System and method for evaluating net rack capability based on multi-dimensional regional portraits

Abstract

The invention discloses a grid capability evaluation system and method based on a multi-dimensional region portrait, and relates to the technical field of electric power. Compared with a previous power distribution network frame capability evaluation method, the power distribution network frame capability evaluation method provided by the invention solves the problems that the existing power distribution network frame capability evaluation cannot timely discover unreasonable parts in a frame structure, accurately identify the root of an electric energy quality problem, determine aged equipment or equipment with insufficient performance and clarify the application effect and defects of an intelligent technology in a power grid; the problems of characteristics and demand differences of different regions are considered; based on six dimensions of grid power supply capability, electric energy quality, grid structure, equipment level, intelligent level and economic operation, the grid capability in a to-be-analyzed area is evaluated, the actual capability and operation condition of the grid can be reflected more comprehensively and accurately, and limitation and one-sidedness of single index evaluation are avoided; and a multi-dimensional regional portrait is established, so that the demand and the development trend of a future power grid can be better predicted, and powerful support is provided for prospective planning.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a grid capacity evaluation system and method based on multi-dimensional regional portraits. Background Art

[0002] At present, power grid companies have invested a lot of manpower and funds in the planning and construction of distribution network frameworks to achieve the goal of safe, reliable, green and efficient distribution network construction. Reasonable grid capacity evaluation methods can help identify the weak links and bottlenecks of the power grid and provide a basis for future grid planning and expansion to meet the growing power demand. However, the existing distribution network framework capacity evaluation cannot timely discover the unreasonable aspects of the grid structure, accurately identify the root causes of power quality problems, determine aging or insufficient performance equipment, clarify the application effects and shortcomings of intelligent technology in the power grid, and consider the characteristics and demand differences of different regions. Therefore, studying targeted distribution network framework structure capacity evaluation methods is a key issue that needs to be urgently solved in the current distribution network planning field.

[0003] In order to solve the above problems, the present invention proposes a grid capacity evaluation system and method based on multi-dimensional regional portraits. Summary of the invention

[0004] The purpose of the present invention is to propose a grid capacity evaluation system and method based on multi-dimensional regional portraits to solve the problems raised in the background technology:

[0005] The existing distribution network capacity evaluation is unable to timely discover the unreasonable aspects of the grid structure, accurately identify the root causes of power quality problems, determine the aging or underperforming equipment, clarify the application effects and shortcomings of intelligent technologies in the power grid, and consider the characteristics and demand differences of different regions.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The grid capacity evaluation system based on multi-dimensional regional portraits includes:

[0008] Data acquisition module: used to collect power grid-related data in the area to be analyzed and perform preprocessing operations;

[0009] Preliminary evaluation module: used to calculate the line contact rate, complex wiring rate, and standard wiring rate in the area to be analyzed, and to make a preliminary evaluation of the area to be analyzed;

[0010] Multi-dimensional portrait module: used to create regional portraits from several dimensions;

[0011] Capacity evaluation module: used to evaluate the grid capacity in the area to be analyzed based on the regional portrait established by the multi-dimensional portrait module;

[0012] The multi-dimensional portrait module includes a grid power supply capacity unit, a power quality unit, a grid structure unit, an equipment level unit, an intelligent level unit and an economic operation unit; and is used to establish regional portraits from six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligent level and economic operation.

[0013] The grid capacity evaluation method based on multi-dimensional regional portrait includes the following steps:

[0014] S1: Collect relevant data of the power grid in the area to be analyzed and perform preprocessing operations;

[0015] S2: Calculate the line connection rate, complex wiring rate, and standard wiring rate in the area to be analyzed, and make a preliminary assessment of the area to be analyzed;

[0016] S3: Evaluate the grid capacity in the area to be analyzed based on six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligence level and economic operation, and establish a regional profile;

[0017] S4: Evaluate the grid capacity in the area to be analyzed based on the DEA-ANN method according to the regional portrait established in S3.

