An anti-electricity-stealing system based on embedded intelligent online detection and situation awareness technology

Abstract

The present application relates to the technical field of power grid sensing, and more particularly to an anti-electricity-stealing system based on embedded intelligent online detection and situation awareness technology. The present application is efficient in post-processing of data collection, reasonable in data analysis, and timely in classified processing of data after analysis, thereby providing certain guarantee for subsequent maintenance and optimization scheme, and greatly improving the safety of power grid use. When processing the collected data, the present application uses the way of geometric invariant moment to quickly compare the picture contour, which is fast in comparison speed and high in accuracy, thereby providing certain guarantee for quick extraction of early warning data, and reducing system loss and operation cost. By calculating the risk probability of failure of the to-be-detected point in advance before maintenance, and setting the sampling fraction of the point according to the risk probability, the risk probability and the collected fraction are proportionally distributed, so as to achieve the purpose of reducing the collected fraction as much as possible while ensuring accurate collection, and achieving efficient maintenance.

Description

Technical Field

[0001] The present invention relates to the field of power grid perception technology, and in particular to an anti-electricity theft system based on embedded intelligent online detection and situational awareness technology. Background Art

[0002] As the scale of distribution networks continues to expand, distributed generation is connected to distribution networks in large quantities. Its uncertainty and randomness have also posed new challenges to the safe, economical and stable operation of distribution networks. The difficulty and complexity of power system operation and control have greatly increased, so it is particularly important to perceive and predict the power grid situation.

[0003] After searching, the patent with application number 202210154688.3 discloses a power grid dispatching situation awareness detection system, including a data acquisition module, a situation awareness module and an output module. The data acquisition module collects data information of legitimate users and enters the situation awareness module. The situation awareness module performs situation assessment through the distribution network real-time detection module, monitors the internal dynamics of the power grid through the load forecasting module and transmits signals to the risk forecasting module for risk forecasting. According to the abnormal event processing module, multi-source heterogeneous power grid status data is collected, and a comprehensive evaluation of the power grid security status is formed through comprehensive analysis and judgment by the security capability optimization module. The output module summarizes and infers the development and change laws of the power grid situation based on the perception, understanding and evaluation results of the situation information, and predicts the development and change trends of the future situation.

[0004] At present, the post-processing operations of power grid data collection are cumbersome and inaccurate, which causes great inconvenience to subsequent power grid maintenance and management, seriously affects the safety of power grid use, and cannot meet people's needs in monitoring. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the present invention provides an anti-electricity theft system based on embedded intelligent online detection and situational awareness technology, which solves the problems of cumbersome operation and poor analysis accuracy in the existing technology.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an anti-electricity theft system based on embedded intelligent online detection and situational awareness technology, comprising the following steps:

[0007] S11, situation data collection, collecting data on the installation, use status and surrounding environment of the overhead cables to obtain data set C;

[0008] S12, situation awareness and warning, processing the collected data set C to obtain the warning data set Y;

[0009] S13, situation accounting and assessment, analyze the early warning data set Y and obtain the risk assessment report set P;

[0010] S14, situation risk classification, classifying the risk assessment report set P into multiple groups according to the incident location and risk status;

[0011] S15. Situation processing and control: Match the corresponding maintenance team and equipment as well as the subsequent optimized configuration plan based on the multiple risk assessment reports classified above.

[0012] Preferably, data collection and prediction is added before step S11, and the specific steps are as follows:

[0013] S101: Retrieving early warning data from previously collected situation data from a database;

[0014] S102: Calculate the risk probability G of the detection point from the warning data in the situation data:

[0015] S103: Calculate the number of data collection copies F according to the risk probability G of the detection point.

[0016] The calculation formula of risk probability G is:

[0017]

[0018] Where ZC is the total number of sampling times, CX is the number of times warning data appears in the total number of sampling times ZC, θ is the risk compensation parameter, θ = 3.5;

[0019] The calculation formula for the number of data collection copies F is:

[0020]

[0021] Among them, CY is the base number for basic data collection. To collect compensation,

[0022] Preferably, the method of collecting situation data in step S11 includes: electronic equipment inspection and manual inspection;

[0023] Electronic equipment inspection: Using electronic equipment installed on drones to collect data on the installation of overhead cables and the surrounding environment;

[0024] Manual inspection: Inspectors collect data on the usage status of overhead cables.

