A power distribution network automation intelligent management system

Abstract

This invention discloses an intelligent management system for power distribution network automation, belonging to the field of power distribution network automation management technology. It includes a distribution network analysis module, an isolation processing module, and an anomaly monitoring and management module. The distribution network analysis module analyzes management objectives and determines the corresponding management areas. It sets up an area information map corresponding to the management objectives, marking each independent objective on the area information map based on a preset independent objective statistics table. The area information map is divided into several unit areas, and each unit area is merged according to each independent objective to obtain the influence area corresponding to each independent objective. Power assessment is performed on each influence area to obtain the corresponding power influence value. The influence areas are then processed accordingly based on the power influence value. The isolation processing module prepares for distribution network isolation based on the influence areas corresponding to each independent objective. The anomaly monitoring and management module performs real-time anomaly monitoring and management of the management objectives.

Description

Technical Field

[0001] This invention belongs to the field of power distribution network automation management technology, specifically a power distribution network automation intelligent management system. Background Technology

[0002] With the continuous growth of electricity demand and the expansion of power system scale, traditional manual operation and management methods can no longer meet the requirements of efficient, reliable, and safe operation. In recent years, power distribution network automation technology has developed rapidly, realizing the automation of monitoring, control, and management of power distribution networks through intelligent devices and advanced communication network technologies. This technology not only improves the operating efficiency of the power grid but also reduces operation and maintenance costs and enhances the safety and stability of the power system.

[0003] Although existing technologies have achieved certain results in the automation of power distribution networks, some shortcomings still exist, such as limited ability to analyze power grid operation data and difficulty in providing accurate optimization scheduling and fault handling solutions. Based on this, the present invention provides an intelligent management system for power distribution network automation. Summary of the Invention

[0004] To address the problems of the above solutions, this invention provides an intelligent management system for power distribution network automation.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] An automated intelligent management system for power distribution networks includes a distribution network analysis module, an isolation processing module, and an anomaly monitoring and management module;

[0007] The power distribution network analysis module is used to analyze management objectives, determine the management areas corresponding to the management objectives, set up an area information map corresponding to the management objectives, and mark each independent objective in the area information map based on a preset independent objective statistics table.

[0008] The regional information map is divided into several unit regions. The unit regions are merged according to each independent target to obtain the influence area corresponding to each independent target.

[0009] A power impact assessment was conducted for each affected area to obtain the corresponding power impact value for each affected area.

[0010] The affected area will be handled accordingly based on the power impact value.

[0011] Furthermore, methods for merging unit regions based on independent objectives include:

[0012] Historical power data corresponding to independent targets are obtained, and an impact assessment model is established. The expression of the impact assessment model is as follows:

[0013] In the formula: LT represents historical power data, si represents the corresponding unit area, i = 1, 2, ..., n, where n is a positive integer; the output data is the influence judgment value 1 or 0;

[0014] The impact judgment model analyzes the historical power data of independent targets for each unit area to obtain the impact judgment value corresponding to each unit area.

[0015] The unit regions with an impact judgment value of 1 are merged into the impact regions corresponding to independent targets.

[0016] Furthermore, the methods for conducting power assessments in each affected area include:

[0017] Identify the information of each unit area within the affected area, evaluate the information of each unit area, and obtain the corresponding fault loss value;

[0018] Estimate the implementation cost for optimizing and segmenting the corresponding affected areas; set fault values ​​for individual devices;

[0019] According to the formula PY = (3.5) GZ The corresponding power impact value is calculated as -1)×GF{SL-CL};

[0020] In the formula: PY is the power impact value; GZ is the fault value; GF{SL-CL} is the piecewise function, and the expression for GF{SL-CL} is: In the formula: SL is the failure loss value; CL is the implementation cost.

[0021] Furthermore, methods for appropriately processing the affected area based on the power impact value include:

[0022] When GF{SL-CL}=1, no corresponding operation is performed;

[0023] When GF{SL-CL}≠1, the power impact value is compared with the threshold X1. When the power impact value is greater than the threshold X1, the corresponding impact area is optimized and segmented; when the power impact value is not greater than the threshold X1, no corresponding operation is performed.

