A ring network box with overload fault monitoring and circuit breaker function

Abstract

The present invention relates to the technical field of power load monitoring, and specifically to a ring network box with an overload fault monitoring and circuit breaker function; obtaining a load filtering time series based on the instantaneous fluctuation characteristics of the load time series; constructing a load characteristic vector based on the load filtering time series; constructing a temperature characteristic vector based on the temperature time series; obtaining temperature differences and date clusters based on the temperature characteristic vector; obtaining load differences based on the load characteristic vector; constructing a difference scatter plot and a load difference prediction model based on the temperature differences and load differences corresponding to the date clusters; obtaining difference weights based on the scatter distribution characteristics and training the load difference prediction model; obtaining a load prediction difference based on a preset temperature characteristic vector for the current time period; obtaining a deviation index based on the load prediction difference, the actual load difference, and the model prediction error characteristics; monitoring the load state based on the change characteristics of the deviation index, thereby improving the accuracy of load monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of electric load monitoring, and in particular to a ring network box with an overload fault monitoring and circuit breaker function. Background Art

[0002] In medium-voltage distribution networks, ring network distribution systems are widely used as a typical distribution solution to ensure reliable and secure power supply to urban high-rise buildings, residential areas, factories, and commercial centers. Ring network boxes with overload fault monitoring and disconnection functions are medium-voltage distribution equipment that integrates advanced sensors and intelligent control technologies. Their primary function is to monitor the load status of different circuits in the ring network distribution system in real time. When overload or other fault conditions occur, they can quickly identify and disconnect the faulty line, improving power supply reliability.

[0003] Existing power grid overload fault monitoring mainly relies on real-time comparison of equipment operating status with rated reference values. Once the load exceeds the rated value, the ring network distribution system will activate the early warning and protection mechanism; however, this method ignores the dynamic changes and patterns of power grid load in actual scenarios. For example, instantaneous load fluctuations such as the startup of large equipment and arc welding work will be misjudged as overload risks, triggering false alarms or unnecessary protection actions, affecting the accuracy of the ring network box in monitoring and controlling the load in the power grid. Summary of the Invention

[0004] In order to solve the above-mentioned technical problem that the existing load monitoring method ignores the dynamic changes and laws of the power grid load in actual scenarios, affecting the accuracy of the ring main box in monitoring and controlling the load in the power grid, the purpose of the present invention is to provide a ring main box with an overload fault monitoring and circuit breaker function. The technical solution adopted is as follows:

[0005] A data acquisition module is used to obtain the load timing of the circuit and the temperature timing of the environment;

[0006] A feature analysis module is configured to perform adaptive filtering based on the instantaneous fluctuation characteristics of the load time series to obtain a load filtering time series; construct a load feature vector based on the data trend characteristics of the current period preset in the load filtering time series; and construct a temperature feature vector based on the data trend characteristics of the current period preset in the temperature time series;

[0007] A feature processing module is configured to obtain different temperature differences and date clusters based on the difference characteristics of temperature feature vectors in the same historical period; obtain load differences based on the difference characteristics of load feature vectors in the same historical period; construct a difference scatter plot and a load difference prediction model based on the distribution characteristics of the temperature differences and load differences corresponding to the date clusters; obtain difference weights based on the scatter distribution characteristics of the scatter plot and train the load difference prediction model;

[0008] The load monitoring module is used to obtain the load prediction difference through the load difference prediction model based on the temperature characteristic vector of the preset current time period; obtain the deviation index at the current moment based on the load prediction difference and the actual load difference of the preset current time period and the model prediction error characteristics; and monitor the load status at the current moment based on the change characteristics of the deviation index.

[0009] Furthermore, the step of performing adaptive filtering according to the instantaneous fluctuation characteristics of the load time series to obtain the load filtering time series includes:

[0010] Calculate the absolute value of the difference between any moment and the previous adjacent moment in the load time series to obtain a first value; calculate the absolute value of the difference between the arbitrary moment and the next adjacent moment to obtain a second value; calculate the average value of the first value and the second value to obtain a fluctuation characteristic value; calculate the absolute value of the difference between the previous adjacent moment and the next adjacent moment to obtain a third value; calculate the ratio of the fluctuation characteristic value to the third value and normalize it to obtain an adjustment value; calculate the product of the adjustment value and the preset window length and take the nearest odd number to obtain an adaptive window, perform median filtering according to different adaptive windows at different moments in the load time series to obtain the load filtering time series.