[0025] Preferably, the specific steps of the situation awareness warning in step S12 are:

[0026] S121: Extracting image features. This includes high-level and low-level features. Low-level features can be obtained by automatic analysis of the image. Low-level features include color, shape, and texture. High-level features are semantically related, such as location information in the image and note points noted during shooting.

[0027] S122: Comparison of image features: comparing the features extracted from the image with those stored in the database to determine whether there are any anomalies in the image. The database is constructed by inspecting the collected graphics of the overhead power grid;

[0028] S123: If there is an abnormality in the comparison of the image features in step S122, the comparison image is stored in the early warning data set Y; if the comparison of the image features is normal, the comparison image is directly discarded.

[0029] In the S122 image feature comparison, the image contour recognition adopts the method of geometric invariant moment:

[0030] Definition of moment: For a two-variable bounded function f(x,y), its (n,j)-order moment is:

[0031]

[0032] Where n,j = 0, 1, 2...

[0033] In particular, the zeroth moment is the area of ​​the object:

[0034]

[0035] The commonly used central moment is calculated with the center of mass as the origin:

[0036] Center of mass location:

[0037]

[0038] Furthermore, the normalized central moment can be defined as;

[0039]

[0040] in:

[0041]

[0042] There is a one-to-one correspondence between a function and its moment set. To describe the shape, assume that f(x,y) takes the value 1 inside the object and takes the value 0 outside it. In this way, it establishes a one-to-one correspondence with the outline of the object, and its moment reflects the outline information of the object. The central moment is position-independent. For the normalized central moment, there are seven invariant moment combinations that are invariant to translation, rotation, and scale changes:

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050] Preferably, the specific operations of the situation calculation and evaluation in S13 are:

[0051] Perform situation calculation on the early warning data set Y according to the situation, calculate the characteristic data measured in the early warning data set Y, and obtain the fault content of each data in the early warning data set Y. Multiple fault contents constitute the risk assessment report set P, and the fault content includes the fault type and the required maintenance method, equipment and personnel.

[0052] Preferably, the situation risk classification operation in S14 is: first classify the fault contents in the risk assessment report set P according to the location of the incident to obtain a maintenance location map, and then group the contents to be repaired on the maintenance location map.

[0053] Preferably, the configuration scheme optimized in S15 adopts an improved tabu search algorithm, and its specific steps are:

[0054] S151: Initial solution coding structure:

[0055] X=[n1,....n m ;p1,....p m ] Among them, n m is the node number; p m is the capacity corresponding to the node;

[0056] Input the original data, including the distribution network line parameters, the limit values ​​of the inequality constraints and the parameters of the taboo search algorithm, and use the forward-backward substitution method to calculate the initial loss of the reactive capacity optimization distribution network;

[0057] S152: Selecting an initial solution for the tabu search;

[0058] S153: Determine the termination condition. The termination condition is that the optimal solution is obtained, or the solution cannot be improved after the specified number of iterations. If the process terminates, the optimal solution is output. Otherwise, the process proceeds to the next step.

[0059] S154: Generate a domain solution and a test solution based on the results of the neighborhood search analysis;

[0060] S155: Calculate candidate solutions. For each trial solution generated above, perform calculations based on the case of a fixed capacitor and calculate the objective function of the trial solution.

[0061] S156: All trial solutions in the neighborhood are tested according to the constraint conditions, and the candidate solution is judged whether the drug criterion is satisfied. If it is satisfied, the current best solution is replaced as the new current solution. If not, the neighborhood search is continued until the condition is satisfied and the search is terminated.

[0062] S157: If the objective function does not change within the number of iterations specified by the algorithm, the iteration ends and the reactive compensation capacity and position vector are output. Otherwise, return to step S152 and continue the tabu search algorithm steps.

[0063] By means of the above technical solution, the present invention provides an anti-electricity theft system, system and device based on embedded intelligent online detection and situational awareness technology, which has at least the following beneficial effects:

[0064] 1. The present invention is efficient in post-data collection processing, reasonable in data analysis, and timely in data classification and processing after analysis, thereby providing certain guarantees for subsequent maintenance and optimization plans, thereby greatly improving the safety of power grid use;

[0065] 2. When processing collected data, the present invention uses geometric invariant moments to quickly compare image contours. This has high speed and accuracy, thus providing a certain guarantee for the rapid extraction of early warning data, thereby reducing system losses and operating costs.

