Power distribution network electricity approaching monitoring and early warning method and system
By collecting and fusion of environmental data in the distribution network environmental monitoring and early warning system, using machine learning algorithms to perform abnormal detection, and dynamically adjusting the threshold, the problem of insufficient data fusion quality and threshold setting in the prior art is solved, and the accuracy and real-time monitoring are improved.
Patent Information
- Application Number
- CN202411650057.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-19
- Publication Date
- 2025-02-28
AI Technical Summary
The existing distribution network environmental monitoring and early warning technology has shortcomings in data fusion quality and dynamic adjustment of thresholds, resulting in insufficient monitoring accuracy and real-timeness.
By collecting environmental monitoring data from key nodes of the distribution network, pre-processing and fusion, abnormal detection is performed using algorithms such as random forest models and support vector machines, and the threshold is dynamically adjusted based on historical data and real-time abnormal signal strength to achieve real-time early warning.
It improves the data fusion quality and flexibility of threshold setting of the distribution network environmental monitoring and early warning system, improves the accuracy and real-time monitoring, and ensures the safe and stable operation of the distribution network.
Smart Images
Figure CN119538155A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power system monitoring and early warning, and in particular to a distribution network near-power monitoring and early warning method and system. Background Art
[0002] With the continuous development of power systems and the improvement of their intelligence level, the safe and stable operation of distribution networks has become a focus of attention in the power industry. As an important part of the power system, the distribution network is responsible for distributing electricity from the transmission network to user terminals. However, due to the wide coverage of the distribution network, the variety of equipment, and the complex operating environment, it is easily affected by various external factors, such as electromagnetic interference, temperature changes, humidity fluctuations, etc., which lead to equipment failures and safety accidents. Therefore, environmental monitoring and abnormal warning of key nodes in the distribution network are particularly important. In recent years, with the rapid development of sensor technology, data processing technology, and machine learning algorithms, the environmental monitoring and early warning technology of distribution networks has made significant progress. Traditional monitoring methods mainly rely on manual inspections and fixed threshold alarms, which have problems such as slow response speed, high false alarm rate, and inability to monitor in real time. Modern monitoring technology integrates multiple sensors to collect environmental data in real time, and uses advanced data processing and analysis methods to improve the accuracy and real-time nature of monitoring. However, the existing distribution network environmental monitoring and early warning technology still has some shortcomings.
[0003] First, the existing monitoring systems have limitations in data fusion. Traditional data fusion methods usually use simple weighted average or linear combination, which fails to fully consider the correlation and nonlinear characteristics between the sensor data, resulting in low quality of fused data and affecting the accuracy of anomaly detection. Secondly, the existing anomaly detection models have defects in threshold setting and adjustment. Most systems use fixed thresholds, which cannot adapt to changes in different environments and working conditions and are prone to false alarms or missed alarms. In addition, the existing threshold adjustment methods are usually based on single historical data, lack a dynamic adjustment mechanism, and are difficult to reflect the current system operation status in real time. These problems limit the practicality and reliability of the distribution network environmental monitoring and early warning system. Summary of the invention
[0004] In view of the above existing problems, the present invention is proposed.
[0005] Therefore, the present invention provides a distribution network near-power monitoring and early warning method and system to solve the problems of data fusion quality and threshold dynamic adjustment, and improve the accuracy and real-time performance of distribution network environment monitoring and early warning.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0007] In a first aspect, the present invention provides a distribution network near-power monitoring and early warning method, which includes collecting environmental monitoring data of key nodes of the distribution network, preprocessing the collected environmental monitoring data; fusing the preprocessed environmental monitoring data; constructing an anomaly detection model, inputting the fused environmental monitoring data, and identifying abnormal signals under normal working conditions; setting an initial threshold based on historical data, and adjusting the initial threshold according to the strength of the abnormal signal; and providing real-time early warning through the adjusted initial threshold.
