An abnormal alarm system for cables used in a smart grid

Abstract

The present invention relates to the technical field of alarm systems, and discloses a cable anomaly alarm system for smart grids. By receiving and processing operating parameters, calculating operation errors, generating correction factors based on this, and then precisely correcting preset thresholds, and finally applying the corrected preset correction thresholds to the alarm trigger judgment module, this process significantly improves the alarm accuracy and reliability of the system. By adopting this method, it can not only dynamically adapt to different working environments and condition changes, but also effectively identify and respond to subtle anomalies in cable operation, avoiding false alarms or missed alarms that may be caused by traditional fixed thresholds. In addition, since the correction process fully considers the analysis results of real-time data and historical data, the system has stronger adaptability and higher stability.

Description

Technical Field

[0001] The present invention relates to the technical field of alarm systems, and particularly to a cable anomaly alarm system for smart grids. Background Art

[0002] During the operation of modern smart grids, cables, as an important part of power transmission, their health status is directly related to the safety and stability of the entire grid. With the continuous expansion of the grid scale and the increase in complexity, real-time monitoring of the cable operation status becomes particularly important. However, traditional cable anomaly detection systems usually rely on fixed preset thresholds to determine whether the cable is in an abnormal condition. These fixed thresholds are set based on historical data or design standards and do not fully consider dynamic change factors in actual operation, such as environmental temperature, humidity, load fluctuations, etc. Therefore, in the face of a complex actual working environment, fixed thresholds often cannot accurately reflect the true operation status of the cable, easily leading to false alarms (i.e., wrongly triggering the alarm) or missed alarms (i.e., failing to detect real faults in a timely manner), which not only increases the maintenance cost but also may cause serious safety accidents.

[0003] In addition, the aging and wear of cables will further affect their electrical performance, making it more difficult for traditional fixed threshold methods to adapt to changing working conditions. To improve the accuracy and reliability of cable anomaly detection, researchers have proposed various improvement schemes, such as using machine learning algorithms for data analysis and prediction, but these methods often require a large amount of computing resources and still face many challenges in actual applications. In view of this, there is an urgent need for a method that can adaptively adjust the alarm threshold to achieve accurate monitoring and rapid response to the cable operation status. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a cable anomaly alarm system for smart grids, including the following modules:

[0005] Data acquisition module: used to collect the operation parameters during the cable operation in real time;

[0006] Preprocessing module: connected to the data acquisition module, used to preprocess the operation parameters collected in real time by the data acquisition module;

[0007] Alarm trigger judgment module: connected to the preprocessing module, used to define a preset threshold, and judge whether to trigger an alarm according to the preprocessed operation parameters and the preset threshold. If so, transmit the preprocessed operation parameters to the correction module; if not, end. The preset threshold includes at least two data points;

[0008] Calibration module: Connected to the alarm trigger judgment module, it is used to receive the pre-processed operating parameters, calculate the operation error, generate a correction factor based on the operation error, correct the preset threshold according to the correction factor to obtain a preset corrected threshold, and transmit the preset corrected threshold to the alarm trigger judgment module, replacing the preset threshold with the preset corrected threshold and applying it to the alarm trigger judgment module.

[0009] Further, the calibration module includes the following sub-modules:

[0010] Difference matrix construction sub-module: Used to collect historical samples and construct a difference matrix by taking the difference between the historical samples and the pre-processed operating parameters;

[0011] Operation error calculation sub-module: Used to calculate the operation error according to the difference matrix;

[0012] Error threshold construction sub-module: Used to construct an error threshold;

[0013] Judgment sub-module: Used to compare the calculated operation error with the error threshold. If the operation error is greater than or equal to the error threshold, the operation error is transmitted to the correction sub-module; if the operation error is less than the error threshold, it ends;

[0014] Correction sub-module: Used to obtain a correction factor according to the operation error and correct the preset threshold according to the correction factor to obtain a preset corrected threshold.