[0066] 3. After extracting the early warning data, the present invention promptly calculates and evaluates the early warning data and calculates a risk assessment report for a single early warning data, thereby facilitating subsequent classification and screening of risk assessment reports for multiple early warning data, thereby obtaining the optimal maintenance plan, greatly reducing the cost of power grid maintenance and improving maintenance efficiency;

[0067] 4. The present invention calculates the risk probability of failure at the point to be inspected before maintenance, and then sets the number of samples taken at the point according to the risk probability, thereby achieving a proportional distribution of the risk probability and the number of samples collected, so as to achieve the goal of ensuring accurate collection while reducing the number of samples collected as much as possible, thereby achieving the purpose of efficient maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0069] Figure 1 This is a flow chart of an anti-electricity theft system based on embedded intelligent online detection and situational awareness technology of the present invention;

[0070] Figure 2 A flowchart of specific steps of situation awareness and early warning in an anti-electricity theft system based on embedded intelligent online detection and situation awareness technology of the present invention;

[0071] Figure 3 The present invention provides a flowchart of an improved tabu search algorithm for optimizing a configuration scheme in an anti-electricity theft system based on embedded intelligent online detection and situational awareness technology.

[0072] Figure 4 This is a flow chart of data collection and prediction in an anti-electricity theft system based on embedded intelligent online detection and situational awareness technology of the present invention. DETAILED DESCRIPTION

[0073] To make the above-mentioned objectives, features, and advantages of the present invention more clearly understood, the present invention is further described below in detail with reference to the accompanying drawings and specific embodiments. This will enable a full understanding of how this application uses technical means to solve technical problems and achieve technical effects, and to implement the invention accordingly.

[0074] Those skilled in the art will appreciate that all or part of the steps in the above-mentioned embodiment methods can be accomplished by instructing the relevant hardware through a program. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0075] Example 1

[0076] Please refer to Figure 1 , an anti-electricity theft system based on embedded intelligent online detection and situational awareness technology, including the following steps:

[0077] S11, situation data collection, collecting data on the installation, use status and surrounding environment of the overhead cables to obtain data set C;

[0078] S12, situation awareness and warning, processing the collected data set C to obtain the warning data set Y;

[0079] S13, situation accounting and assessment, analyze the early warning data set Y and obtain the risk assessment report set P;

[0080] S14, situation risk classification, classifying the risk assessment report set P into multiple groups according to the incident location and risk status;

[0081] S15. Situation processing and control: Match the corresponding maintenance team and equipment as well as the subsequent optimized configuration plan based on the multiple risk assessment reports classified above.

[0082] The methods for collecting situation data in step S11 include: electronic equipment inspection and manual inspection;

[0083] Electronic equipment inspection: Using electronic equipment mounted on drones to collect data on the installation of overhead cables and the surrounding environment, this inspection primarily focuses on detecting changes in the status of overhead power grid poles and the surrounding environment, such as the presence of flammable materials, dangerous temporary structures, and tree branches and kites hanging on overhead cables.

[0084] Manual inspection: Inspectors collect data on the use status of overhead cables, mainly to check whether the binding wires are tight, whether the interface contacts are good, and whether the lightning rods are well grounded. Whether the interface contacts are good includes overheating, reddening, aging corrosion, breakage, pollution of insulators, and discharge points.

[0085] The specific steps of the situation awareness and warning in step S12 are:

[0086] S121: Extracting image features. This includes high-level and low-level features. Low-level features can be obtained by automatic analysis of the image. Low-level features include color, shape, and texture. High-level features are semantically related, such as location information in the image and note points noted during shooting.

[0087] S122: Comparison of image features: comparing the features extracted from the image with those stored in the database to determine whether there are any anomalies in the image. The database is constructed by inspecting the collected graphics of the overhead power grid;

[0088] S123: If there is an abnormality in the comparison of the image features in step S122, the comparison image is stored in the early warning data set Y; if the comparison of the image features is normal, the comparison image is directly discarded.