[0008] As a preferred solution of the distribution network near-power monitoring and early warning method described in the present invention, the environmental monitoring data includes electromagnetic field strength, current, voltage, ambient temperature, ambient humidity, equipment surface temperature, thermal imaging images and images of surrounding objects.
[0009] As a preferred solution of the distribution network near-power monitoring and early warning method described in the present invention, the preprocessing includes data cleaning, data conversion, data standardization and feature engineering.
[0010] As a preferred solution of the power distribution network near-electricity monitoring and early warning method of the present invention, the specific steps of fusing the pre-processed environmental monitoring data are as follows:
[0011] Use the preprocessed environmental monitoring data to train a random forest model;
[0012] According to the Gini importance, the Gini index reduction of each feature on all nodes that use the feature is calculated, and the Gini index reduction of each feature is summed to obtain the overall importance score of the feature. After normalization, the feature importance score is obtained;
[0013] A nonlinear function is introduced to enhance the nonlinear characteristics of the feature. The weighting coefficient is calculated according to the feature importance score and the nonlinear characteristics. Based on the weighting coefficient and the nonlinear characteristics, the fused environmental monitoring data S is calculated.
[0014] As a preferred solution of the distribution network near-power monitoring and early warning method of the present invention, wherein: the abnormal detection model is constructed, the fused environmental monitoring data is input, and the abnormal signal under normal working state is identified. The specific steps are as follows:
[0015] The support vector machine is selected as the anomaly detection model, and the fused environmental monitoring data is input to identify abnormal signals under normal working conditions.
[0016] As a preferred solution of the distribution network near-power monitoring and early warning method of the present invention, wherein: the initial threshold is set based on historical data, and the initial threshold is adjusted according to the abnormal signal strength. The specific steps are as follows:
[0017] The initial threshold T0 is set according to historical data, and the initial threshold is adjusted according to the abnormal signal strength A.
[0018] As a preferred solution of the distribution network near-power monitoring and early warning method of the present invention, wherein: the real-time early warning is performed through the adjusted initial threshold, and the specific steps are as follows:
[0019] By setting a time window, calculating the abnormal signal strength at each time point in the time window, and comparing the calculated abnormal signal strength with the adjusted threshold, a multi-time point judgment is performed;
[0020] When the abnormal signal strength is greater than the adjusted threshold, the current state is considered abnormal, an early warning is triggered, and an early warning message is output.
[0021] In the second aspect, the present invention provides a distribution network near-power monitoring and early warning system, including: a data acquisition module, which collects environmental monitoring data of key nodes of the distribution network and preprocesses the collected environmental monitoring data; a data fusion module, which fuses the preprocessed environmental monitoring data; an anomaly detection module, which constructs an anomaly detection model, inputs the fused environmental monitoring data, and identifies abnormal signals under normal working conditions; a threshold management module, which sets an initial threshold based on historical data and adjusts the initial threshold according to the strength of the abnormal signal; and a real-time early warning module, which performs real-time early warning through the adjusted initial threshold.
[0022] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the distribution network near-power monitoring and early warning method as described in the first aspect of the present invention is implemented.
[0023] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the distribution network near-power monitoring and early warning method as described in the first aspect of the present invention is implemented.
[0024] The beneficial effects of the present invention are as follows: by improving the data fusion method and the dynamic threshold adjustment mechanism, the deficiencies of the prior art in data fusion quality and threshold setting are solved. Specifically, the data fusion step improves the quality and reliability of the fused data through scientific weight distribution and the introduction of nonlinear functions; the dynamic threshold adjustment step realizes the dynamic adjustment of the threshold through the combination of historical data and real-time abnormal signal strength, thereby improving the flexibility and adaptability of the system. These innovative steps jointly improve the accuracy and real-time performance of the distribution network near-power monitoring and early warning system, ensuring the safe and stable operation of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0026] Figure 1 This is a flow chart of the distribution network near-power monitoring and early warning method in Example 1.