[0018] Preferably, the analysis indicators of the grid power supply capacity include maximum power supply load, load margin and power supply reliability;

[0019] The analysis indicators of the power quality include voltage deviation, frequency deviation and harmonic content;

[0020] The analysis indicators in the grid structure include wiring mode, line corridor layout and network topology;

[0021] The analysis indicators in the equipment level include transformer performance, switch reliability and transmission line material and insulation level;

[0022] The analytical indicators in the intelligence level include automated monitoring system, intelligent scheduling capability and big data analysis application;

[0023] The analysis indicators in the economic operation include network loss rate, return on investment and operation and maintenance level.

[0024] Preferably, in S3, the six dimensional indicator results of the evaluation are comprehensively analyzed and a regional portrait is drawn, as follows:

[0025] Quantify each dimension indicator and set corresponding scoring indicators to calculate the score of each dimension. Calculate the weight value of different indicators in each dimension based on the improved cuckoo algorithm;

[0026] The scores for each of the six dimensions were calculated and plotted;

[0027] According to the development of the power grid and the updating of data, the assessment data and regional portraits are updated regularly, and the assessment indicators are optimized.

[0028] Preferably, the S4 is specifically as follows:

[0029] The grid capacity is regarded as a production process, and the input index and output index of the area to be analyzed are defined. For each determined index, its fuzzy set and membership function are defined; the fuzzified index is divided into input variables and output variables: n DMUs are set, and the i-th DMU is defined as DMU i , which consumes m input variables a ij , generating the output variable b ik , the relative efficiency value of the grid in each area is calculated based on the DEA model:

[0030]

[0031] The constraints are as follows:

[0032]

[0033] Among them, α k , β j are the weights of input variables and output variables respectively;

[0034] After calculating the efficiency values ​​of different DMUs, feature extraction is performed and input into the multi-layer perceptron to evaluate the grid capacity. The weight parameters in DEA and the parameters in the multi-layer perceptron are optimized based on the improved cuckoo algorithm.

[0035] Preferably, the improved cuckoo algorithm is as follows:

[0036] Taking the inverse of the evaluation error as the fitness value, randomly generate NP individuals, set the maximum number of iterations T, and execute the cuckoo algorithm to generate a new solution for the parameters:

[0037] S x,t+1 =S x,t +ξ t +L(χ)+(S x,t- S best )+θ(S x,t- S y,t )(μ x,t -μ y,t )

[0038] ξ t =ξ max -(ξ max -ξ min ) / T×t

[0039] μ x =|f best -f x | / |f best -f worst |

[0040] Among them, S x,t , S x,t+1 is the parameter value of the xth parameter before and after the tth iteration; ξ t is the iteration factor; max , min are the maximum and minimum values ​​of the iteration factor respectively; t is the number of iterations; L(χ) is the Levy flight formula; S best is the optimal parameter value in the tth iteration during the iteration process; θ is a random number between [0,1]; S y,t is the parameter value of the yth parameter at the tth iteration; μ x,t , μ y,t is the weight factor of the xth parameter and the yth parameter at the tth iteration; f best 、f worst is the upper and lower limits of parameter fitness; f x is the fitness value of the xth parameter;

[0041] Based on the particle swarm algorithm, a new solution for the individual is generated:

[0042] W x,t+1 =W x,t +γ x (W x,t -W y,t )

[0043] Among them, W x,t+1 , W x,t is the parameter value of the xth parameter before and after the tth iteration in the particle swarm algorithm; γ x Represents a random number between the interval [-1,1], W y,t is the parameter value of the tth iteration of the yth parameter in the particle swarm algorithm;

[0044] Q x,t+1 =ω×S x,t+1 +(1-ω)×W x,t+1

[0045] Among them, Q x,t+1 is the parameter value of the mixed iteration; ω is the weight coefficient;

[0046] The optimization selection is performed according to the fitness value, and the merge sorting is performed by mixing the previous generation population and the next generation population. The fitness values ​​are sorted from best to worst, and the NP parameter values ​​with the best fitness values ​​are selected to continue the subsequent iterations, and whether to end the iteration is determined according to the termination condition.

