Fault diagnosis method and system based on electrical box parameter prediction

Abstract

The invention discloses a fault diagnosis method and system based on electrical box parameter prediction. The method specifically comprises the steps that 1, actual parameters of an electrical box are collected; 2, preprocessing the collected data; judging whether abnormal parameters exist or not based on the data collected at the current moment, and if yes, turning to the step 5; otherwise, turning to the step 3; 3, constructing an autoregressive moving average model by adopting a time sequence analysis method, and predicting parameters of the next moment by adopting the autoregressive moving average model based on the collected data; 4, judging whether abnormal parameters exist in the predicted parameters at the next moment or not, and if yes, turning to the step 5; otherwise, turning to the step 1; 5, if yes, abnormal parameters are extracted, the fault level is judged according to the abnormal parameters, then an abnormal parameter matrix is constructed, and data in the abnormal parameter matrix are subjected to standardization processing; and 6, carrying out feature extraction on the abnormal parameter matrix so as to judge the fault type and position of the electrical box at the current moment, and carrying out early warning based on the fault type.

Description

Technical Field

[0001] The present invention belongs to the technical field of electrical equipment, and particularly relates to a fault diagnosis method and system based on the prediction of electrical box parameters. Background Art

[0002] In modern industries, buildings, and power systems, as the core equipment for power distribution and control, the stable operation of electrical boxes is crucial for the reliability and safety of the entire system. However, during long-term operation, electrical boxes are prone to various faults due to reasons such as aging of electrical components, overload, and environmental factors (such as high temperature, humidity, and erosion by harmful gases), which may even lead to serious accidents such as fires and equipment damage.

[0003] Traditional monitoring methods for electrical boxes mainly rely on manual inspections, which have many limitations. On the one hand, the time interval for manual inspections is relatively long, making it difficult to capture abnormal changes during the operation of electrical boxes in real time and unable to detect potential fault hazards in a timely manner. On the other hand, the accuracy of manual judgment is easily affected by the experience and subjective factors of the inspection personnel, posing risks of missed inspections and misjudgments. In addition, some existing monitoring systems can only monitor single or a few parameters (such as current and voltage), unable to comprehensively reflect the operating status of electrical boxes. Even if some systems can achieve multi-parameter monitoring, in terms of data processing and fault diagnosis, they mostly use simple threshold comparison methods, lacking the ability to accurately identify complex fault patterns, making it difficult to effectively warn and diagnose faults in the early stage and unable to meet the requirements of modern power systems for high reliability and intelligent operation and maintenance of electrical boxes. Summary of the Invention

[0004] Object of the Invention: To solve the problems existing in the above-mentioned prior art, the present invention provides a fault diagnosis method and system based on the prediction of electrical box parameters.

[0005] Technical Solution: The present invention discloses a fault diagnosis method based on the prediction of electrical box parameters, which specifically includes the following steps:

[0006] Step 1: Real-time collect the actual parameters of the electrical box;

[0007] Step 2: Preprocess the collected data. For any one of the preprocessed parameters, construct it into a multi-dimensional time series , where t represents time; based on the data collected at the current moment, determine whether there are abnormal parameters. If so, go to Step 5; otherwise, go to Step 3;

[0008] Step 3: Use the time series analysis method to construct an autoregressive moving average model, and based on the collected data, use the autoregressive moving average model to predict the parameters at the next moment;

[0009] Step 4: Determine whether there are abnormal parameters among the predicted parameters for the next moment. If so, go to Step 5; otherwise, go to Step 1;

[0010] Step 5: Then extract the abnormal parameters, judge the fault level according to the abnormal parameters, then construct an abnormal parameter matrix, and perform normalization processing on the data in the abnormal parameter matrix;

[0011] Step 6: Extract features from the abnormal parameter matrix, thereby judge the fault type and location of the electrical box at the current moment, and give an early warning based on the fault type.

[0012] Furthermore, a corresponding first threshold is set for each type of parameter, and the abnormal parameter is a parameter whose value exceeds the corresponding first threshold.

[0013] Furthermore, the parameters of the electrical box in Step 1 include electrical parameters and environmental parameters. The electrical parameters include current, voltage, active power, reactive power, apparent power, and frequency. The environmental parameters include air humidity, temperature, and gas composition inside the electrical box.