[0089] In the S122 image feature comparison, the image contour recognition adopts the method of geometric invariant moment:

[0090] Definition of moment: For a two-variable bounded function f(x,y), its (n,j)-order moment is:

[0091]

[0092] Where n,j = 0, 1, 2...

[0093] In particular, the zeroth moment is the area of ​​the object:

[0094]

[0095] The commonly used central moment is calculated with the center of mass as the origin:

[0096] Center of mass location:

[0097]

[0098] Furthermore, the normalized central moment can be defined as;

[0099]

[0100] in:

[0101]

[0102] There is a one-to-one correspondence between a function and its moment set. To describe the shape, assume that f(x,y) takes the value 1 inside the object and takes the value 0 outside it. In this way, it establishes a one-to-one correspondence with the outline of the object, and its moment reflects the outline information of the object. The central moment is position-independent. For the normalized central moment, there are seven invariant moment combinations that are invariant to translation, rotation, and scale changes:

[0103]

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110] When processing the collected data, the geometric invariant moment method is used to quickly compare the image contours. The comparison speed is fast and the accuracy is high, which provides a certain guarantee for the rapid extraction of warning data, thereby reducing system losses and operating costs.

[0111] The specific operations of situation accounting assessment in S13 are:

[0112] Perform situation calculation on the early warning data set Y according to the situation, calculate the characteristic data measured in the early warning data set Y, and obtain the fault content of each data in the early warning data set Y. Multiple fault contents constitute the risk assessment report set P, and the fault content includes the fault type and the required maintenance method, equipment and personnel.

[0113] The operation of situation risk classification in S14 is as follows: first, the fault contents in the risk assessment report set P are classified according to the location of the incident to obtain a maintenance location map, and then grouped according to the contents to be maintained on the maintenance location map, wherein a point can be set as the center of a circle, and a circular maintenance range is formed by a set radius. The maintenance contents within the range that are similar or identical are divided into the same group. The maintenance contents can also be directly grouped according to the distance segments of a single maintenance line, and can also be directly grouped according to the difficulty of the maintenance contents, so as to subsequently assign the corresponding maintenance team.

[0114] The optimized configuration scheme in S15 uses an improved tabu search algorithm, and its specific steps are as follows:

[0115] S151: Initial solution coding structure:

[0116] X=[n1,....n m ;p1,....p m ] Among them, n m is the node number; p m is the capacity corresponding to the node;

[0117] Input the original data, including the distribution network line parameters, the limit values ​​of the inequality constraints and the parameters of the taboo search algorithm, and use the forward-backward substitution method to calculate the initial loss of the reactive capacity optimization distribution network;

[0118] S152: Selecting an initial solution for the tabu search;

[0119] S153: Determine the termination condition. The termination condition is that the optimal solution is obtained, or the solution cannot be improved after the specified number of iterations. If the process terminates, the optimal solution is output. Otherwise, the process proceeds to the next step.

[0120] S154: Generate a domain solution and a test solution based on the results of the neighborhood search analysis;

[0121] S155: Calculate candidate solutions. For each trial solution generated above, perform calculations based on the case of a fixed capacitor and calculate the objective function of the trial solution.

[0122] S156: All trial solutions in the neighborhood are tested according to the constraint conditions, and the candidate solution is judged whether the drug criterion is satisfied. If it is satisfied, the current best solution is replaced as the new current solution. If not, the neighborhood search is continued until the condition is satisfied and the search is terminated.

[0123] S157: If the objective function does not change within the number of iterations specified by the algorithm, the iteration ends and the reactive compensation capacity and position vector are output. Otherwise, return to step S152 and continue the tabu search algorithm steps.

[0124] After extracting the early warning data, the early warning data is promptly calculated and evaluated, and a risk assessment report for a single early warning data is calculated, which facilitates subsequent classification and screening of risk assessment reports for multiple early warning data, and thus obtains the optimal maintenance plan, greatly reducing the cost of power grid maintenance and improving maintenance efficiency.

[0125] Example 2

[0126] Please refer to Figure 1 Preferably, data collection and prediction is added before step S11, and the specific steps are as follows:

[0127] S101: Retrieving early warning data from previously collected situation data from a database;

[0128] S102: Calculate the risk probability G of the detection point from the warning data in the situation data:

[0129] S103: Calculate the number of data collection copies F according to the risk probability G of the detection point.