[0011] Furthermore, the step of constructing a load feature vector according to the data trend characteristics of the preset current period in the load filtering time series includes:

[0012] Calculate the load average value of the preset current period and normalize it to obtain a first load characteristic value; calculate the average value of the load difference between adjacent moments in the preset current period and normalize it to obtain a second load characteristic value; calculate the average value of the absolute value of the load difference between adjacent moments in the preset current period and normalize it to obtain a third load characteristic value; construct the load characteristic vector based on the first load characteristic value, the second load characteristic value and the third load characteristic value.

[0013] Furthermore, the step of obtaining different temperature difference degrees and date clusters based on the difference characteristics of the temperature feature vectors in the same historical period includes:

[0014] The cosine similarity of the temperature feature vectors of any two identical historical periods is calculated and negatively correlated to obtain the temperature difference between the two identical historical periods. Different historical dates are clustered using the mean shift clustering algorithm based on the temperature difference to obtain different date clusters.

[0015] Furthermore, the step of constructing a difference scatter plot and a load difference prediction model according to the distribution characteristics of the temperature difference and the load difference corresponding to the date clusters includes:

[0016] A rectangular coordinate system is constructed, with the horizontal axis representing temperature difference and the vertical axis representing load difference; coordinate points are generated in the rectangular coordinate system according to the temperature difference and load difference of the same period of any two dates in the date cluster to obtain a difference scatter plot; and a load difference prediction model is constructed in the date cluster using the SVR algorithm to predict the load difference based on the temperature difference.

[0017] Furthermore, the step of obtaining a difference weight according to the scatter distribution characteristics in the difference scatter plot and training a load difference prediction model includes:

[0018] Calculate the ratio of the temperature difference and the load difference of the scatter point in the difference scatter plot to obtain the ratio coefficient of the scatter point; calculate the absolute value of the difference between the ratio coefficients of the scatter point and other scatter points to obtain the ratio difference; calculate the minimum value of the absolute value of the difference between the temperature difference and the absolute value of the load difference between the scatter point and the other scatter points to obtain a difference characterization value; calculate the ratio of the ratio difference to the difference characterization value to obtain the degree of difference between the scatter point and the other scatter points; calculate the average value of the degree of difference between the scatter point and all other scatter points to obtain the average degree of difference of the scatter point; calculate the proportion of the average degree of difference of the scatter point in all scatter points and negatively correlate and map them to obtain the difference weight of the scatter point;

[0019] The temperature difference value corresponding to the scatter point is calculated by calculating the square of the difference between the predicted value obtained by the load difference prediction model and the actual load difference to obtain the prediction difference; the average value of the product of the prediction difference of all scatter points and the difference weight is calculated to obtain the comprehensive prediction difference; the load difference prediction model trained when the comprehensive prediction difference is the smallest is used as the final load difference prediction model.

[0020] Furthermore, the step of obtaining the load prediction difference by using the load difference prediction model according to the temperature characteristic vector of the preset current period includes:

[0021] The date cluster in which the temperature characteristic vector of the preset current period is located is determined, and the temperature characteristic vectors of the two periods corresponding to the scatter points with the largest difference weight in the date cluster in which the preset current period is located are used as temperature target vectors; the average value of the temperature difference between the temperature characteristic vector of the preset current period and the temperature target vector is calculated to obtain an input value; and the load prediction difference is obtained through the final load difference prediction model based on the input value.

[0022] Furthermore, the step of obtaining the deviation index at the current moment according to the load forecast difference, the actual load difference in the preset current period, and the model prediction error characteristics includes:

[0023] The load characteristic vectors of the two time periods corresponding to the scatter points with the largest difference weight in the date cluster where the preset current time period is located are used as load target vectors; the average value of the load difference between the load characteristic vector of the preset current time period and the load target vector is calculated to obtain the actual load difference; the square of the difference between the actual load difference and the load forecast difference is calculated to obtain the degree of deviation; the ratio of the degree of deviation to the minimum value of the comprehensive forecast difference is calculated to obtain the deviation index at the current moment.

[0024] Furthermore, the step of monitoring the load state at the current moment according to the variation characteristics of the deviation index includes:

[0025] Calculate the average difference between the deviation index of each moment and the next moment in the preset adjacent time period of the current moment and normalize them to obtain the deviation change degree; normalize the deviation index of the current moment to obtain the deviation characterization value; calculate the average value of the deviation characterization value and the deviation change degree to obtain the load abnormality; when the load abnormality exceeds the preset abnormality threshold, the load state at the current moment is abnormal.