[0015] Further, the difference matrix construction sub-module includes the following units:

[0016] Historical normal operating parameter set acquisition unit: Used to acquire the historical normal operating parameter set; the historical normal operating parameter set includes several historical samples, and each historical sample includes several historical operating parameters;

[0017] Euclidean distance calculation unit: Used to calculate the Euclidean distance between the pre-processed operating parameters and the historical samples;

[0018] Mahalanobis distance calculation unit: Used to calculate the Mahalanobis distance between the pre-processed operating parameters and the historical samples;

[0019] Adjustment coefficient determination unit: Used to collect the historical temperature under the historical operating parameters, take the average of the historical temperature as the reference temperature; obtain the real-time temperature under the pre-processed operating parameters, calculate the temperature difference according to the real-time temperature and the reference temperature, and calculate the adjustment coefficients of the Euclidean distance and the Mahalanobis distance according to the temperature difference;

[0020] Distance index calculation unit: Used to calculate the distance index according to the Euclidean distance, Mahalanobis distance and adjustment coefficient, and take the historical sample with the smallest distance index as the historical sub-sample;

[0021] Difference matrix generation unit: used to subtract the preprocessed operating parameters from each historical operating parameter of the historical sub-samples to obtain the operating parameter differences, and integrate all the operating parameter differences to obtain a difference matrix.

[0022] Further, the operation error calculation sub-module includes the following units:

[0023] Difference matrix mean calculation unit: used to add up all the operating parameter differences in the difference matrix and then calculate the mean value to obtain the difference matrix mean value;

[0024] Maximum operating parameter difference acquisition unit: used to obtain the maximum operating parameter difference from the difference matrix;

[0025] Operation error calculation unit: used to calculate the operation error according to the operating parameter differences, the difference matrix mean value, and the maximum operating parameter difference.

[0026] Further, the calculation formula for the operation error is:

[0027] ;

[0028] In the formula, E represents the operation error, represents the i-th operating parameter difference in the difference matrix, represents the difference matrix mean value, represents the maximum operating parameter difference, represents the weight of the i-th operating parameter difference, a represents the adjustment coefficient of the maximum operating parameter difference, n represents the total number of operating parameter differences in the difference matrix, and log is the logarithmic function.

[0029] Further, the error threshold construction sub-module includes the following units:

[0030] Error parameter acquisition unit: used to obtain the historical operation errors of the historical normal operating parameter set and determine the historical standard deviation according to the historical operation errors;

[0031] Historical average temperature acquisition unit: used to calculate the historical average temperature according to the historical temperatures under the historical operating parameters;

[0032] Error threshold calculation unit: used to calculate the error threshold according to the preset threshold, the temperature difference, the historical average temperature, and the historical standard deviation.

[0033] Further, the calculation formula for the error threshold is;

[0034] ;

[0035] In the formula, P represents the error threshold, represents the maximum value of the preset threshold, represents the temperature difference, represents the historical standard deviation, represents the historical average temperature, represents the adjustment coefficient.

[0036] Furthermore, the corrector sub-module includes the following units:

[0037] Correction factor calculation unit: used to calculate the correction factor according to the operation error;

[0038] Preset threshold correction module: used to correct the preset threshold according to the correction factor to obtain the preset correction threshold.

[0039] Furthermore, the calculation formula of the correction factor is:

[0040] ;

[0041] In the formula, F represents the correction factor, k represents the attenuation coefficient, represents the exponential decay function.

[0042] Furthermore, the calculation formula of the preset correction threshold is:

[0043] ;

[0044] Among them, represents the preset correction threshold, represents the preset threshold.

[0045] The embodiments of the present invention have the following technical effects:

[0046] By receiving and processing the operation parameters, calculating the operation error, generating a correction factor based on this, and then accurately correcting the preset threshold, and finally applying the corrected preset correction threshold to the alarm trigger judgment module, this process significantly improves the alarm accuracy and reliability of the system. By adopting this method, it can not only dynamically adapt to different working environments and condition changes, but also effectively identify and respond to subtle abnormalities in cable operation, avoiding false alarms or missed alarms that may be caused by traditional fixed thresholds. In addition, since the correction process fully considers the analysis results of real-time data and historical data, the system has stronger adaptability and higher stability. This not only helps to detect potential faults in a timely manner, take maintenance measures in advance, reduce the risk of unexpected shutdowns, but also provides strong support for the optimized management and intelligent monitoring of the power system. Therefore, the application of this invention greatly improves the accuracy and efficiency of cable monitoring, and is of great significance for ensuring the safe and stable operation of the power system, reducing operation and maintenance costs, and enhancing the overall economic benefits. Description of the Drawings

[0047] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0048] Figure 1 It is a structural diagram of a cable anomaly alarm system for a smart grid provided by an embodiment of the present invention. Specific embodiments

[0049] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present invention.