[0014] Furthermore, the preprocessing includes adaptive filtering and normalization.

[0015] Furthermore, Step 3 is specifically as follows:

[0016] Step 3.1: Based on the historical data of any parameter, establish an autoregressive moving average model for this parameter:

[0017] ;

[0018] Among them, is the autoregressive order, is the moving average order, is the autoregressive coefficient, is the moving average coefficient, is the white noise sequence, is the forgetting factor;

[0019] Step 3.2: Dynamically determine the optimal autoregressive order and the moving average order using the following formula:

[0020]

[0021] Among them, represents the Akaike information criterion value, ; is the likelihood function value, and select the values of p and q corresponding to the minimum value of ;

[0022] Step 3.3: Update the parameters in the model of Step 3.1 in real time using the following formula:

[0023] ;

[0024] where, is the gain matrix, is the covariance matrix, is the parameter vector, , is the historical data vector, , T represents transpose;

[0025] Calculate the absolute value of the difference between the predicted value and the actual value of the parameter at time t , if ; then perform interpolation processing on the predicted value of the parameter at time t; k takes 2 or 3, is the residual standard deviation;

[0026] Step 3.4: Based on the model parameters at time t, predict the parameters of the electrical box at time t + 1:

[0027] ;

[0028] where, is the multi-dimensional time series composed of the predicted parameters.

[0029] Furthermore, the following formula is used to standardize the abnormal data in Step 5:

[0030] ;

[0031] where, is the data after standardization processing, is the data in the a-th row and b-th column of the abnormal parameter matrix, is the mean value, is the standard deviation.

[0032] Furthermore, Step 6 is specifically:

[0033] Step 5.1: Use the CNN model to predict the fault type probability and the fault location probability. The expression of the convolution used by the CNN model is as follows:

[0034] ;

[0035] where, is the th feature output by the th row and th The weight corresponding to the element in the m-th row and n-th column of the th feature in the convolutional layer, is the input feature of the th convolutional layer, is the bias term, and

[0036]

[0036] the pooling layer in the CNN model adopts adaptive pooling;

[0037] Step 5.2: Use the fault type and fault location corresponding to the maximum probability output by the CNN model as the final fault type and fault location.

[0038] Furthermore, the early warning based on the fault type in step 6 is specifically as follows: the first threshold, the second threshold, and the third threshold corresponding to each parameter setting; for each abnormal parameter among all abnormal parameters, if the value of the abnormal parameter is greater than the corresponding first threshold and less than or equal to the corresponding second threshold ; then the fault is determined as a first-level fault, and the warning light shows orange, denotes the first threshold corresponding to the th abnormal parameter, denotes the second threshold corresponding to the th abnormal parameter, f = 1, 2,..., F; F represents the total number of abnormal parameters. If there exists an abnormal parameter whose value is greater than the second threshold and less than or equal to the third threshold , then the fault is determined as a second-level fault, the warning light shows yellow, and the staff is reminded with a volume of Q decibels, is the second threshold corresponding to the abnormal parameter Y, is the third threshold corresponding to the abnormal parameter Y; if there exists an abnormal parameter whose value is greater than the third threshold , the fault is determined as a third-level fault, then the warning light shows red, and the staff is reminded with a volume greater than Q decibels.

[0039] A fault diagnosis system based on electrical box parameter prediction, comprising:

[0040] A parameter monitoring module for real-time monitoring of the actual parameters of the electrical box and transmitting them to the data processing module;

[0041] A data processing module for preprocessing the received data;

[0042] A data prediction module for constructing an autoregressive moving average model and predicting the value of the parameter at the next moment based on the autoregressive moving average model;

[0043] A fault diagnosis module, which is used to extract abnormal parameters and determine the fault type and fault location according to the abnormal parameters;

[0044] An alarm module, which is used to send alarm information according to the fault type;

[0045] A power supply module, including a main power supply and a backup battery. When the main power supply is normal, the main power supply powers the system and charges the backup battery; when the main power supply fails, the backup battery supplies power;

[0046] A remote monitoring module, which is used to transmit the operating status, monitoring data, and fault diagnosis report of the system to a remote monitoring center in real time through network communication. Managers can access the remote monitoring center through terminal devices to view the operating conditions of the electrical box. Managers can also adjust parameters and restart the device of the electrical box through the remote monitoring module.