[0026] The present invention has the following beneficial effects:

[0027] In the present invention, obtaining a load filtering time series can adaptively weaken transient fluctuation characteristics while preserving normal load fluctuation characteristics, thereby improving the accuracy of subsequent load analysis and monitoring. Obtaining a load feature vector can characterize multiple load characteristics over a period of time; obtaining a temperature feature vector can characterize multiple temperature characteristics over a period of time, thereby enabling load status monitoring based on the load and temperature feature vectors. Obtaining temperature variability can characterize the degree of temperature variability between historically identical time periods; obtaining date clusters can cluster dates with similar temperature characteristics. Because weather temperature significantly affects load magnitude, analyzing the relationship between temperature and load characteristics within a cluster can improve the accuracy of subsequent predictions and load monitoring. Obtaining load variability can characterize the degree of load variability between historically identical time periods; obtaining a scatter plot can determine the distribution of temperature and load variability between the same time periods within a date cluster; obtaining a load variability prediction model can be used to predict the load variability for a preset current time period, thereby enabling load status monitoring; obtaining variability weights can train and optimize the load variability prediction model, improving model prediction accuracy. Obtaining the load forecast difference can characterize the relatively normal load difference degree between the preset current period and the same period in history under the same temperature conditions; obtaining the deviation index can characterize the deviation of the load status at the current moment from the same period in history. Finally, the load status at the current moment is monitored according to the change characteristics of the deviation index, so that the load status monitoring at the current moment takes into account the dynamic changes and regular characteristics of the load characteristics under the same temperature conditions in the same period in history, thereby improving the accuracy of the load status monitoring of the ring network box. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0029] Figure 1 This is a block diagram of a ring network box module with overload fault monitoring and circuit breaker function provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0030] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a ring main unit (RME) with overload fault monitoring and circuit breaking functions. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0031] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0032] The following describes in detail a specific solution of a ring network box with overload fault monitoring and circuit breaker function provided by the present invention with reference to the accompanying drawings.

[0033] See also Figure 1 , which shows a module block diagram of a ring network box with an overload fault monitoring and circuit breaker function provided by an embodiment of the present invention. The ring network box includes the following modules:

[0034] The data acquisition module S1 is used to obtain the load timing of the circuit and the temperature timing of the environment.

[0035] In an embodiment of the present invention, the implementation scenario is to monitor the load status of the circuit through the ring network box to improve the accuracy of load monitoring and control; first, the load timing of the circuit and the temperature timing of the environment are obtained. The load timing is constructed by the power of the circuit passing through the ring network box, and the temperature is the outdoor temperature of the area where the circuit is located; in an embodiment of the present invention, the load timing and temperature timing within one month before the current moment are collected as the analysis range. The implementer can determine the data collection duration and frequency according to the implementation scenario.

[0036] The feature analysis module S2 is used to perform adaptive filtering based on the instantaneous fluctuation characteristics of the load time series to obtain the load filtering time series; construct a load feature vector based on the data trend characteristics of the current period preset in the load filtering time series; and construct a temperature feature vector based on the data trend characteristics of the current period preset in the temperature time series.

[0037] During circuit overload fault monitoring, if the grid load experiences rapid increases and decreases within a short period of time, such as during large equipment startup or arc welding, the ring network box monitoring system can easily misinterpret these non-sustained fluctuations as overload conditions, triggering false alarms or unnecessary protection mechanisms. To improve monitoring accuracy and anti-interference capabilities, and enhance the ability to identify true overload faults, the transient characteristics of the collected load time series are first analyzed. Median filtering is used to effectively remove these fluctuations while retaining the normal fluctuation characteristics of the grid load. Therefore, adaptive filtering is performed based on the transient fluctuation characteristics of the load time series to obtain the load filter time series.