[0050] Figure 1 It is a structural diagram of a cable anomaly alarm system for a smart grid provided by an embodiment of the present invention. Refer to Figure 1 , and specifically includes the following modules:

[0051] Data acquisition module: used to collect the operating parameters during the operation of the cable in real time.

[0052] The operating parameters include but are not limited to: temperature, voltage, current, vibration signal, etc.

[0053] Preprocessing module: connected to the data acquisition module, and used to preprocess the operating parameters collected in real time by the data acquisition module.

[0054] Use a low-pass filter to filter the collected voltage and current signals to remove high-frequency noise;

[0055] Apply a normalization algorithm to map the filtered signal values to the interval [0,1];

[0056] Use wavelet transform to perform noise reduction processing on the temperature and vibration signals, and retain the effective signal components.

[0057] Alarm trigger judgment module: connected to the preprocessing module, and used to define a preset threshold, and judge whether to trigger an alarm according to the preprocessed operating parameters and the preset threshold. If so, transmit the preprocessed operating parameters to the calibration module. If not, end; the preset threshold includes at least two data points.

[0058] The alarm trigger judgment module is a prior art. It mainly monitors the cable operation parameters. When a certain operation parameter exceeds the preset threshold, it is considered that the current operation parameter is abnormal, which means that the cable has a fault or there is an abnormal risk in the future. The alarm trigger judgment module receives the abnormal signal and issues an alarm. After the staff receives the alarm, they repair the cable. The abnormal monitoring and alarm process is a prior art and will not be elaborated in this embodiment.

[0059] The innovation lies in that there are operation errors in the process of abnormal alarm of operation parameters, and this embodiment corrects based on the operation errors.

[0060] Among them, the prior art obtains the preset threshold through the mean value of historical operation parameters ± 2 times the standard deviation. The preset threshold is a value range, including an upper limit and a lower limit.

[0061] Correction module: Connected to the alarm trigger judgment module, it is used to receive the preprocessed operation parameters, calculate the operation error, generate a correction factor according to the operation error, obtain a preset correction threshold for the preset threshold according to the correction factor, and transmit the preset correction threshold to the alarm trigger judgment module, replacing the preset threshold with the preset correction threshold and applying it to the alarm trigger judgment module.

[0062] The correction module includes the following sub-modules:

[0063] Difference matrix construction sub-module: Used to collect historical samples and construct a difference matrix by taking the difference between the historical samples and the preprocessed operation parameters.

[0064] The difference matrix construction sub-module includes the following units:

[0065] Historical normal operation parameter set acquisition unit: Used to obtain the historical normal operation parameter set; the historical normal operation parameter set includes several historical samples, and each historical sample includes several historical operation parameters.

[0066] The historical normal operation parameter set is the operation parameters generated when the distribution network cable is operating normally and no alarm occurs.

[0067] Euclidean distance calculation unit: Used to calculate the Euclidean distance between the preprocessed operation parameters and the historical samples.

[0068] The calculation formula of Euclidean distance is prior art. In the intelligent grid cable anomaly alarm system, in this embodiment, the prior art formula is used to evaluate the similarity between the preprocessed operating parameters and historical samples. Measuring the similarity between the preprocessed operating parameters and historical samples can significantly identify anomalies between them. When the Euclidean distance is smaller, the similarity between the preprocessed operating parameters and historical samples is higher; when the Euclidean distance is larger, the similarity between the preprocessed operating parameters and historical samples is lower, indicating that the preprocessed operating parameters are abnormal.

[0069] ;

[0070] In the formula, represents the j-th operating parameter after preprocessing and the i-th historical sample the Euclidean distance between them; represents the weight coefficient of the j-th operating parameter, represents the parameter value of the j-th operating parameter, represents the parameter value of the p-th historical operating parameter in the i-th historical sample, represents the total number of operating parameters.

[0071] Among them, the process of obtaining the weight coefficient: taking the preprocessed operating parameters as independent variables and the failure rate after anomaly monitoring of the preprocessed operating parameters (anomaly monitoring is not an innovation point of this embodiment, and prior art often uses models to achieve monitoring and prediction to obtain the failure rate. Therefore, the process of obtaining this failure rate is prior art and will not be elaborated in detail in this embodiment) as the dependent variable, constructing a multiple linear regression analysis model and performing regression analysis, and calculating the coefficient of determination of the model as the weight coefficient. The above process of regression analysis is prior art and will not be elaborated in detail in this embodiment.

