet defect prediction system also includes a position analysis module, which analyzes the position of the equipment at the detected sound source according to the different times when the abnormal sounds are collected by the detection modules a, b, c, and d.

[0034] The location of the device at the sound source is analyzed based on the different times when the four detection modules collect abnormal sounds. This method is usually called the time difference positioning method. The time difference positioning method measures the time difference when the abnormal sound reaches different sensors and uses geometric relationships to calculate the specific location of the sound source. Compared with other positioning methods, the time difference positioning method can provide higher positioning accuracy, especially inside smaller high-voltage switch cabinets.

[0035] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A switch cabinet defect prediction system, characterized in that: It includes a signal filtering module, a feature extraction module, a comparison and analysis module, an alarm and feedback module, and detection modules a, b, c and d placed in the switch cabinet; The detection module a, detection module b, detection module c, and detection module d are all used to detect the sound changes of the switch cabinet during partial discharge and non-partial discharge at a power frequency of 50 Hz; The feature extraction module is used to extract the change characteristics of the sound loudness and frequency during partial discharge and non-partial discharge; The comparative analysis module estimates the energy coefficient of the sound source according to the change characteristics of the sound emitted by the sound source, and predicts whether the equipment at the sound source detected in the switch cabinet will fail according to whether the energy coefficient exceeds a set threshold.

2. The switch cabinet defect prediction system according to claim 1, characterized in that: The comparative analysis module estimates the energy coefficient of the sound emitted by the sound source according to the condition coefficient of the abnormal equipment at the sound source during partial discharge and the condition coefficient of the abnormal equipment at the sound source during non-partial discharge, which is: In the formula, It represents the energy coefficient at the sound source at 50Hz power frequency. The condition coefficient indicating the abnormality of the equipment at the sound source during partial discharge in the switch cabinet. The condition coefficient indicating the abnormality of equipment at the sound source when there is no partial discharge in the switch cabinet.

3. The switch cabinet defect prediction system according to claim 2, characterized in that: The condition coefficient of equipment abnormality at the sound source during partial discharge is obtained based on the average increment of the center frequency of the sound wave emitted by the detected sound source, the duration of the pulse sound wave emitted by the sound source, and the repetition rate of the pulse sound wave, which is: In the formula, The condition coefficient indicating the abnormality of the equipment at the sound source during partial discharge in the switch cabinet. represents the average increment of the center frequency of the sound wave, represents the duration of the pulse sound wave, Indicates the repetition rate of the pulse sound wave.

4. The switch cabinet defect prediction system according to claim 2, characterized in that: The condition coefficient of equipment abnormality at the sound source during non-partial discharge is obtained based on the change in the frequency range of the sound wave of the detected sound source and the change in the peak amplitude of the sound wave, which is: In the formula, Indicates the condition coefficient of abnormal equipment at the sound source when there is no partial discharge in the switch cabinet, Indicates the change in the frequency range of the sound wave. Indicates the change in the peak amplitude of the sound wave.

5. The switch cabinet defect prediction system according to claim 2, characterized in that: The switch cabinet defect prediction system further includes a threshold setting module, which is used to determine whether a fault will occur at the detected sound source according to a set threshold; when When , it means that there is an abnormality in the equipment at the sound source of the switch cabinet under the 50Hz power frequency, and the equipment at the detected sound source will fail.

6. The switch cabinet defect prediction system according to claim 5, characterized in that: The switch cabinet defect prediction system also includes a sample analysis module, which analyzes the size of the threshold used for judgment based on a large amount of sample data.

7. The switch cabinet defect prediction system according to claim 1, characterized in that: The alarm and feedback module sends early warning feedback to the remote monitoring terminal according to the analysis results of the comparison and analysis module.

8. The switch cabinet defect prediction system according to claim 1, characterized in that: The signal filtering module uses the Kalman filtering method to reduce noise to filter the external noise detected by the combination of the detection module a, the detection module b, the detection module c, and the detection module d.

9. The switch cabinet defect prediction system according to claim 1, characterized in that: The switch cabinet defect prediction system also includes a position analysis module, which analyzes the position of the equipment at the detected sound source according to the different times when the abnormal sounds are collected by the detection modules a, b, c and d.

10. A switch cabinet defect prediction method, characterized in that: Applied to the prediction system according to any one of claims 1 to 9, the switch cabinet defect prediction method comprises the following steps: S1, detection module a, detection module b, detection module c, detection module d collect the sound of each unit module in the switch cabinet at a power frequency of 50Hz, and after the external noise is removed by the signal filtering module, it is fed back to the feature extraction module; S2, the feature extraction module extracts the change characteristics of the loudness and frequency in the collected sound, and feeds the change characteristics back to the comparison and analysis module for analyzing whether the unit module emitting the sound wave is abnormal; S3, the position analysis module analyzes the position of the device at the detected sound source according to the different times when the abnormal sound is collected by the detection module a, the detection module b, the detection module c, and the detection module d; S4. Finally, the alarm and analysis module will feedback the analyzed anomaly and the location of the anomaly.

Images