[0038] Preferably, in an embodiment of the present invention, the step of obtaining the load filtering time series includes: calculating the absolute value of the difference between any moment and the previous adjacent moment in the load time series to obtain a first value; calculating the absolute value of the difference between the arbitrary moment and the next adjacent moment to obtain a second value; calculating the average of the first value and the second value to obtain a fluctuation characteristic value; the first value and the second value characterize the instantaneous fluctuation degree at the arbitrary moment, and the larger the fluctuation characteristic value, the more likely it is that the load will rise or fall sharply at the arbitrary moment. Calculating the absolute value of the difference between the previous adjacent moment and the next adjacent moment to obtain a third value; the third value characterizes the load difference characteristics before and after the arbitrary moment. Calculating the ratio of the fluctuation characteristic value to the third value and normalizing it to obtain an adjustment value; it should be noted that in all calculation processes in the embodiment of the present invention, if the denominator is 0, for example, the third value is 0, then a preset minimum positive number is added to the denominator to avoid the denominator being 0. In the embodiment of the present invention, the preset minimum positive number is 0.01. When the fluctuation characteristic value is larger and the third value is smaller, it means that the instantaneous fluctuation characteristics at any moment are more obvious, the adjustment value is larger, and the window for median filtering at any moment is larger, which can more effectively remove the instantaneous fluctuation characteristics at any moment. Calculate the product of the adjustment value and the preset window length and take the nearest odd number to obtain an adaptive window. In the embodiment of the present invention, the preset window length is 11 acquisition moments, and the implementer can determine it according to the implementation scenario. When the adaptive window is larger, it means that more adjacent data are needed to perform median filtering at any moment, smooth the instantaneous fluctuation characteristics, and improve the accuracy of subsequent load state analysis. Median filtering is performed according to different adaptive windows at different moments in the load time series to obtain a load filtering time series. It should be noted that median filtering belongs to the existing technology and the specific steps will not be repeated. The load filtering time series can filter the instantaneous fluctuation characteristics of the load and retain the characteristics of normal load fluctuations, which can enhance the ability to identify real overload conditions.

[0039] Furthermore, weather temperature data is a crucial reference factor in overload monitoring, as temperature fluctuations significantly impact power load fluctuations, especially under conditions such as high summer temperatures or severe winter cold. For example, during periods of sustained high temperatures, the extensive use of high-power equipment like air conditioners can lead to a significant increase in power load in residential and commercial areas, potentially causing equipment to operate near or above rated load for extended periods, increasing the risk of overload failures. Introducing temperature as a factor in load characteristic analysis can help identify the normal load fluctuation range under different temperature conditions. When the latest load data deviates significantly from the historical distribution under current temperature conditions, for example, the load is significantly higher than the load characteristics under similar temperatures in the past, it can be judged as a potential overload trend. First, the load characteristics are determined, and a load characteristic vector is constructed based on the data trend characteristics preset for the current time period in the load filtering time series.

[0040] Preferably, in an embodiment of the present invention, the step of obtaining a load characteristic vector includes: calculating and normalizing the load average value for a preset current time period to obtain a first load characteristic value; in an embodiment of the present invention, the preset current time is within three hours before the current time period, which is used to reflect the local load characteristics before the current time period, and the implementer can determine it according to the implementation scenario; the first load characteristic value represents the average load level of the time period; a larger value indicates long-term continuous high-load operation. Calculating and normalizing the average value of the load difference between adjacent moments in the preset current time period to obtain a second load characteristic value; the load difference between adjacent moments is the load difference between any moment and the previous moment; the second load characteristic value represents whether the load shows a continuous upward trend; the larger the second load characteristic value, the more obvious the load upward trend in the time period. Calculating and normalizing the average value of the absolute value of the load difference between adjacent moments in the preset current time period to obtain a third load characteristic value; the third load characteristic value represents the degree of load fluctuation in the time period; the larger the third load characteristic value, the more obvious the load fluctuation characteristics in the time period. A load characteristic vector is constructed according to the first load characteristic value, the second load characteristic value, and the third load characteristic value; the load characteristic vector represents a plurality of load characteristics of a preset current time period.

[0041] Similarly, a temperature characteristic vector is constructed based on the data trend characteristics of the current time period preset in the temperature time series; it should be noted that the steps for obtaining the temperature characteristic vector and the load characteristic vector are the same, and the first temperature characteristic value, the second temperature characteristic value and the third temperature characteristic value are obtained based on the temperature time series preset for the current time period to construct the temperature characteristic vector. The specific steps will not be repeated here; the temperature characteristic vector represents a variety of temperature characteristics preset at the current moment.

[0042] Feature processing module S3 is used to obtain different temperature differences and date clusters based on the difference characteristics of temperature feature vectors in the same historical period; obtain load differences based on the difference characteristics of load feature vectors in the same historical period; construct a difference scatter plot and a load difference prediction model based on the distribution characteristics of temperature differences and load differences corresponding to the date clusters; obtain difference weights based on the scatter distribution characteristics in the difference scatter plot and train the load difference prediction model.