[0072] Mahalanobis distance calculation unit: used to calculate the Mahalanobis distance between the preprocessed operating parameters and historical samples.

[0073] First, subtract the preprocessed operating parameters from the historical operating parameters of each historical sample to obtain a difference vector, then construct a covariance matrix based on the historical operating parameters in the historical normal operating parameter set, and finally obtain the Mahalanobis distance according to the covariance matrix and the difference vector.

[0074] Mahalanobis distance is a method for measuring the difference between data points, especially suitable for multivariate data sets, and takes into account the correlation between variables. The main advantage of Mahalanobis distance is that it can consider the correlation between variables, rather than just their absolute differences. This makes it very useful when dealing with multivariate data, especially when there is a correlation between these variables. In a cable monitoring system, Mahalanobis distance can be used to detect anomalies. By comparing the Mahalanobis distance between the current operating parameters and the historical normal operating parameters, outliers can be identified.

[0075] ;

[0076] In the formula, represents the j-th operating parameter after preprocessing and the i-th historical sample the Mahalanobis distance between them; represents the difference vector, T represents the transpose operation, represents the covariance matrix.

[0077] represents the difference vector, indicating the difference between the current operating parameters and the historical sample. Through the subtraction operation, a vector can be obtained, and each element of this vector represents the difference of the corresponding parameter. represents the covariance matrix, indicating the covariance relationship between the historical operating parameters in the set of historical normal operating parameters. The covariance matrix reflects the correlation and variation degree between each variable. represents the inverse matrix of the covariance matrix, which is used to standardize the difference vector. By multiplying the inverse matrix by the difference vector, the influence of different scales between variables can be eliminated, and the correlation between variables can be taken into account. , representing the standardized distance between the current operating parameters and the historical sample. Calculate the historical Mahalanobis distance of the historical operating parameters in the set of historical normal operating parameters, and take the 95% quantile of the historical Mahalanobis distance as the Mahalanobis distance threshold. When the Mahalanobis distance obtained through the above calculation is greater than or equal to the Mahalanobis distance threshold, it means that the current preprocessed operating parameter is an abnormal data point.

[0078] Adjustment coefficient determination unit: used to collect the historical temperature under the historical operating parameters, take the average value of the historical temperature as the reference temperature; obtain the real-time temperature under the preprocessed operating parameters, calculate the temperature difference according to the real-time temperature and the reference temperature, and calculate the adjustment coefficient of the Euclidean distance and the Mahalanobis distance according to the temperature difference.

[0079] ;

[0080] Among them, f represents the adjustment coefficient, represents the temperature difference, represents the adjustment factor of the temperature difference.

[0081] By experimentally measuring the corresponding adjustment coefficients at different temperature differences and then using these data points for curve fitting, we obtain .

[0082] In a cable monitoring system, temperature is a key parameter as it directly affects the operating status and performance of the cable. By considering the temperature difference, it is possible to more accurately evaluate whether the current operating parameters are abnormal.

[0083] Distance index calculation unit: used to calculate the distance index based on the Euclidean distance, Mahalanobis distance, and adjustment coefficient, and take the historical sample with the minimum distance index as the historical sub-sample.

[0084] ;

[0085] In the formula, represents the distance index between the j-th preprocessed operating parameter and the i-th historical sample.

[0086] Difference matrix generation unit: used to subtract each historical operating parameter of the preprocessed operating parameter from the historical sub-sample to obtain the operating parameter difference, and integrate all the operating parameter differences to obtain the difference matrix.

[0087] By combining the Euclidean distance and the Mahalanobis distance and introducing the temperature difference as an adjustment coefficient to calculate the distance index, it is possible to more accurately screen out the most suitable reference historical samples. First, by combining the two distance metrics, it is possible to capture both the local characteristics and the global structure of the data simultaneously - the Euclidean distance emphasizes the absolute difference and is suitable for quick preliminary screening; the Mahalanobis distance conducts fine screening by considering the correlation between variables and the characteristics of the data distribution. Second, temperature is one of the key factors affecting the operating status of the cable, and there may be significant differences in the performance of the cable at different temperatures. Therefore, introducing the temperature difference as an adjustment coefficient can dynamically adjust the weight according to the current environmental conditions to adapt to different operating scenarios. Specifically, the adjustment coefficient in the distance index formula adjusts the contribution ratio of the Euclidean distance and the Mahalanobis distance based on the temperature difference, ensuring that the most suitable historical samples can be selected under different temperature conditions. This method not only enhances the flexibility and adaptability of the model but also improves the accuracy of the screening process because it is based on a comprehensive consideration of multiple factors rather than a single criterion. In this way, the selected samples can better reflect the real operating status and contribute to improving the quality of subsequent analysis and decision-making.