[0024] Furthermore, methods for optimizing the segmentation of the affected area include:

[0025] In the regional information map, each affected area that needs to be optimized is marked as an optimized area, and the affected areas of non-optimized areas are marked as normal areas; the power impact values ​​of each optimized area and normal area are marked.

[0026] Based on the regional information map, select the candidate optimization schemes for the corresponding optimization areas, evaluate each candidate optimization scheme, and obtain the corresponding implementation costs and effect values;

[0027] Calculate the priority value of each candidate optimization scheme according to the formula QY = b1×XG - b2×μ×CB;

[0028] In the formula: QY is the priority value; b1 and b2 are both proportionality coefficients, and the value ranges are 0 < b1 ≤ 1, 0 < b2 ≤ 1; μ is the unit conversion coefficient; CB is the application cost;

[0029] Select the candidate optimization scheme with the highest priority value as the target optimization scheme;

[0030] Set the optimization segmentation scheme according to the target optimization scheme, and optimize and segment the influence area according to the obtained optimization segmentation scheme.

[0031] Furthermore, the setting method of the candidate optimization scheme includes:

[0032] Identify the candidate areas corresponding to each optimization area, perform segmentation and merger analysis on each unit area in the optimization area and the candidate areas, and determine various segmentation and merger methods; determine various internal segmentation methods based on the independent segmentation method of the optimization area;

[0033] Evaluate each segmentation and merger method and internal segmentation method, and obtain the segmentation evaluation values corresponding to each segmentation and merger method and internal segmentation method; mark the segmentation and merger methods with a segmentation evaluation value of 0 as candidate optimization methods.

[0034] Furthermore, the method for evaluating each segmentation and merger method includes:

[0035] Establish a corresponding segmentation evaluation model, and the expression of the segmentation evaluation model is

[0036] In the formula: kf is the segmentation and merger method or internal segmentation method, and the output data is the segmentation evaluation value 1 or 0;

[0037] Evaluate each segmentation and merger method or internal segmentation method through the segmentation evaluation model, and obtain the segmentation evaluation values corresponding to each segmentation and merger method or internal segmentation method.

[0038] The isolation processing module is used to prepare for power distribution isolation according to the influence areas corresponding to the current independent targets.

[0039] The abnormal monitoring and management module is used to perform real-time abnormal monitoring and management on the management target, obtain the monitoring data of each independent target in real time, perform abnormal identification on the obtained monitoring data, and obtain the corresponding abnormal identification result. The abnormal identification result includes monitoring abnormal and monitoring normal;

[0040] When the abnormal identification result is monitoring normal, no corresponding operation is performed;

[0041] When the anomaly identification result is a monitoring anomaly, the corresponding independent device is marked as an abnormal device; the anomaly type of the abnormal device is determined; when the anomaly type is a Class I anomaly, the affected area corresponding to the abnormal device is identified, and power distribution isolation is performed on the affected area; when the anomaly type is a Class II anomaly, the corresponding power consumption impact range is identified; the affected area corresponding to the abnormal device is identified, and the recovery area is determined based on the affected area; power distribution isolation is performed on the affected area, and power supply is restored to the recovery area.

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] By coordinating the distribution network analysis module, isolation processing module, and anomaly monitoring and management module, automated and intelligent management of the power distribution network can be achieved; power fault handling methods can be optimized to reduce the adverse effects of power equipment failures; and by conducting power assessments of each affected area, it can be understood whether the allocation settings of each affected area are reasonable, so as to avoid large-scale power outages and significant economic losses caused by abnormal failures of corresponding independent targets. Therefore, it is necessary to minimize the adverse effects based on the actual situation. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a block diagram illustrating the principle of the present invention. Detailed Implementation

[0046] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0047] like Figure 1 As shown, a power distribution network automation intelligent management system includes a distribution network analysis module, an isolation processing module, and an anomaly monitoring and management module.