[0047] Compared with the prior art, the present invention provides a grid capacity evaluation system and method based on multi-dimensional regional portraits, which has the following beneficial effects:

[0048] The present invention evaluates the grid capacity in the area to be analyzed based on six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligence level and economic operation. It can more comprehensively and accurately reflect the actual capacity and operation status of the grid, and avoid the limitations and one-sidedness of single indicator evaluation; timely discover the unreasonable aspects of the grid structure, so as to optimize it and improve the reliability and stability of power supply; accurately identify the root causes of power quality problems, take corresponding governance measures, and improve power quality; determine aging or underperforming equipment, and provide a scientific basis for equipment updates and upgrades; evaluate the intelligence level, promote the development and application of intelligent technology, and formulate grid development strategies suitable for various regions; and establish multi-dimensional regional portraits, which can better predict future power grid needs and development trends, and provide strong support for forward-looking planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a system block diagram mentioned in Embodiment 1 of the present invention;

[0050] Figure 2 This is a flow chart of the method mentioned in Example 2 of the present invention;

[0051] Figure 3 It is the DMU total efficiency grid diagram mentioned in Example 2 of the present invention;

[0052] Figure 4 This is a schematic diagram of the structure of the multi-layer perceptron mentioned in Example 2 of the present invention. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0054] The present invention evaluates the grid capacity in the area to be analyzed based on six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligence level and economic operation. It can more comprehensively and accurately reflect the actual capacity and operation status of the grid, avoid the limitations and one-sidedness of single indicator evaluation; timely discover the unreasonable aspects of the grid structure, so as to optimize it and improve the reliability and stability of power supply; accurately identify the root causes of power quality problems, take corresponding governance measures, and improve power quality; determine aging or insufficient equipment, and provide a scientific basis for equipment updates and upgrades; evaluate the intelligence level, promote the development and application of intelligent technology, and formulate grid development strategies suitable for various regions; and establish a multi-dimensional regional portrait, which can better predict the future demand and development trend of the power grid and provide strong support for forward-looking planning. Specifically include the following contents.

[0055] Embodiment 1:

[0056] See also Figure 1 The grid capacity evaluation system based on multi-dimensional regional portrait of the present invention includes:

[0057] Data acquisition module 100: used to collect power grid related data in the area to be analyzed and perform preprocessing operations;

[0058] Preliminary evaluation module 200: used to calculate the line connection rate, complex wiring rate, and standard wiring rate in the area to be analyzed, and to perform a preliminary evaluation on the area to be analyzed;

[0059] Multi-dimensional portrait module 300: used to establish regional portraits from several dimensions respectively;

[0060] Capacity evaluation module 400: used to evaluate the grid capacity in the area to be analyzed based on the regional portrait established by the multi-dimensional portrait module 300;

[0061] The multi-dimensional portrait module 300 includes a grid power supply capacity unit 310, a power quality unit 320, a grid structure unit 330, an equipment level unit 340, an intelligence level unit 350 and an economic operation unit 360; they are used to establish regional portraits from six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligence level and economic operation.

[0062] Embodiment 2:

[0063] See also Figure 2-4 The grid capacity evaluation method based on multi-dimensional regional portrait of the present invention comprises the following steps:

[0064] S1: Collect the relevant data of the power grid in the area to be analyzed and perform preprocessing operations; the details are as follows:

[0065] Collect data such as the historical maximum power supply load, load forecast data, power outage frequency, duration, voltage deviation, frequency deviation, harmonic content, power grid connection mode, line corridor layout, network topology structure, etc. in the area, and perform pre-processing operations such as data cleaning and standardization.