[0088] Operation error calculation sub-module: used to calculate the operation error based on the difference matrix.

[0089] The operation error calculation sub-module includes the following units:

[0090] Difference matrix mean calculation unit: used to sum all operation parameter differences in the difference matrix and then calculate the mean to obtain the difference matrix mean.

[0091] Maximum operation parameter difference acquisition unit: used to obtain the maximum operation parameter difference from the difference matrix;

[0092] Operation error calculation unit: used to calculate the operation error based on the operation parameter difference, the difference matrix mean, and the maximum operation parameter difference.

[0093] ;

[0094] In the formula, E represents the operation error, represents the i-th operation parameter difference in the difference matrix, represents the difference matrix mean, represents the maximum operation parameter difference, represents the weight of the i-th operation parameter difference, a represents the adjustment factor of the maximum operation parameter difference, , n represents the total number of operation parameter differences in the difference matrix. Take the reciprocal of the variance of the operation parameter differences in the difference matrix as the weight , is the logarithmic function.

[0095] , by considering the deviation of each operation parameter difference from its mean and assigning different weights, effectively reflects the relative importance of each parameter in the overall error, enabling those parameters that have a greater impact on the system performance to receive more attention in the error calculation.

[0096] , a correction mechanism for extreme values (i.e., the maximum operation parameter difference in the difference matrix) is introduced. It can not only reduce the impact of individual extremely large or small outliers on the total error, but also maintain sensitivity to significant deviations from the normal range, thus avoiding misjudgment caused by extreme changes in a single parameter.

[0097] The combination of these two parts not only ensures the comprehensiveness and accuracy of error evaluation, but also enhances the adaptability of the model to complex and changeable conditions in the actual environment, providing a more reliable basis for anomaly detection in application scenarios such as cable monitoring. This comprehensive consideration method helps to more accurately identify potential problem points and improves the stability and reliability of the system.

[0098] The operating error is calculated through three parameters: the difference in operating parameters, the mean of the difference matrix, and the maximum difference in operating parameters, which can effectively capture the deviation between the current state of the system and its historical normal state. First, the difference in operating parameters reflects the deviation degree of each specific parameter relative to its historical sample value, enabling us to evaluate the health status of the system separately for each dimension. The mean of the difference matrix provides an overall perspective, summarizing the average deviation degree of all parameters and helping to identify whether there are general offsets or anomalies in the system. Considering the maximum difference in operating parameters allows us to focus on extreme deviations that may pose a significant threat to the system's stability. The advantage of combining these three parameters to calculate the operating error is that this comprehensive method can not only cover multi-level analysis from individual to overall, ensuring that no important deviation information is missed; but also, by quantifying these differences, it can more accurately reflect the gap between the actual operating condition of the system and the ideal state. In addition, this method improves the detection ability of abnormal behaviors, making the operating error a sensitive indicator for measuring the system's performance and reliability, and thus supporting more timely and effective maintenance decisions.

[0099] Error threshold construction sub-module: used to construct the error threshold.

[0100] The error threshold construction sub-module includes the following units:

[0101] Error parameter acquisition unit: used to obtain the historical operating error of the historical normal operating parameter set and determine the historical standard deviation according to the historical operating error.

[0102] Historical average temperature acquisition unit: used to calculate the historical average temperature based on the historical temperature under the historical operating parameters;

[0103] Error threshold calculation unit: used to calculate the error threshold according to the preset threshold, temperature difference, historical average temperature, and historical standard deviation.

[0104] ;

[0105] In the formula, P represents the error threshold, represents the maximum value of the preset threshold, represents the temperature difference, represents the historical standard deviation, represents the historical average temperature, represents the adjustment coefficient.

[0106] In this embodiment, by combining temperature changes and historical standard deviations to dynamically adjust the error threshold, the system can automatically adapt and set a reasonable error range under different temperature conditions. This dynamic adjustment mechanism helps to improve the robustness and reliability of the system, ensuring that operation errors can be effectively monitored and controlled in various operating environments, thereby promptly detecting and handling abnormal situations and guaranteeing the stable operation of the system.