[0043] Since weather temperature is an important factor affecting load differences, it is necessary to analyze load characteristic differences under similar weather temperature conditions. Therefore, different temperature differences and date clusters are obtained based on the difference characteristics of the temperature feature vectors of the same historical period. Preferably, in an embodiment of the present invention, the step of obtaining the temperature difference and date cluster includes: calculating the cosine similarity of the temperature feature vectors of any two same historical periods and performing negative correlation mapping to obtain the temperature difference of the arbitrary two same historical periods. It should be noted that cosine similarity belongs to the prior art, and the specific calculation steps are not repeated. The more similar the two temperature feature vectors are, the greater the cosine similarity is, which means that the temperature features of the two same periods are more similar and the temperature difference is smaller. Based on the temperature difference, different historical dates are clustered using the mean shift clustering algorithm to obtain different date clusters. It should be noted that the mean shift clustering algorithm belongs to the prior art, and the specific steps are not repeated. The smaller the temperature difference between the two temperature feature vectors, the more likely the dates corresponding to the two temperature feature vectors are in the same date cluster. The temperature feature vectors in each date cluster are similar, and the weather temperature conditions of the same period of the corresponding dates are similar.

[0044] Furthermore, after obtaining different date clusters, it is necessary to analyze the differences in load characteristic vectors corresponding to different dates in the date cluster, so as to subsequently determine whether an overload situation occurs at the current moment; therefore, the load difference is obtained based on the difference characteristics of the load characteristic vectors in the same historical period. It should be noted that the load difference is calculated in the same way as the temperature difference, and is obtained through the cosine similarity and negative correlation mapping between the load characteristic vectors. The greater the load difference, the more obvious the difference in load characteristics between the two same periods.

[0045] If the temperature characteristics of the same time period on different dates within the same date cluster are similar, then the load characteristics of electricity consumption should also be similar, and thus the load characteristic differences can be predicted based on the temperature characteristic differences. Therefore, a difference scatter plot and a load difference prediction model are constructed based on the distribution characteristics of the temperature difference and load difference corresponding to the date cluster. Preferably, in an embodiment of the present invention, the steps of constructing the difference scatter plot and the load difference prediction model include: constructing a rectangular coordinate system with the horizontal axis being the temperature difference and the vertical axis being the load difference; generating coordinate points in the rectangular coordinate system based on the temperature difference and load difference of the same time period on any two dates in the date cluster to obtain a difference scatter plot; the horizontal and vertical coordinate values ​​of each scatter point in the difference scatter plot represent the temperature characteristic difference and load characteristic difference corresponding to the same time period on the two dates. The positions of the scatter points in the difference scatter plot corresponding to the same date cluster are relatively similar, and thus a load difference prediction model can be constructed in the date cluster using the SVR algorithm to predict the load difference based on the temperature difference. It should be noted that the SVR support vector regression algorithm belongs to the existing technology, and the algorithm can perform data prediction. The specific steps will not be repeated here. The load difference prediction model can predict the load difference of the scatter point's vertical coordinate based on the temperature difference of the scatter point's horizontal coordinate.

[0046] Furthermore, to improve the accuracy of the prediction model, avoid the model falling into a local optimum or the distribution of some scattered points affecting overall prediction accuracy, and ensure the accuracy of the final overload status monitoring, it is necessary to train and optimize the load difference prediction model. Therefore, a difference weight is obtained based on the distribution characteristics of the scattered points in the difference scatter plot, and the load difference prediction model is trained. Preferably, in an embodiment of the present invention, the steps of obtaining the difference weight and training the load difference prediction model include: calculating the ratio of the temperature difference and the load difference of the scattered points in the difference scatter plot to obtain the ratio coefficient of the scattered point; within the same date cluster, since the temperature feature vector and the load feature vector are similar, the ratio coefficients of most scattered points are relatively close. The absolute value of the difference between the ratio coefficients of the scattered point and other scattered points is calculated to obtain the ratio difference; a larger ratio difference indicates a larger difference in the ratio coefficients corresponding to the two scattered points. The minimum value of the absolute value of the difference between the temperature difference and the absolute value of the difference between the scattered point and the other scattered points is calculated to obtain a difference representation value; a smaller difference representation value indicates a more similar temperature difference and load difference between the two scattered points. Calculate the ratio of the ratio difference to the difference representation value to obtain the degree of difference between the scatter point and the other scatter points. The difference representation value serves as the weight for calculating the degree of difference between the two scatter points. The more similar the horizontal or vertical axes of the two scatter points are, the more important the ratio difference result is. Calculate the average of the degree of difference between the scatter point and all other scatter points to obtain the average degree of difference for the scatter point. If the distribution location of the scatter point is close to that of the other scatter points, the smaller the ratio difference, the smaller the average degree of difference. If the distribution location of the scatter point is significantly different from that of the other scatter points, the larger the ratio difference, the larger the average degree of difference, and the greater the difference between the temperature difference and load difference represented by the scatter point and the other scatter points, the lower the data reliability of the scatter point during the prediction process. Calculate the proportion of the scatter point's average degree of difference among all scatter points and negatively correlate them to obtain the degree of difference weight for the scatter point. A larger degree of difference weight means that the temperature difference and load difference corresponding to the scatter point are more common, and its importance in the prediction model training process is higher.