[0066] S2: Calculate the line connection rate, complex wiring rate, and standard wiring rate in the area to be analyzed, and make a preliminary assessment of the area to be analyzed; the details are as follows:

[0067] The line contact rate is calculated as follows:

[0068] Line connection rate = number of reasonable lines / total number of lines × 100%

[0069] The number of lines with reasonable connection refers to the number of lines that meet the specific connection mode in the technical guidelines for distribution networks. For example, the overhead network is three-connection or single-connection, the cable network is a single-ring network or a double-ring network (single-connection), and the interconnected lines are the outgoing lines of different substations.

[0070] The complex wiring rate is calculated as follows:

[0071] The complex wiring rate may vary depending on the grid system, assessment purpose or specific study.

[0072] Complex wiring rate = (number of elements with complex wiring characteristics / total number of elements) × 100%

[0073] The number of elements with complex wiring characteristics is determined according to the specific definition, and the total number of elements is the corresponding reference total.

[0074] For example, if complex wiring is defined as wiring layouts with more than a certain number of cross-connections or circuitous connections, the number of wiring layouts with this characteristic can be counted and then divided by the total number of wiring layouts to obtain the complex wiring rate.

[0075] The standard wiring ratio is calculated as follows:

[0076] Standard wiring rate = number of large feeders using standard wiring / total number of large feeders × 100%

[0077] Among them, the large feeder with standard wiring refers to the large feeder that complies with the typical wiring method specified in the technical guidelines for distribution networks. For example, the overhead network is three-segment, three-connection or three-segment, single-connection, and the cable network is a single-ring network or a double-ring network.

[0078] It should be noted that in actual applications, specific definitions and calculation methods may vary according to different standards, regions or specific project requirements. The above formula only provides a common reference. In specific power grid analysis and evaluation, accurate calculation methods and indicator definitions can be determined based on relevant specifications, guidelines or specific research purposes. At the same time, the evaluation of these indicators also needs to be comprehensively analyzed in combination with the actual situation of the power grid to fully understand the performance and characteristics of the power grid.

[0079] For example, in power supply areas above Class C, the evaluation of line connection rate can be done by counting the number of 10 (20) kV public lines with reasonable connection, dividing it by the total number of 10 (20) kV public lines in the power supply area, and then multiplying it by 100% to get the line connection rate. Then, according to the set evaluation method (such as the five-segment evaluation method), it is judged whether the line connection situation in the area is good. Similarly, for the standard connection rate, it is also calculated by counting the number of large feeders using the specified standard connection method, dividing it by the total number of large feeders and multiplying it by 100%, and then evaluating it according to the corresponding evaluation method.

[0080] S3: Evaluate the grid capacity in the area to be analyzed based on six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligence level and economic operation, and establish a regional profile; the details are as follows:

[0081] Dimension 1: Grid power supply capacity, analysis indicators include maximum power supply load, load margin and power supply reliability;

[0082] Maximum power supply load: evaluate the maximum load that the regional power grid can bear and whether it can meet the current and future power demand. For example, the maximum power supply load in a certain area is 1 million kilowatts, but it is expected to increase to 1.2 million kilowatts in the next two years, which means there is a risk of insufficient power supply capacity.

[0083] Load margin: Calculates the difference between the grid's power supply capacity and actual load under normal operating conditions. For example, if the load margin in a certain area is only 10%, it means that the grid has less flexibility in dealing with sudden load increases.

[0084] Power supply reliability indicators: including power outage frequency, power outage duration, and average power supply availability, etc. For example, the region has more than 5 power outages per year, with an average power outage duration of 1.5 hours each time, and an average power supply availability of 99.5%, which is still far from the industry's best level.

[0085] Dimension 2: Power quality, analysis indicators include voltage deviation, frequency deviation and harmonic content;

[0086] Voltage deviation: monitors the degree to which the voltage of each node of the power grid deviates from the rated value. For example, in a certain industrial area, the voltage deviation of some nodes exceeds ±5%, affecting the normal operation of production equipment.