[0047] Advantageous effects: The present invention realizes multi-parameter real-time monitoring of the electrical parameters and environmental parameters of the electrical box, and can comprehensively and accurately master the operating status of the electrical box. The adaptive filtering algorithm is used to process the data, effectively removing noise interference and improving the accuracy and reliability of the data. By establishing a change trend model of the operating parameters of the electrical box, the operating trend of the electrical box can be predicted, and potential faults can be detected in advance. The artificial intelligence algorithm is used for fault diagnosis, improving the accuracy and efficiency of fault diagnosis, and being able to quickly and accurately predict the fault type, fault location, and fault severity. There are multiple warning methods, and graded warnings are carried out according to the fault severity, which can timely notify relevant personnel and facilitate taking corresponding measures. The remote monitoring function enables managers to understand the operating conditions of the electrical box anytime and anywhere and can perform remote operations, improving the convenience and efficiency of management. Description of the Drawings

[0048] Figure 1 is the system block diagram of the present invention.

[0049] Figure 2 is the flowchart of the present invention. Detailed Embodiment

[0050] The drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention.

[0051] As Figure 1-2 shown, this embodiment specifically includes the following parts:

[0052] Multi-parameter monitoring module: It is used to monitor the electrical parameters and environmental parameters of the electrical cabinet. Among them, the electrical parameter monitoring unit includes a current sensor, a voltage sensor, a power sensor, and a frequency sensor installed inside the electrical cabinet. The power sensor calculates the active power, reactive power, and apparent power of the electrical cabinet based on the data measured by the current sensor and the voltage sensor. The frequency sensor monitors the fluctuation of the power supply frequency. The electrical parameter monitoring unit converts the collected electrical parameter information into electrical signals and transmits them to the data processing module by wired or wireless means; the environmental parameter monitoring unit includes a temperature sensor, a humidity sensor, and a gas sensor. The temperature sensor detects the temperature near the electronic components inside the electrical cabinet. The humidity sensor measures the air humidity inside the electrical cabinet. The gas sensor detects the harmful gases generated by overheating of electrical components and decomposition of insulating materials inside the electrical cabinet. The environmental parameter monitoring unit transmits the collected data to the data processing module.

[0053] Data processing module: It receives the electrical parameter and environmental parameter data transmitted by the multi-parameter monitoring module, filters the data to remove noise interference, and adopts an adaptive filtering algorithm to automatically adjust the filtering parameters according to the real-time changes of the data. It performs real-time analysis and calculation on the processed data, judges whether the current operating state of the electrical cabinet is normal by setting the normal operating parameter range, and stores and analyzes the historical data. It establishes a change trend model of the operating parameters of the electrical cabinet. The specific method is as follows: perform normalization processing on the historical data. For the gth type of electrical parameter , the normalization formula is ; is the minimum value of the gth type of electrical parameter, is the maximum value of the gth type of electrical parameter, g = 1, 2,..., G; G represents the total category of electrical parameters; for environmental parameters , the normalization formula is ; represents the normalized environmental parameter. represents the minimum value of the hth type of environmental parameter, is the maximum value of the hth type of environmental parameter, h = 1, 2,..., H; H represents the total category of environmental parameters.

[0054] Adopt the time series analysis method to construct an autoregressive moving average model , and estimate the model parameters by the least squares method to determine the optimal and values, so as to establish a trend model of the operating parameters of the electrical cabinet changing with time; specifically:

[0055] Obtain historical electrical parameters and environmental parameters from the data processing module, construct a multi-dimensional time series for any one parameter, and perform standardization processing on parameters with different dimensions: , is the mean value, and is the standard deviation; is the value after standardization.

[0056] Dynamically determine the optimal autoregressive order and the moving average order ;

[0057] ;

[0058] where, ; is the likelihood function value.

[0059] For and calculate AIC, and select the corresponding to the minimum value.

[0060] Introduce a forgetting factor , assign higher weights to recent data, suppress the influence of historical noise, and thus establish a trend model for the operating parameters of the electrical box changing with time: .

[0061] where, is the autoregressive coefficient, is the moving average coefficient, is the white noise term;

[0062] Use the recursive least squares method to update the parameters, and the iterative formula is:

[0063] ;

[0064] where, is the parameter vector.