[0107] Judgment sub-module: used to compare the calculated operation error with the error threshold. If the operation error is greater than or equal to the error threshold, the operation error is transmitted to the correction sub-module; if the operation error is less than the error threshold, it ends.

[0108] Correction sub-module: used to obtain a correction factor based on the operation error and correct the preset threshold according to the correction factor to obtain a preset correction threshold.

[0109] The correction sub-module includes the following units:

[0110] Correction factor calculation unit: used to calculate the correction factor according to the operation error;

[0111] Preset threshold correction module: used to correct the preset threshold according to the correction factor to obtain a preset correction threshold.

[0112] ;

[0113] ;

[0114] In the formula, F represents the correction factor, k represents the attenuation coefficient, and the attenuation coefficient is obtained through experimental determination, k = 2.197. represents the preset correction threshold. represents the preset threshold. represents the exponential decay function.

[0115] Through the operation error is mapped to a value between 0 and 1. When the operation error is small (0.1 and within) or almost 0, no correction is required; when E is close to 1, based on the experimentally determined value of k, the actual correction factor is calculated to be 0.902, that is, the preset threshold is corrected by means of 0.902.

[0116] When the operation error is greater than 0.1 (in this embodiment, it is considered that the current operation error is relatively large), it indicates that the system has deviated from the normal operating state and entered a large error range. In this case, using a larger correction factor to adjust the preset threshold can effectively amplify the system's response to these significant deviations. This not only helps to identify potential problems or abnormal situations in a timely manner, but also improves the system's early warning ability and stability. For example, in a cable monitoring system, if the difference between the real-time monitored operation parameters and the historical samples increases significantly, the system can use a higher correction factor to adjust its judgment criteria in a timely manner, so as to issue an alarm or take preventive measures earlier, avoid the occurrence or expansion of faults, ensure the system sensitivity, enhance its adaptability to changes in actual operating conditions, and improve the overall reliability and safety of the system.

[0117] When the preset correction threshold after correction is applied to the alarm trigger judgment module, the monitoring ability for abnormal operating parameters is significantly enhanced. By dynamically adjusting the preset threshold to reflect the changes in the current system state and environmental conditions, it is possible to more accurately judge whether the cable is abnormal. This adaptive mechanism not only improves the detection accuracy, reduces the possibility of false alarms and missed alarms, but also can be optimized for different working environments and operating conditions to ensure that potential problems can be identified in a timely manner under any circumstances. In addition, since the correction process takes into account the actual error distribution and historical data, the system has stronger robustness and higher reliability, thus providing a scientific basis for cable maintenance, helping to detect and solve possible faults in advance, avoid the occurrence of major losses, and extend the service life of the cable. Therefore, using the corrected preset threshold can not only improve the system's immediate response ability and stability, but also lay a solid foundation for realizing intelligent monitoring and management. Finally, these improvement measures work together to effectively ensure the safe and stable operation of the power system.