[0087] Frequency deviation: examines the stability of the grid frequency. For example, the frequency deviation in this area is always controlled within ±0.2Hz, and the power quality is good.

[0088] Harmonic content: Analyze the harmonic level in the power grid to determine whether it meets national standards. For example, the harmonic content in a commercial area exceeds the standard, causing some electronic equipment to malfunction.

[0089] Dimension 3: Grid structure: analysis indicators include wiring mode, line corridor layout and network topology;

[0090] Wiring mode: Study whether the power grid adopts radial, ring or grid wiring. For example, if a certain area mainly adopts radial wiring, a partial line failure may cause a large-scale power outage.

[0091] Line corridor layout: evaluate the rationality of line corridors and the efficiency of land resource utilization. For example, unreasonable line corridor planning conflicts with urban construction, resulting in increased line relocation costs.

[0092] Network topology: Analyze whether the topology of the power grid is simple, clear, easy to maintain and expand. For example, the network topology of a certain area is complex, which increases the difficulty of operation and management.

[0093] Dimension 4: Equipment level analysis indicators include transformer performance, switch reliability, and transmission line material and insulation level;

[0094] Transformer performance: Inspect the capacity, loss and service life of the transformer. For example, some transformers have been in operation for more than 20 years and have large losses, so they need to be replaced.

[0095] Switchgear reliability: Evaluate the number of switchgear operations, number of failures, and repair costs. For example, a high failure rate of switchgear in a certain area affects the normal operation of the power grid.

[0096] Transmission line material and insulation level: Analyze whether the material and insulation performance of the transmission line meet the requirements. For example, the material and insulation level of some old lines are low, which poses a safety hazard.

[0097] Dimension 5: Intelligent level: Analysis indicators include automated monitoring systems, intelligent scheduling capabilities, and big data analysis applications;

[0098] Automated monitoring system: Check whether the power grid has a complete automated monitoring system that can monitor the power grid operation status in real time. For example, a certain area has achieved real-time monitoring and data collection of the entire network, which can quickly discover and handle faults.

[0099] Intelligent dispatching capability: evaluates the intelligent dispatching level of the power grid and whether it can optimize dispatching according to load changes and operating conditions. For example, through the intelligent dispatching system, the region reasonably arranges the shutdown of units during the low electricity consumption period, reducing operating costs.

[0100] Big data analysis and application: examine the power grid's ability to analyze and apply big data to support decision-making. For example, use big data analysis to predict load changes and make grid planning and equipment maintenance arrangements in advance.

[0101] Dimension 6: Economic operation: analysis indicators include network loss rate, return on investment and operation and maintenance level;

[0102] Network loss rate: Calculate the loss ratio of the power grid during the transmission of electric energy. For example, the network loss rate in a certain area is 5%, which is higher than the industry average, and measures need to be taken to reduce the loss.

[0103] Return on investment: Evaluate the return on investment of power grid construction and transformation projects. For example, the return on investment of a power grid upgrade project reaches 15%, which has good economic benefits.

[0104] Operation and maintenance costs: Analyze the daily operation and maintenance costs of the power grid, including human, material and financial inputs. For example, a region reduced its operation and maintenance costs by 10% by optimizing operation and maintenance management.

[0105] Comprehensive evaluation of the above six dimensions can fully understand the capacity and existing problems of the regional grid, and provide a basis for further optimization and improvement. For example, if there is a shortage in the grid power supply capacity, you can consider building a new substation or upgrading the transmission line; if there is a problem with the power quality, you can install reactive power compensation devices or filtering equipment; when the grid structure is unreasonable, you can carry out line transformation and optimize the layout. By continuously improving the indicators of each dimension, the overall performance of the regional grid can be improved to ensure the safe, reliable, high-quality and economical supply of electricity.