[0027] Figure 2 This is a flow chart of the distribution network near-power monitoring and early warning system in Example 1. DETAILED DESCRIPTION
[0028] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.
[0029] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a distribution network near-power monitoring and early warning method, comprising the following steps:
[0030] Collect environmental monitoring data from key nodes of the distribution network and pre-process the collected environmental monitoring data;
[0031] Furthermore, electromagnetic field strength, current and voltage are collected through electromagnetic induction sensors;
[0032] Collect ambient temperature and equipment surface temperature through temperature sensors;
[0033] Collect environmental humidity through humidity sensor;
[0034] Collect thermal imaging images and surrounding objects through infrared sensors;
[0035] Process the collected environmental monitoring data through data cleaning, data conversion, data standardization and feature engineering;
[0036] It should be noted that data cleaning is used to check and process missing values, detect and process outliers, and eliminate noise data; data conversion is used to unify the units of different sensor data, align data timestamps, and convert data formats; data standardization is used to eliminate dimensional differences and perform logarithmic transformation on skewed data; and feature engineering is used to calculate feature importance scores, construct new combined features, extract useful information from the data, and select the most relevant features.
[0037] Fusing the pre-processed environmental monitoring data;
[0038] Furthermore, the preprocessed environmental monitoring data is used to train the random forest model. The importance score of each feature is obtained through the random forest model. According to the Gini importance, the reduction of the Gini index of each feature on all nodes using the feature is calculated. The expression is:
[0039]
[0040] Among them, K f is the set of nodes using feature f, G p is the Gini index of the parent node, G k is the Gini index of child node k, n k is the number of samples in child node k, n is the number of samples in the parent node, C k is the set of child nodes k;
[0041] The overall importance score of each feature is obtained by summing the reduction of the Gini index of each feature, and the feature importance score is obtained after normalization;
[0042] The nonlinear function is introduced to enhance the nonlinear characteristics of the features. The weighting coefficient is calculated according to the feature importance score and the nonlinear characteristics. Based on the weighting coefficient and the nonlinear characteristics, the fused environmental monitoring data S is calculated. The expression is:
[0043]
[0044] Among them, S is the fused environmental monitoring data, N is the number of sensors, and X i is the i-th sensor data, W i is the weighting coefficient of the i-th sensor data, Var(X i ) is the variance of the i-th sensor data, μ i is the mean value of the i-th sensor data, σ i The standard deviation of the i-th sensor data, γ is the adjustment factor.
[0045] It should be noted that the random forest model is selected as a tool for feature importance scoring. The preprocessed data is divided into a training set and a test set. The training set data is used to train the random forest model. After training, the random forest model outputs the importance score of each feature. The logarithmic function increases the nonlinear characteristics of the data, making it more suitable for complex environmental monitoring data. The weighting coefficient W is calculated based on the feature importance score and the nonlinear function. i, through standardization, nonlinear transformation and weighted combination, the fused environmental monitoring data S is finally calculated. The adjustment factor γ is used to control the influence of the nonlinear function to ensure that the fused environmental monitoring data S can better reflect the complexity and nonlinear characteristics of the actual environment. Generally, the initial setting is made through experience, and then the cross-validation method is used to evaluate the impact on the model. Finally, the grid search method is used to find the optimal adjustment factor γ.