[0047] After obtaining the difference weight for each scatter point in the difference scatter plot corresponding to the date cluster, the predicted difference is calculated by squaredly comparing the temperature difference value corresponding to the scatter point, as predicted by the load difference prediction model, with the actual load difference. The greater the difference between the predicted value and the actual load difference, the greater the prediction difference and the worse the prediction effect. The average of the predicted differences multiplied by the difference weight for all scatter points is calculated to obtain the comprehensive predicted difference. The difference weight weights weight the predicted differences to prevent the temperature and load differences corresponding to scatter points with relatively unusual distributions in the scatter plot from affecting the prediction accuracy. Consequently, the larger the difference weight, the more significant the predicted difference of the scatter point. The load difference prediction model trained when the comprehensive predicted difference is minimized is used as the final load difference prediction model. This load difference prediction model achieves the highest prediction accuracy, thereby improving the accuracy of monitoring the load status at the latest moment.

[0048] The load monitoring module S4 is used to obtain the load prediction difference through the load difference prediction model based on the temperature characteristic vector of the preset current time period; obtain the deviation index at the current moment based on the load prediction difference and the actual load difference of the preset current time period and the model prediction error characteristics; and monitor the load status at the current moment based on the change characteristics of the deviation index.

[0049] After obtaining the load difference prediction model for the date cluster, the load prediction difference can be obtained using the load difference prediction model based on the temperature feature vector of the preset current period. Preferably, in an embodiment of the present invention, the step of obtaining the load prediction difference includes: determining the date cluster based on the temperature feature vector of the preset current period, clustering the temperature feature vector of the preset current period based on the temperature difference between the temperature feature vector of the preset current period and the temperature feature vector of the same historical period to determine the date cluster. The specific steps are not repeated here. The temperature feature vectors of the two time periods corresponding to the scatter point with the largest difference weight in the date cluster of the preset current period are used as the temperature target vectors. The scatter point with the largest difference weight means that the temperature difference and load difference with all other scatter points in the date cluster are closest, and can represent the comprehensive temperature difference level and load difference level in the date cluster. The temperature feature vectors of the two time periods corresponding to the scatter point are then used as the temperature target vectors. The temperature target vector represents the representative temperature characteristics of the date cluster. The average temperature difference between the temperature feature vector of the preset current period and the temperature target vector is calculated to obtain an input value; and the load prediction difference is obtained using the final load difference prediction model based on the input value.

[0050] Furthermore, after obtaining the load forecast difference at the preset current time, a deviation index for the current time can be obtained based on the load forecast difference, the actual load difference during the preset current time period, and the model prediction error characteristics. Preferably, in an embodiment of the present invention, the step of obtaining the deviation index includes: using the load characteristic vectors of the two time periods corresponding to the scatter points with the largest difference weights in the date cluster in which the preset current time period is located as load target vectors; the load target vectors represent representative load characteristics in the date cluster, and the two time periods are also the two time periods referenced in the temperature difference prediction process. Calculating the average of the load difference between the load characteristic vector and the load target vector for the preset current time period to obtain the actual load difference; the actual load difference represents the difference between the load characteristic vector and the load target vector for the preset current time period. Calculating the square of the difference between the actual load difference and the load forecast difference to obtain the degree of deviation; the more abnormal the load characteristic vector and load status during the preset current time period, the greater the difference between the actual load difference and the load forecast difference, and the greater the degree of deviation. Calculate the ratio of the degree of deviation to the minimum value of the comprehensive prediction difference to obtain the deviation index at the current moment; the comprehensive prediction difference represents the prediction accuracy of the load difference prediction model. The smaller the value of the comprehensive prediction difference, the higher the prediction accuracy and the more reliable the deviation result. If the degree of deviation is greater at this time, the deviation index is larger.