[0048] The power distribution network analysis module is used to analyze the power distribution network, marking the corresponding power distribution network as management targets; determining the area range corresponding to the management targets and marking it as the management area; setting up an area information map of the management targets, which is used to display relevant information of various power equipment, lines, sensors, etc. within the management area, such as model, location, corresponding lines, etc.; marking each independent target in the area information map, which refers to which equipment in the management area will affect power use after a failure, such as transmission lines, transformers, switching stations, etc., and marking the corresponding equipment as independent equipment. Specifically, professionals will establish a corresponding independent target statistical table. Power equipment that cannot be optimized or adjusted is not included in the independent target statistical table; subsequent identification and marking are based on the independent target statistical table.

[0049] The regional information map is divided into several unit areas, that is, the division is based on the user terminal, such as a residential community, shopping mall, office building, hospital, factory, etc.; a residential community is a unit area;

[0050] Identify the impact of each independent target on each unit area, and then merge the unit areas to form the impact area corresponding to each independent target;

[0051] A power impact assessment was conducted for each affected area to obtain the corresponding power impact value for each affected area.

[0052] The affected area will be handled accordingly based on the power impact value.

[0053] By conducting power assessments of each affected area, we can understand whether the power allocation settings for each affected area are reasonable, and avoid large-scale power outages and significant economic losses caused by abnormal failures of corresponding independent targets. Therefore, it is necessary to minimize the adverse impact based on the actual situation.

[0054] Methods for determining the influence area of ​​independent targets include:

[0055] Historical power data corresponding to an independent target is acquired, primarily including power data during normal operation and failure. This data is used to assess which unit areas are affected when the independent target fails, thereby determining which unit areas are truly impacted. An impact assessment model is established to determine whether each unit area is affected by the independent target based on historical power data. Specifically, this model is built using existing data processing techniques, and its expression is as follows: In the formula: LT represents historical power data, si represents the corresponding unit area, i = 1, 2, ..., n, where n is a positive integer; the output data is the influence judgment value 1 or 0;

[0056] The impact judgment model analyzes the historical power data of independent targets for each unit area to obtain the impact judgment value corresponding to each unit area.

[0057] The unit regions with an impact judgment value of 1 are merged into the impact regions corresponding to independent targets.

[0058] Methods for conducting power assessments in various affected areas include:

[0059] Identify information about each unit area within the affected area. This information includes type, average electricity consumption, electricity usage, and backup power. Evaluate the information of each unit area to obtain the corresponding fault loss value. The fault loss value is assessed based on its electricity consumption, electricity usage, and backup power. When the electricity consumption of a unit area is affected by the fault of this independent target, the potential loss of the corresponding unit area is estimated according to the existing loss estimation method.

[0060] Based on the independent target statistics table, several optimized segmentation methods for the impact areas corresponding to different independent targets are preset. This involves reducing the impact area corresponding to each independent target by adding appropriate power equipment, optimizing lines, etc. The segmented impact area can be allocated to other impact areas or formed into a new impact area independently. Specific existing applicable technologies are used to set these methods. The impact area corresponding to each independent target is analyzed using the preset optimized segmentation methods to estimate the implementation cost required for optimized segmentation, and the lowest cost is selected as the implementation cost. For example, a corresponding implementation cost estimation model can be established based on neural networks such as DNN networks. The model is trained manually using a training set, which includes input and output data. The input data includes information about each unit area within the impact area and the preset optimized segmentation methods; the output data is the implementation cost.

[0061] Set the fault value for each individual device. The fault value is the failure rate of that individual device. The failure rate can also be corrected based on the actual application environment of the individual device to make the fault value more accurate.

[0062] According to the formula PY = (3.5) GZ The corresponding power impact value is calculated as -1)×GF{SL-CL};

[0063] In the formula: PY is the power impact value; GZ is the fault value; GF{SL-CL} is the piecewise function, and the expression for GF{SL-CL} is: In the formula: SL is the failure loss value; CL is the implementation cost.