[0065] is the historical data vector.

[0066] is the gain matrix, is the covariance matrix;

[0067] Outlier detection and correction

[0068] Calculate the model prediction residual , represents the parameter predicted at time t. If , take 2 or 3, is the residual standard deviation, then the predicted parameter is identified as an outlier; replace the outlier with the linear interpolation of adjacent time data.

[0069] Trend prediction and model update

[0070] Predict the value at the next moment based on the current parameters: ;

[0071] After collecting a certain number of new samples, re - execute the order optimization and parameter estimation to adapt to the change of the running state.

[0072] The data processing module also performs pattern recognition on the abnormal data through artificial intelligence algorithms and transmits it to the fault diagnosis module.

[0073] Fault diagnosis module: Receive abnormal data, judge the fault type, fault location and fault severity, and generate a fault diagnosis report. The specific method is as follows:

[0074] Pre - process the abnormal data, and denote the abnormal data matrix as , where is the number of samples, is the number of features; Use the standardization method to process the data. For each feature , its standardization formula is: , where is the mean value, is the standard deviation.

[0075] Use CNN for feature extraction and pattern recognition. Introduce dilated convolution in the convolutional layer to expand the receptive field. The expression is as follows:

[0076] ;

[0077] Among them, is the th feature output by the th row and th column element feature of the is the th element of the th feature in the convolutional layer corresponding weight, is the input feature of the th convolutional layer, is the bias term, is the size of the convolutional kernel.

[0078] Use adaptive pooling to replace fixed - size pooling, and dynamically adjust the pooling window size according to the feature map; Add a GAP layer after the last convolution layer to reduce parameters and enhance translational invariance: , where, is the feature map size, is the th global mean value of the feature map.

[0079] After being processed by the convolutional layer and the pooling layer, the feature map is flattened into a one-dimensional vector and input into the fully connected layer;

[0080] The fully connected layer uses multi-task learning to simultaneously predict the fault type and the location :

[0081]

[0082] where is the one-dimensional feature vector output by the pooling layer, is the number of input neurons.

[0083] Apply Softmax to each task branch respectively to convert the output of the fully connected layer into a probability distribution:

[0084] ;

[0085] Output vector: ; represents the type probability vector, represents the location probability vector, and select the type and location corresponding to the maximum probability in and .

[0086] Early warning module: According to the severity of the fault in the fault diagnosis report, different levels of early warning methods are adopted. Specifically, the first threshold, the second threshold, and the third threshold corresponding to each parameter setting; for each parameter among all abnormal parameters, if the value of the parameter is greater than the corresponding first threshold less than or equal to the corresponding second threshold ; then the fault is determined to be a first-level fault, and the warning light shows orange, represents the first threshold corresponding to the th abnormal parameter, represents the second threshold corresponding to the th abnormal parameter, f = 1, 2,..., F; F represents the total number of abnormal parameters. If there exists an abnormal parameter whose value is greater than the second threshold less than or equal to the third threshold , then the fault is determined to be a second-level fault, the warning light shows yellow, and the staff is reminded with a volume of I decibels, is the second threshold corresponding to the abnormal parameter Y, is the third threshold corresponding to the abnormal parameter Y; if there exists an abnormal parameter whose value is greater than the third threshold , the fault is determined to be a third-level fault, then the warning light shows red, and the staff is reminded with a volume greater than I decibels.

[0087] Remote monitoring module: It transmits the operating status, monitoring data, and fault diagnosis reports of the system to the remote monitoring center in real time through network communication, allowing management personnel to access the remote monitoring center through terminal devices to view the operating conditions of the electrical box, and allowing management personnel to remotely adjust parameters and restart the device for the electrical box.

[0088] Power supply module: It provides stable power supply for the entire system. It adopts dual power supply, including mains power supply and backup battery power supply. When the mains power is normal, the mains power supplies the system and charges the backup battery; when the mains power fails, the backup battery automatically switches to supply power.

[0089] In addition, the current sensor uses a closed-loop Hall current sensor; the voltage sensor uses a resistive voltage divider voltage sensor; the temperature sensor uses a digital temperature sensor; the gas sensor includes a carbon monoxide sensor and a hydrogen sensor, and uses a semiconductor gas sensor.