[0051] Since the deviation index at the current moment is obtained based on the load characteristics of the preset current time period, if there is a significant overload situation in the recent time period before the current moment, the value of the deviation index will gradually increase. In order to improve the accuracy of load status monitoring, the load status at the current moment can be monitored based on the change characteristics of the deviation index. Preferably, in an embodiment of the present invention, the steps include: calculating the average difference between the deviation index of each moment and the next moment in the preset adjacent time period of the current moment and normalizing it to obtain the deviation change degree. In this embodiment of the present invention, the preset adjacent time period is the period 30 minutes before the current moment, and the implementer can determine it according to the implementation scenario. The larger the deviation change degree, the higher the deviation index is, which means that there is an upward trend in the preset adjacent time period, and the abnormal trend of the load characteristics gradually increases. The deviation index at the current moment is normalized to obtain a deviation characterization value. The average of the deviation characterization value and the deviation change degree is calculated to obtain the load abnormality degree. The larger the load abnormality degree is, the load characteristics have had a gradual abnormal change trend before and the current abnormality is more obvious. When the load anomaly exceeds the preset anomaly threshold, the current load state is abnormal, and the ring network box needs to promptly activate the early warning and protection mechanism. In the embodiment of the present invention, the preset anomaly threshold is 0.4, which can be determined by the implementer based on the implementation scenario. The embodiment of the present invention monitors the load state through the load anomaly degree, taking into account the impact of weather and temperature factors on load characteristics, and analyzes load differences under similar temperature conditions, and determines whether there is a load anomaly based on the load differences. This allows the current load state monitoring to combine the load changes and patterns of the same historical period, thereby improving the accuracy of load state monitoring.

[0052] In summary, the embodiments of the present invention provide a ring network box with an overload fault monitoring and circuit breaker function; obtain a load filtering time series based on the instantaneous fluctuation characteristics of the load time series; construct a load characteristic vector based on the load filtering time series; construct a temperature characteristic vector based on the temperature time series; obtain temperature differences and date clusters based on the temperature characteristic vector; obtain load differences based on the load characteristic vector; construct a difference scatter plot and a load difference prediction model based on the temperature differences and load differences corresponding to the date clusters; obtain difference weights based on the scatter distribution characteristics and train the load difference prediction model. Obtain a load prediction difference based on the temperature characteristic vector of the preset current time period; obtain a deviation index based on the load prediction difference, the actual load difference, and the model prediction error characteristics; monitor the load status based on the change characteristics of the deviation index, thereby improving the accuracy of load monitoring.

[0053] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0054] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A ring network box with overload fault monitoring and circuit breaker function, characterized in that: The ring network box includes the following modules: A data acquisition module is used to obtain the load timing of the circuit and the temperature timing of the environment; A feature analysis module, configured to perform adaptive filtering based on the instantaneous fluctuation characteristics of the load time series to obtain a load filtering time series; Constructing a load feature vector based on the data trend characteristics of the current period preset in the load filtering time series; constructing a temperature feature vector based on the data trend characteristics of the current period preset in the temperature time series; A feature processing module is configured to obtain different temperature differences and date clusters based on the difference characteristics of temperature feature vectors in the same historical period; obtain load differences based on the difference characteristics of load feature vectors in the same historical period; construct a difference scatter plot and a load difference prediction model based on the distribution characteristics of the temperature differences and load differences corresponding to the date clusters; obtain difference weights based on the scatter distribution characteristics of the scatter plot and train the load difference prediction model; The load monitoring module is used to obtain the load prediction difference through the load difference prediction model based on the temperature characteristic vector of the preset current time period; obtain the deviation index at the current moment based on the load prediction difference and the actual load difference of the preset current time period and the model prediction error characteristics; and monitor the load status at the current moment based on the change characteristics of the deviation index.

2. The ring main box with overload fault monitoring and circuit breaker function according to claim 1, characterized in that: The step of performing adaptive filtering according to the instantaneous fluctuation characteristics of the load time series to obtain the load filtering time series includes: Calculate the absolute value of the difference between any moment and the previous adjacent moment in the load time series to obtain a first value; calculate the absolute value of the difference between the arbitrary moment and the next adjacent moment to obtain a second value; calculate the average value of the first value and the second value to obtain a fluctuation characteristic value; calculate the absolute value of the difference between the previous adjacent moment and the next adjacent moment to obtain a third value; calculate the ratio of the fluctuation characteristic value to the third value and normalize it to obtain an adjustment value; calculate the product of the adjustment value and the preset window length and take the nearest odd number to obtain an adaptive window, perform median filtering according to different adaptive windows at different moments in the load time series to obtain the load filtering time series.

3. The ring main box with overload fault monitoring and circuit breaker function according to claim 1, characterized in that: The step of constructing a load feature vector according to the data trend characteristics of the current period preset in the load filtering time series includes: Calculate the load average value of the preset current period and normalize it to obtain a first load characteristic value; calculate the average value of the load difference between adjacent moments in the preset current period and normalize it to obtain a second load characteristic value; calculate the average value of the absolute value of the load difference between adjacent moments in the preset current period and normalize it to obtain a third load characteristic value; construct the load characteristic vector based on the first load characteristic value, the second load characteristic value and the third load characteristic value.