[0130] The calculation formula of risk probability G is:

[0131]

[0132] Where ZC is the total number of sampling times, CX is the number of times warning data appears in the total number of sampling times ZC, θ is the risk compensation parameter, θ = 3.5;

[0133] For example, if a total of ZC = 100 pieces of data are selected, and 45 pieces of data have a total of CX = 76 pieces of warning data, then

[0134]

[0135] The calculation formula for the number of data collection copies F is:

[0136]

[0137] Among them, CY is the base number for basic data collection. To collect compensation,

[0138] Take CY=50, then

[0139] F = round (50 * 38.4%) + 5 = 24

[0140] The number of sampling times for this point is 24.

[0141] In this embodiment, the risk probability of failure at the point to be inspected is calculated in advance before maintenance, and the number of samples taken at the point is set according to the level of the risk probability, thereby achieving a proportional distribution of the risk probability and the number of samples collected, so as to achieve the goal of reducing the number of samples collected as much as possible while ensuring accurate collection, thereby achieving the purpose of efficient maintenance.

[0142] The above embodiments provide a detailed introduction to the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. An anti-electricity theft system based on embedded intelligent online detection and situational awareness technology, characterized in that: The following steps are involved: S11, situation data collection, collecting data on the installation, use status and surrounding environment of the overhead cables to obtain data set C; S12, situation awareness and warning, processing the collected data set C to obtain the warning data set Y; S13, situation accounting and assessment, analyze the early warning data set Y and obtain the risk assessment report set P; S14, situation risk classification, classifying the risk assessment report set P into multiple groups according to the incident location and risk status; S15: Situation management and control: Matching the corresponding maintenance team and equipment, as well as subsequent optimized configuration solutions, based on the multiple risk assessment reports classified above; Data collection and prediction are added before step S11, and the specific steps are as follows: S101: Retrieving early warning data from previously collected situation data from a database; S102: Calculate the risk probability G of the detection point from the warning data in the situation data: S103: Calculate the number of data collection copies F based on the risk probability G of the detection point; The calculation formula of risk probability G is: ； in is the total number of sampling times, The total number of sampling The number of times warning data appears in is the risk compensation parameter, =3.5; The calculation formula for the number of data collection copies F is: ； in, The base number for basic data collection, To collect compensation, =5; The specific steps of the situation awareness and warning in step S12 are: S121: Extracting image features. This extraction includes high-level and low-level features. Low-level features are obtained by automatic analysis of the image. Low-level features include color, shape, and texture. High-level features are semantically related and include location information in the image and note points noted during shooting. S122: Comparing the features of the image, comparing the features extracted from the image with those stored in the database to determine whether there are any abnormalities in the image; S123: If there is an abnormality in the comparison of the image features in step S122, the compared image is stored in the warning data set Y; if the comparison of the image features is normal, the compared image is directly discarded; In the comparison of S122 image features, the image contour recognition adopts the method of geometric invariant moment.

2. The anti-electricity theft system based on embedded intelligent online detection and situational awareness technology according to claim 1 is characterized in that: The methods for collecting situation data in step S11 include: electronic equipment inspection and manual inspection; Electronic equipment inspection: Using electronic equipment installed on drones to collect data on the installation of overhead cables and the surrounding environment; Manual inspection: Inspectors collect data on the usage status of overhead cables.

3. The anti-electricity theft system based on embedded intelligent online detection and situational awareness technology according to claim 1 is characterized in that: The specific operations of situation accounting assessment in S13 are: Perform situation calculation on the early warning data set Y according to the situation, calculate the characteristic data measured by the early warning data set Y, and obtain the fault content of each data in the early warning data set Y. Multiple fault contents constitute the risk assessment report set P.

4. The anti-electricity theft system based on embedded intelligent online detection and situational awareness technology according to claim 3 is characterized in that: The fault content includes the fault type and the required maintenance method, equipment and personnel.

5. The anti-electricity theft system based on embedded intelligent online detection and situational awareness technology according to claim 4 is characterized in that: The operation of situation risk classification in S14 is as follows: first, the fault contents in the risk assessment report set P are classified according to the location of the incident to obtain a maintenance location map, and then the contents to be repaired on the maintenance location map are grouped.

Images