[0064] Methods for appropriately handling the affected area based on the power impact value include:

[0065] When GF{SL-CL}=1, no corresponding operation is performed;

[0066] When GF{SL-CL}≠1, the power impact value is compared with the threshold X1. When the power impact value is greater than the threshold X1, the corresponding impact area is optimized and segmented; when the power impact value is not greater than the threshold X1, no corresponding operation is performed.

[0067] Methods for optimizing the segmentation of the affected area include:

[0068] The power impact value is estimated based on the minimum standard during the assessment and calculation process. However, in actual application, it will be changed according to the actual situation. That is, when the minimum standard can reach the optimized segmentation standard, other methods are also acceptable.

[0069] The detailed process includes: marking the affected areas that need to be optimized and segmented in the regional information map; marking the affected areas that need to be optimized and segmented as optimized areas, and marking other affected areas as normal areas; marking the power impact values ​​of each optimized area and normal area;

[0070] Based on the regional information map, set the corresponding candidate optimization schemes for each optimized region;

[0071] Each candidate optimization scheme is comprehensively evaluated from both cost and effectiveness perspectives. Cost is estimated based on the available implementation methods for each candidate optimization scheme. Generally, existing analytical techniques are used to determine which application implementation method best meets the user's requirements under current conditions, and cost is estimated accordingly. Alternatively, the user's staff can specify the possible implementation methods when optimizing and segmenting this type of independent objective; there can be one or more such methods. If none of the preset implementation methods are applicable to the candidate optimization scheme, it is directly eliminated. If it is applicable, the cost corresponding to the applicable implementation method is evaluated. If multiple implementation methods exist, multiple costs are evaluated, and then... The implementation method determines the corresponding effect value, which ranges from [1, 10]. First, all possible effect scenarios are identified and sorted from highest to lowest. The effect value for the highest-ranked scenario is 10, and the lowest is 1. Other effects are set according to their differences. Specifically, a corresponding effect evaluation model can be built based on a CNN or DNN network. A training set is manually created and used for training. The training set includes input and output data. Input data includes the implementation method, optimization region information, and candidate optimization schemes. Output data is the effect value. The effect value is then evaluated using the successfully trained effect evaluation model. Other methods can also be used to evaluate the effect value.

[0072] The priority value of each candidate optimization scheme is calculated according to the formula QY=b1×XG-b2×μ×CB;

[0073] Where: QY is the priority value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 ≤ 1, 0 < b2 ≤ 1; μ is the unit conversion coefficient, which is used to convert the unit of the application cost and is set by the expert group through discussion; CB is the application cost;

[0074] Select the candidate optimization plan with the highest priority value as the target optimization plan;

[0075] Set the optimization segmentation plan according to the target optimization plan, that is, set the optimization segmentation plan manually in combination with the implementation method; optimize and segment the affected area according to the obtained optimization segmentation plan.

[0076] The method for setting the candidate optimization plans for the optimization area according to the area information map includes:

[0077] Identify the normal areas of the same type adjacent to the optimization area, where the same type refers to the same type of independent target; mark the corresponding normal areas as candidate areas, and perform segmentation evaluation on each unit area in the optimization area with the candidate areas, that is, segment and merge the corresponding unit area into the candidate area, and evaluate whether the power influence values of the two meet the requirements. When the requirements are met, a candidate optimization plan is formed. By analogy, several candidate optimization plans are obtained, including the segmentation and merger analysis between the optimization area and one or more candidate areas. According to this method, several candidate optimization plans are formed; in the actual application process, a corresponding segmentation evaluation model can be established, and the power influence value is quickly evaluated based on the evaluation method of the power influence value, and then it is judged whether the segmentation and merger requirements are met; the expression of the segmentation evaluation model is Where: kf is the segmentation and merger method to be evaluated, and the output data is the segmentation evaluation value 1 or 0; determine the various segmentation and merger methods according to the optimization area and the candidate area, and evaluate through the segmentation evaluation model to obtain the segmentation evaluation values corresponding to each segmentation and merger method; mark the segmentation and merger methods with a segmentation evaluation value of 0 as candidate optimization methods;