[0090] The specific implementation process of this embodiment is as follows:

[0091] In practical applications, each sensor of the multi-parameter monitoring module is installed at a suitable position inside the electrical box. The current sensor (closed-loop Hall current sensor), voltage sensor (resistive voltage divider voltage sensor), power sensor, and frequency sensor collect electrical parameters such as current, voltage, power, and frequency of the electrical box in real time; the temperature sensor (digital temperature sensor), humidity sensor, and gas sensor (carbon monoxide sensor and hydrogen sensor, using semiconductor gas sensors) monitor environmental parameters such as temperature, humidity, and harmful gases inside the electrical box in real time. After the collected data is converted into electrical signals, it is transmitted to the data processing module in a wired or wireless manner.

[0092] After receiving the data, the data processing module uses an adaptive filtering algorithm to filter the data to remove noise interference. Then, according to the set normal operating parameter range, it performs real-time analysis and calculation on the processed data to determine whether the current operating status of the electrical box is normal. At the same time, it normalizes the historical data and constructs an autoregressive moving average model using time series analysis methods , estimates the model parameters by the least squares method, and establishes a change trend model of the operating parameters of the electrical box.

[0093] When the data processing module detects abnormal data, it transmits it to the fault diagnosis module. The fault diagnosis module preprocesses and standardizes the abnormal data, uses CNN for feature extraction and pattern recognition. After processing through the convolutional layer, pooling layer, and fully connected layer, it uses the softmax function to convert the output into a probability distribution to determine the fault type, fault location, and fault severity; and generates a fault diagnosis report.

[0094] The warning module takes corresponding warning methods according to the severity of the faults in the fault diagnosis report.

[0095] The remote monitoring module transmits the operating status, monitoring data, and fault diagnosis report of the system to the remote monitoring center in real time through network communication. Managers can access the remote monitoring center through terminal devices such as mobile phones and computers to view the operating conditions of the electrical box and can also remotely perform operations such as adjusting parameters and restarting the device on the electrical box.

[0096] When the mains power is normal, the power supply module is powered by the mains power and charges the backup battery; when a fault occurs in the mains power, the backup battery automatically switches to supply power to ensure the normal operation of the system.

[0097] In addition, it should be noted that, in the above specific embodiments, the various specific technical features described can be combined in any suitable manner without conflict. To avoid unnecessary repetition, the present invention does not further describe various possible combination methods.

Claims

1. A fault diagnosis method based on the prediction of electrical box parameters, characterized in that, Specifically, it includes the following steps: Step 1: Collect the actual parameters of the electrical cabinet in real time; Step 2: Preprocess the collected data. For any parameter after preprocessing, construct it into a multi-dimensional time series , where t represents time; Based on the data collected at the current moment, determine whether there are abnormal parameters. If there are, go to Step 5; otherwise, go to Step 3; Step 3: Adopt the time series analysis method to construct an autoregressive moving average model. Based on the collected data, use the autoregressive moving average model to predict the parameters at the next moment; Step 4: Determine whether there are abnormal parameters in the predicted parameters at the next moment. If there are, go to Step 5; otherwise, go to Step 1; Step 5: Extract the abnormal parameters, judge the fault level according to the abnormal parameters, then construct an abnormal parameter matrix, and perform normalization processing on the data in the abnormal parameter matrix; Step 6: Extract the features of the abnormal parameter matrix, thereby judge the fault type and location of the electrical cabinet at the current moment, and give an early warning based on the fault type.

2. The fault diagnosis method based on the prediction of electrical box parameters according to claim 1, wherein Set a corresponding first threshold for each parameter, and the abnormal parameter is the parameter whose value exceeds the corresponding first threshold.

3. A fault diagnosis method based on electrical box parameter prediction according to claim 1, characterized in that The parameters of the electrical cabinet in Step 1 include electrical parameters and environmental parameters. The electrical parameters include current, voltage, active power, reactive power, apparent power, and frequency. The environmental parameters include air humidity, temperature, and gas composition inside the electrical cabinet.

4. A fault diagnosis method based on electrical box parameter prediction according to claim 1, characterized in that, The preprocessing includes adaptive filtering and normalization.