[0046] Build an anomaly detection model, input the fused environmental monitoring data, and identify abnormal signals under normal working conditions;
[0047] Furthermore, the support vector machine is selected as the anomaly detection model, and the fused environmental monitoring data is input to identify abnormal signals under normal working conditions. The expression is:
[0048]
[0049] Among them, A is the abnormal signal strength, M is the number of environmental monitoring data elements after fusion, S j is the jth element in the fused environmental monitoring data, μ S is the mean of the fused environmental monitoring data, σ S is the standard deviation of the environmental monitoring data after fusion, β is the adjustment factor for adjusting the influence of the hyperbolic sine function, and sinh is the hyperbolic sine function;
[0050] It should be noted that the support vector machine identifies abnormal data by learning the boundaries of normal data and is suitable for high-dimensional and nonlinear data. The fused environmental monitoring data is used as input to ensure the high quality of the data. The support vector machine model is trained using data under normal working conditions to ensure that the model can accurately identify abnormal data. The data is standardized by calculating the mean and standard deviation to eliminate dimensional differences. The nonlinear characteristics of the data are enhanced by calculating the absolute value and the hyperbolic sine function value. The abnormal signal strength of each element is calculated by weighted combination of the absolute value and the hyperbolic sine function value. The total abnormal signal strength is calculated by summing and averaging. The larger the abnormal signal strength value, the more obvious the abnormal signal. β is used to control the influence of the hyperbolic sine function sinh. The optimal value is usually determined through cross-validation and parameter tuning. Sinh is used to enhance the nonlinear characteristics of the data so that the model can better capture abnormal signals.
[0051] Set the initial threshold based on historical data and adjust the initial threshold based on the abnormal signal strength;
[0052] Furthermore, the initial threshold T0 is set according to historical data, and the initial threshold is adjusted by the abnormal signal strength A. The expression is:
[0053]
[0054] Where T is the adjusted threshold, log is the natural logarithm function with e as the base, and max(A) is the maximum value of the abnormal signal strength in the historical data;
[0055] It should be noted that historical data is collected and cleaned, a reasonable initial threshold T0 is set according to the abnormal signal strength in the historical data, the initial threshold is adjusted by the abnormal signal strength A, the maximum abnormal signal strength in the historical data is calculated, and the threshold is dynamically adjusted to ensure that the threshold can adapt to changes in the current environment.
[0056] Real-time warning through adjusted initial thresholds;
[0057] Furthermore, by setting a time window, calculating the abnormal signal strength at each time point in the time window, and comparing the calculated abnormal signal strength with the adjusted threshold, a multi-time point judgment is performed;
[0058] When the abnormal signal strength is greater than the adjusted threshold, the current state is considered abnormal, an early warning is triggered, and an early warning message is output;
[0059] It should be noted that the selection of an appropriate time window length, such as 30 seconds, 1 minute, 5 minutes, etc., depends on the specific monitoring needs and system response speed. The time window length should be long enough to capture the duration of the abnormal signal, but not too long to avoid delayed warning. The abnormal signal strength is calculated at each time point in the time window, and the calculated abnormal signal strength is compared with the adjusted threshold. A multi-time point judgment mechanism is set within a time window. If the abnormal signal strength is greater than the adjusted threshold at multiple consecutive time points, the current state is considered abnormal and a warning is triggered. This multi-time point judgment mechanism can reduce false alarms and missed alarms and improve the accuracy and reliability of the warning.
[0060] This embodiment also provides a distribution network near-power monitoring and early warning system, including:
[0061] The data acquisition module collects environmental monitoring data of key nodes in the distribution network and pre-processes the collected environmental monitoring data;
[0062] Data fusion module, which fuses the pre-processed environmental monitoring data;
[0063] The anomaly detection module builds an anomaly detection model, inputs the fused environmental monitoring data, and identifies abnormal signals under normal working conditions;
[0064] Threshold management module, which sets the initial threshold based on historical data and adjusts the initial threshold according to the abnormal signal strength;
[0065] The real-time warning module provides real-time warning through the adjusted initial threshold.
[0066] This embodiment also provides a computer device, which is suitable for the distribution network near-power monitoring and early warning method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the distribution network near-power monitoring and early warning method proposed in the above embodiment.
[0067] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.
[0068] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for realizing near-power monitoring and early warning of a distribution network as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.