[0118] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A cable abnormality alarm system for smart grid, characterized in that: Includes the following modules: Data acquisition module: used to collect the operating parameters of the cable operation process in real time; Preprocessing module: connected with the data acquisition module, used to preprocess the operating parameters collected in real time in the data acquisition module; Alarm trigger judgment module: connected to the preprocessing module, used to define a preset threshold, and judge whether to trigger an alarm according to the preprocessed operating parameters and the preset threshold. If yes, the preprocessed operating parameters are transmitted to the correction module, if not, the process ends; the preset threshold includes at least two data points; Correction module: connected to the alarm trigger judgment module, used to receive the pre-processed operating parameters and calculate the operation error, generate a correction factor according to the operation error, correct the preset threshold according to the correction factor to obtain a preset correction threshold, and transmit the preset correction threshold to the alarm trigger judgment module, replace the preset threshold with the preset correction threshold and apply it to the alarm trigger judgment module; Correction module, including the following sub-modules: Difference matrix construction submodule: used to collect historical samples and construct a difference matrix by subtracting the historical samples from the preprocessed operating parameters; The difference matrix construction submodule includes the following units: A historical normal operation parameter set acquisition unit is used to acquire a historical normal operation parameter set; the historical normal operation parameter set includes a number of historical samples, and each historical sample includes a number of historical operation parameters; Euclidean distance calculation unit: used to calculate the Euclidean distance between the pre-processed operating parameters and historical samples; Mahalanobis distance calculation unit: used to calculate the Mahalanobis distance between the pre-processed operating parameters and historical samples; Adjustment coefficient determination unit: used to collect historical temperatures under historical operating parameters, calculate the average of the historical temperatures as the reference temperature; obtain the real-time temperature under the preprocessed operating parameters, calculate the temperature difference between the real-time temperature and the reference temperature, and calculate the adjustment coefficients of the Euclidean distance and the Mahalanobis distance based on the temperature difference; Distance index calculation unit: used to calculate the distance index according to the Euclidean distance, Mahalanobis distance and adjustment coefficient, and take the historical sample with the smallest distance index as the historical sub-sample; ； In the formula, Represents the distance index between the jth operating parameter after preprocessing and the ith historical sample, Represents the jth running parameter after preprocessing With the i-th historical sample The Mahalanobis distance between Represents the jth running parameter after preprocessing With the i-th historical sample The Euclidean distance between them, f represents the adjustment coefficient; Difference matrix generation unit: used to subtract the preprocessed operating parameters from each historical operating parameter of the historical subsample to obtain the operating parameter difference, and integrate all operating parameter differences to obtain the difference matrix; combined with the distance measurement method, it can capture the local characteristics and global structure of the data at the same time: Euclidean distance emphasizes absolute differences and is suitable for rapid preliminary screening; Mahalanobis distance performs fine screening by considering the correlation between variables and data distribution characteristics; the adjustment coefficient adjusts the contribution ratio of Euclidean distance and Mahalanobis distance according to the temperature difference, so that suitable historical samples can be selected under different temperature conditions; Operation error calculation submodule: used to calculate the operation error according to the difference matrix; Error threshold construction submodule: used to construct the error threshold; Judgment submodule: used to compare the calculated operation error with the error threshold. If the operation error is greater than or equal to the error threshold, the operation error is transmitted to the correction submodule; if the operation error is less than the error threshold, the process ends. Correction submodule: used to obtain a correction factor according to an operation error, and to correct a preset threshold according to the correction factor to obtain a preset correction threshold.

2. A cable abnormality alarm system for smart grid according to claim 1, characterized in that: The operation error calculation submodule includes the following units: A difference matrix mean calculation unit; used for adding up all the operating parameter differences in the difference matrix and calculating the mean to obtain the difference matrix mean; Maximum operating parameter difference obtaining unit: used for obtaining the maximum operating parameter difference from the difference matrix; An operation error calculation unit is used to calculate the operation error according to the operation parameter difference, the mean of the difference matrix and the maximum operation parameter difference.

3. A cable abnormality alarm system for smart grid according to claim 2, characterized in that: The calculation formula of operating error is: ； Where E represents the operating error, represents the difference of the i-th running parameter in the difference matrix, represents the mean of the difference matrix, Represents the maximum operating parameter difference, represents the weight of the ith operating parameter difference, a represents the adjustment coefficient of the maximum operating parameter difference, n represents the total number of operating parameter differences in the difference matrix, is a logarithmic function.

4. A cable abnormality alarm system for smart grid according to claim 3, characterized in that: The error threshold construction submodule includes the following units: Error parameter acquisition unit: used to obtain the historical operation error of the historical normal operation parameter set, and determine the historical standard deviation according to the historical operation error; Historical average temperature acquisition unit: used to calculate the historical average temperature according to the historical temperature under the historical operating parameters; Error threshold calculation unit: used to calculate the error threshold according to the preset threshold, temperature difference, historical average temperature and historical standard deviation.

5. A cable abnormality alarm system for smart grid according to claim 4, characterized in that: The calculation formula of the error threshold is: ； Where P represents the error threshold, Represents the maximum value of the preset threshold, represents the temperature difference, represents the historical standard deviation, represents the historical average temperature, Represents the adjustment factor.

6. A cable abnormality alarm system for smart grid according to claim 5, characterized in that: The correction submodule includes the following units: Correction factor calculation unit: used to calculate the correction factor according to the operation error; Preset threshold correction module: used to correct the preset threshold according to the correction factor to obtain the preset correction threshold.

7. A cable abnormality alarm system for smart grid according to claim 6, characterized in that: The correction factor is calculated as: ； In the formula, F represents the correction factor, k represents the attenuation coefficient, Represents an exponential decay function.

8. The cable abnormality alarm system for smart grid according to claim 7, characterized in that: The calculation formula for the preset correction threshold is: ； in, represents the preset correction threshold, Represents the preset threshold.

Images