[0078] Then, based on the independent segmentation method of the optimization area, set the candidate optimization methods, that is, do not segment and merge the unit areas in the optimization area into the adjacent normal areas, but establish a new affected area by adding corresponding power equipment and re-segmenting, which is equivalent to adding a set of power configurations corresponding to this independent target, then there will be multiple segmentation methods, marked as internal segmentation methods. Similarly, evaluate through the segmentation evaluation model to obtain each candidate optimization method; the setting of the internal segmentation method does not involve specific plans, that is, which plan is used to implement the application of this internal segmentation method is considered in the subsequent corresponding plan setting process.

[0079] The isolation processing module is used to prepare for distribution network isolation based on the affected areas corresponding to each independent target. This involves setting up appropriate power equipment so that when distribution network isolation is required for a specific affected area, that area can be isolated from the distribution network, with power cut off only to that affected area while power is restored to other areas. This is implemented using existing power technologies.

[0080] The anomaly monitoring and management module is used to perform real-time anomaly monitoring and management of the management targets, acquire monitoring data of each independent target in real time, identify anomalies in the acquired monitoring data, and obtain corresponding anomaly identification results. The anomaly identification results include monitoring anomalies and monitoring normality. The specific anomaly identification analysis can be performed using existing power monitoring and analysis technologies. However, monitoring anomalies are divided into two parts: one is monitoring anomalies that are still running, and the other is monitoring anomalies that have stopped running, which are marked as Class I anomalies and Class II anomalies, respectively.

[0081] When the anomaly identification result is normal, no corresponding operation is performed;

[0082] When the anomaly identification result is a monitoring anomaly, the corresponding independent device is marked as an abnormal device; the anomaly type of the abnormal device is determined; when the anomaly type is a Class I anomaly, the affected area corresponding to the abnormal device is identified, and the affected area is isolated from the power distribution network, that is, the area is disconnected from the power distribution network and power is cut off; when the anomaly type is a Class II anomaly, the current power consumption affected area, that is, the actual affected area, is identified and marked as the power consumption affected range; the affected area corresponding to the abnormal device is identified, and the recovery area is determined based on the affected area, that is, the area outside the affected area within the power consumption affected range; the affected area is isolated from the power distribution network, and power is restored to the recovery area.

[0083] By cooperating with the distribution network analysis module, isolation processing module, and anomaly monitoring and management module, automated and intelligent management of the power distribution network can be achieved; power fault handling methods can be optimized, and the adverse effects of power equipment failures can be reduced.

[0084] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.