[0069] In summary, the present invention solves the deficiencies of the prior art in data fusion quality and threshold setting by: improving the data fusion method and dynamically adjusting the threshold mechanism. Specifically, the data fusion step improves the quality and reliability of the fused data through scientific weight distribution and the introduction of nonlinear functions; the dynamic threshold adjustment step realizes the dynamic adjustment of the threshold through the combination of historical data and real-time abnormal signal strength, thereby improving the flexibility and adaptability of the system. These innovative steps jointly improve the accuracy and real-time performance of the distribution network near-power monitoring and early warning system, ensuring the safe and stable operation of the distribution network.
[0070] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A distribution network near-power monitoring and early warning method, characterized by: include, Collect environmental monitoring data from key nodes of the distribution network and pre-process the collected environmental monitoring data; Fusing the pre-processed environmental monitoring data; Build an anomaly detection model, input the fused environmental monitoring data, and identify abnormal signals under normal working conditions; Set the initial threshold based on historical data and adjust the initial threshold based on the abnormal signal strength; Real-time warning is provided through adjusted initial thresholds.
2. The distribution network near-current monitoring and early warning method according to claim 1, characterized in that: The environmental monitoring data includes electromagnetic field strength, current, voltage, ambient temperature, ambient humidity, equipment surface temperature, thermal imaging images and images of surrounding objects.
3. The distribution network near-power monitoring and early warning method according to claim 2, characterized in that: The preprocessing includes data cleaning, data conversion, data standardization and feature engineering.
4. The distribution network near-current monitoring and early warning method according to claim 3, characterized in that: The specific steps of fusing the pre-processed environmental monitoring data are as follows: Use the preprocessed environmental monitoring data to train a random forest model; According to the Gini importance, the Gini index reduction of each feature on all nodes that use the feature is calculated, and the Gini index reduction of each feature is summed to obtain the overall importance score of the feature. After normalization, the feature importance score is obtained; A nonlinear function is introduced to enhance the nonlinear characteristics of the feature. The weighting coefficient is calculated according to the feature importance score and the nonlinear characteristics. Based on the weighting coefficient and the nonlinear characteristics, the fused environmental monitoring data S is calculated.
5. The power distribution network near-current monitoring and early warning method according to claim 4, characterized in that: The anomaly detection model is constructed, the fused environmental monitoring data is input, and the abnormal signals under normal working conditions are identified. The specific steps are as follows: The support vector machine is selected as the anomaly detection model, and the fused environmental monitoring data is input to identify abnormal signals under normal working conditions.
6. The method for monitoring and early warning of power shortage in a distribution network according to claim 5, characterized in that: The initial threshold is set based on historical data, and the initial threshold is adjusted according to the abnormal signal strength. The specific steps are as follows: The initial threshold T0 is set according to historical data, and the initial threshold is adjusted according to the abnormal signal strength A.
7. The power distribution network near-current monitoring and early warning method according to claim 6, characterized in that: The specific steps of performing real-time warning by adjusting the initial threshold are as follows: By setting a time window, calculating the abnormal signal strength at each time point in the time window, and comparing the calculated abnormal signal strength with the adjusted threshold, a multi-time point judgment is performed; When the abnormal signal strength is greater than the adjusted threshold, the current state is considered abnormal, an early warning is triggered, and an early warning message is output.
8. A distribution network near-power monitoring and early warning system, based on the distribution network near-power monitoring and early warning method according to any one of claims 1 to 7, characterized in that: It includes data acquisition module, data fusion module, anomaly detection module, threshold management module and real-time warning module. The data acquisition module collects environmental monitoring data of key nodes in the distribution network and pre-processes the collected environmental monitoring data; Data fusion module, which fuses the pre-processed environmental monitoring data; The anomaly detection module builds an anomaly detection model, inputs the fused environmental monitoring data, and identifies abnormal signals under normal working conditions; Threshold management module, which sets the initial threshold based on historical data and adjusts the initial threshold according to the abnormal signal strength; The real-time warning module provides real-time warning through the adjusted initial threshold.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the distribution network near-power monitoring and early warning method described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distribution network near-current monitoring and early warning method according to any one of claims 1 to 7 are implemented.