5. A fault diagnosis method based on electrical cabinet parameter prediction according to claim 1, characterized in that Specifically, Step 3 is as follows: Step 3.1: Based on the historical data of any parameter, establish an autoregressive moving average model for this parameter: ； Among them, is the autoregressive order, is the moving average order, is the autoregressive coefficient, is the moving average coefficient, is the white noise sequence, is the forgetting factor; Step 3.2: Dynamically determine the optimal autoregressive order using the following formula and the moving average order : ； Among them, represents the Akaike information criterion value, ; is the likelihood function value, select the minimum value corresponding to and the value of; Step 3.3: Use the following formula to update the parameters in the model of Step 3.1 in real time: ； Among them, is the gain matrix, is the covariance matrix, is the parameter vector, , is the historical data vector, , where T represents the transpose; Calculate the absolute value of the difference between the predicted value and the actual value of the parameter at time t , if ; then perform interpolation on the predicted value of the parameter at time t; where k takes 2 or 3, is the standard deviation of the residuals Step 3.4: Based on the model parameters at time t, predict the parameters of the electrical cabinet at time t + 1: ； wherein, is a multi-dimensional time series composed of predicted parameters.

6. A fault diagnosis method based on electrical box parameter prediction according to claim 1, characterized in that, In Step 5, the following formula is used to perform normalization processing on the abnormal data: ； Among them, is the data after standardization, is the data in the ath row and bth column of the abnormal parameter matrix, is the mean value, is the standard deviation.

7. A fault diagnosis method based on parameter prediction of an electrical box according to claim 1, characterized in that, Specifically, Step 6 is as follows: Step 5.1: Use the CNN model to predict the fault type probability and fault location probability. The expression of the convolution used in the CNN model is as follows: ； Among them, is the feature of the element in the th row and th column of the th feature output by the convolutional layer, is the weight corresponding to the element in the th row and th column of the th feature in the convolutional layer, is the input feature of the th convolutional layer, is the size of the convolutional kernel; The pooling layer in the CNN model uses adaptive pooling; Step 5.2: Take the fault type and fault location corresponding to the maximum probability output by the CNN model as the final fault type and fault location.

8. A fault diagnosis method based on electrical box parameter prediction according to claim 1, characterized in that In step 6, the early warning based on the fault type is specifically as follows: for each parameter setting, there are corresponding first, second, and third thresholds; for each abnormal parameter among all abnormal parameters, if the value of the abnormal parameter is greater than its corresponding first threshold less than or equal to its corresponding second threshold ; then the fault is determined as a first-level fault, and the warning light shows orange, indicating the first threshold corresponding to the f-th abnormal parameter, indicating the second threshold corresponding to the f-th abnormal parameter, where f = 1, 2, …, F; F represents the total number of abnormal parameters. If there exists an abnormal parameter whose value is greater than the second threshold and less than or equal to the third threshold , then the fault is determined as a second-level fault, the warning light shows yellow, and the staff is reminded with a volume of Q decibels, being the second threshold corresponding to the abnormal parameter Y, being the third threshold corresponding to the abnormal parameter Y; if there exists an abnormal parameter whose value is greater than the third threshold , the fault is determined as a third-level fault, then the warning light shows red, and the staff is reminded with a volume greater than Q decibels.

9. A system for implementing the fault diagnosis method based on prediction of electrical cabinet parameters according to claim 1, characterized in that, It includes: A parameter monitoring module, which is used to monitor the actual parameters of the electrical cabinet in real time and transmit them to the data processing module; A data processing module, which is used to preprocess the received data; A data prediction module, which is used to construct an autoregressive moving average model and predict the value of the parameters at the next moment based on the autoregressive moving average model; A fault diagnosis module, which is used to extract abnormal parameters and judge the fault type and location according to the abnormal parameters; An alarm module, which is used to send an alarm message according to the fault type; A power supply module, which includes a main power supply and a backup battery. When the main power supply is normal, the main power supply powers the system and charges the backup battery; when the main power supply fails, the backup battery supplies power; A remote monitoring module, which is used to transmit the operating status, monitoring data, and fault diagnosis report of the system to the remote monitoring center in real time through network communication. Managers can access the remote monitoring center through terminal devices to view the operating conditions of the electrical cabinet. Managers can also adjust the parameters and restart the device of the electrical cabinet through the remote monitoring module.

Images