[0085] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. 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 method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A power distribution network automation intelligent management system, characterized in that, It includes a power distribution network analysis module, an isolation processing module, and an anomaly monitoring and management module; The power distribution network analysis module is used to analyze management objectives, determine the management areas corresponding to the management objectives, set up an area information map corresponding to the management objectives, and mark each independent objective in the area information map based on a preset independent objective statistics table. The regional information map is divided into several unit regions. The unit regions are merged according to each independent target to obtain the influence area corresponding to each independent target. A power impact assessment was conducted for each affected area to obtain the corresponding power impact value for each affected area. The affected area will be treated accordingly based on the power impact value; The isolation processing module is used to prepare for power distribution network isolation based on the affected areas corresponding to each independent target. The anomaly monitoring and management module is used to perform real-time anomaly monitoring and management of the management target, acquire monitoring data of each independent target in real time, identify anomalies in the acquired monitoring data, and obtain corresponding anomaly identification results. The anomaly identification results include monitoring anomalies and monitoring normality. When the anomaly identification result is normal, no corresponding operation is performed; When the anomaly identification result is a monitoring anomaly, the corresponding independent device is marked as an abnormal device; the anomaly type of the abnormal device is determined; when the anomaly type is a Class I anomaly, the affected area corresponding to the abnormal device is identified, and power distribution isolation is performed on the affected area; when the anomaly type is a Class II anomaly, the corresponding power consumption impact range is identified; the affected area corresponding to the abnormal device is identified, and the recovery area is determined based on the affected area. Power distribution was isolated in the affected area, and power was restored to the affected area. Methods for conducting power assessments in various affected areas include: Identify the information of each unit area within the affected area, evaluate the information of each unit area, and obtain the corresponding fault loss value; Estimate the implementation cost for optimizing and segmenting the corresponding affected areas; set fault values ​​for individual devices; According to the formula Calculate the corresponding power impact value; In the formula: PY is the power impact value; GZ is the fault value; GF{SL-CL} is the piecewise function, and the expression for GF{SL-CL} is: In the formula: SL represents the failure loss value; CL represents the implementation cost; Methods for appropriately handling the affected area based on the power impact value include: When GF{SL-CL}=1, no corresponding operation is performed; When GF{SL-CL}≠1, the power impact value is compared with the threshold X1. When the power impact value is greater than the threshold X1, the corresponding impact area is optimized and segmented; when the power impact value is not greater than the threshold X1, no corresponding operation is performed.

2. The intelligent management system for power distribution network automation according to claim 1, characterized in that, Methods for merging unit regions based on independent objectives include: Historical power data corresponding to independent targets are obtained, and an impact assessment model is established. The expression of the impact assessment model is as follows: ; In the formula: LT represents historical power data, si represents the corresponding unit area, i=1, 2, ..., n, where n is a positive integer; the output data is the influence judgment value 1 or 0; The impact judgment model analyzes the historical power data of independent targets for each unit area to obtain the impact judgment value corresponding to each unit area. The unit regions with an impact judgment value of 1 are merged into the impact regions corresponding to independent targets.

3. The intelligent management system for power distribution network automation according to claim 1, characterized in that, Methods for optimizing the segmentation of the affected area include: In the regional information map, each affected area that needs to be optimized is marked as an optimized area, and the affected areas of non-optimized areas are marked as normal areas; the power impact values ​​of each optimized area and normal area are marked. Set each candidate optimization plan corresponding to the optimization area according to the area information map, evaluate each candidate optimization plan, and obtain the corresponding implementation cost and effect value; Calculate the priority value of each candidate optimization plan according to the formula QY = b1×XG - b2×μ×CB; In the formula: QY is the priority value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 ≤ 1, 0 < b2 ≤ 1; μ is the unit conversion coefficient; CB is the application cost; Select the candidate optimization plan with the highest priority value as the target optimization plan; Set the optimization segmentation plan according to the target optimization plan, and optimize and segment the influence area according to the obtained optimization segmentation plan.

4. The power distribution network automation intelligent management system according to claim 3, characterized in that, The setting method of the candidate optimization plan includes: Identify the candidate areas corresponding to each optimization area, perform segmentation and merging analysis on each unit area in the optimization area and the candidate areas, and determine various segmentation and merging methods; determine various internal segmentation methods based on the independent segmentation method of the optimization area; Evaluate each segmentation and merging method and internal segmentation method, and obtain the segmentation evaluation value corresponding to each segmentation and merging method and internal segmentation method; mark the segmentation and merging method with a segmentation evaluation value of 0 as the candidate optimization method.

5. The intelligent management system for power distribution network automation according to claim 4, characterized in that, The method for evaluating each segmentation and merging method includes: Establish a corresponding segmentation evaluation model, the expression of which is: ; In the formula: kf is the segmentation and merging method or internal segmentation method, and the output data is the segmentation evaluation value 1 or 0; Evaluate each segmentation and merging method or internal segmentation method through the segmentation evaluation model, and obtain the segmentation evaluation value corresponding to each segmentation and merging method or internal segmentation method.

Images