4. The ring main box with overload fault monitoring and circuit breaker function according to claim 1, characterized in that: The step of obtaining different temperature difference degrees and date clusters based on the difference characteristics of the temperature feature vectors of the same historical period includes: The cosine similarity of the temperature feature vectors of any two identical historical periods is calculated and negatively correlated to obtain the temperature difference between the two identical historical periods. Different historical dates are clustered using the mean shift clustering algorithm based on the temperature difference to obtain different date clusters.

5. The ring main box with overload fault monitoring and circuit breaker function according to claim 1, characterized in that: The step of constructing a difference scatter plot and a load difference prediction model according to the distribution characteristics of the temperature difference and the load difference corresponding to the date clusters includes: A rectangular coordinate system is constructed, with the horizontal axis representing temperature difference and the vertical axis representing load difference; coordinate points are generated in the rectangular coordinate system according to the temperature difference and load difference of the same period of any two dates in the date cluster to obtain a difference scatter plot; and a load difference prediction model is constructed in the date cluster using the SVR algorithm to predict the load difference based on the temperature difference.

6. The ring main box with overload fault monitoring and circuit breaker function according to claim 5, characterized in that: The step of obtaining a difference weight according to the scatter distribution characteristics in the difference scatter plot and training a load difference prediction model comprises: Calculate the ratio of the temperature difference and the load difference of the scatter point in the difference scatter plot to obtain the ratio coefficient of the scatter point; calculate the absolute value of the difference between the ratio coefficients of the scatter point and other scatter points to obtain the ratio difference; calculate the minimum value of the absolute value of the difference between the temperature difference and the absolute value of the load difference between the scatter point and the other scatter points to obtain a difference characterization value; calculate the ratio of the ratio difference to the difference characterization value to obtain the degree of difference between the scatter point and the other scatter points; calculate the average value of the degree of difference between the scatter point and all other scatter points to obtain the average degree of difference of the scatter point; calculate the proportion of the average degree of difference of the scatter point in all scatter points and negatively correlate and map them to obtain the difference weight of the scatter point; The temperature difference value corresponding to the scatter point is calculated by calculating the square of the difference between the predicted value obtained by the load difference prediction model and the actual load difference to obtain the prediction difference; the average value of the product of the prediction difference of all scatter points and the difference weight is calculated to obtain the comprehensive prediction difference; the load difference prediction model trained when the comprehensive prediction difference is the smallest is used as the final load difference prediction model.

7. The ring main box with overload fault monitoring and circuit breaker function according to claim 6, characterized in that: The step of obtaining the load prediction difference by using the load difference prediction model according to the temperature characteristic vector of the preset current period includes: The date cluster in which the temperature characteristic vector of the preset current period is located is determined, and the temperature characteristic vectors of the two periods corresponding to the scatter points with the largest difference weight in the date cluster in which the preset current period is located are used as temperature target vectors; the average value of the temperature difference between the temperature characteristic vector of the preset current period and the temperature target vector is calculated to obtain an input value; and the load prediction difference is obtained through the final load difference prediction model based on the input value.

8. The ring main box with overload fault monitoring and circuit breaker function according to claim 7, characterized in that: The step of obtaining the deviation index at the current moment according to the load prediction difference, the actual load difference in the preset current period, and the model prediction error characteristics includes: The load characteristic vectors of the two time periods corresponding to the scatter points with the largest difference weight in the date cluster where the preset current time period is located are used as load target vectors; the average value of the load difference between the load characteristic vector of the preset current time period and the load target vector is calculated to obtain the actual load difference; the square of the difference between the actual load difference and the load forecast difference is calculated to obtain the degree of deviation; the ratio of the degree of deviation to the minimum value of the comprehensive forecast difference is calculated to obtain the deviation index at the current moment.

9. The ring main box with overload fault monitoring and circuit breaker function according to claim 1, characterized in that: The step of monitoring the load state at the current moment according to the variation characteristics of the deviation index includes: Calculate the average difference between the deviation index of each moment and the next moment in the preset adjacent time period of the current moment and normalize them to obtain the deviation change degree; normalize the deviation index of the current moment to obtain the deviation characterization value; calculate the average value of the deviation characterization value and the deviation change degree to obtain the load abnormality; when the load abnormality exceeds the preset abnormality threshold, the load state at the current moment is abnormal.

Images