[0106] Quantify each dimension indicator and set the corresponding scoring indicator to calculate the score of each dimension, for example:

[0107] Grid power supply capacity: maximum power supply load satisfaction (actual maximum load / predicted maximum load) × 100%, 1 point for less than 80%, 2 points for 80%-90%, 3 points for 90%-100%, and 4 points for more than 100%.

[0108] Power quality: 4 points for voltage deviation within ±5%, 3 points for ±5%-±7%, 2 points for ±7%-±10%, and 1 point for voltage deviation exceeding ±10%.

[0109] For some indicators that are difficult to quantify directly, evaluation can be carried out through expert scoring or questionnaire surveys.

[0110] Then, the weight values ​​of different indicators in each dimension are calculated based on the improved cuckoo algorithm; the improved cuckoo algorithm is as follows:

[0111] Taking the inverse of the evaluation error as the fitness value, randomly generate NP individuals, set the maximum number of iterations T, and execute the cuckoo algorithm to generate a new solution for the parameters:

[0112] S x,t+1 =S x,t +ξ t +L(χ)+(S x,t- S best )+θ(S x,t- S y,t )(μ x,t -μ y,t )

[0113] ξ t =ξ max -(ξ max -ξ min ) / T×t

[0114] μ x =|f best -f x | / |f best -f worst |

[0115] Among them, S x,t , S x,t+1 is the parameter value of the xth parameter before and after the tth iteration; ξ t is the iteration factor; max , min are the maximum and minimum values ​​of the iteration factor respectively; t is the number of iterations; L(χ) is the Levy flight formula; S best is the optimal parameter value in the tth iteration during the iteration process; θ is a random number between [0,1]; S y,t is the parameter value of the yth parameter at the tth iteration; μ x,t , μ y,t is the weight factor of the xth parameter and the yth parameter at the tth iteration; f best 、f worst is the upper and lower limits of parameter fitness; f x is the fitness value of the xth parameter;

[0116] Based on the particle swarm algorithm, a new solution for the individual is generated:

[0117] W x,t+1 =W x,t +γ x (W x,t -W y,t )

[0118] Among them, W x,t+1 , W x,t is the parameter value of the xth parameter before and after the tth iteration in the particle swarm algorithm; γ x Represents a random number between the interval [-1,1], W y,t is the parameter value of the tth iteration of the yth parameter in the particle swarm algorithm;

[0119] Q x,t+1 =ω×S x,t+1 +(1-ω)×W x,t+1

[0120] Among them, Q x,t+1 is the parameter value of the mixed iteration; ω is the weight coefficient;

[0121] The optimization selection is performed according to the fitness value, and the merge sorting is performed by mixing the previous generation population and the next generation population. The fitness values ​​are sorted from best to worst, and the NP parameter values ​​with the best fitness values ​​are selected to continue the subsequent iterations, and whether to end the iteration is determined according to the termination condition.

[0122] The scores of each of the six dimensions can be calculated comprehensively, and the scores of different regions on the same dimension can be compared to find out the advantages and disadvantages. A radar chart or bar chart can be drawn to intuitively display the scores of regional power grids.

[0123] Based on the data analysis results, each dimension is described in detail, including the current status, existing problems and directions for improvement. For example, in terms of grid power supply capacity, the current maximum power supply load can meet current demand, but with the development of regional economy, there may be a gap in the next two years, and new substation construction needs to be planned in advance. In terms of power quality, voltage deviation is well controlled, but the harmonic content exceeds the standard in some areas, and harmonic control measures need to be strengthened.

[0124] A complete regional power grid capability portrait report is formed in text form to provide reference for power grid planning, construction and operation.

[0125] As the power grid develops and data accumulates, assessment data and portraits are updated regularly.

[0126] According to new situations and needs, optimize evaluation indicators and methods to make the portrait more accurate and practical.

[0127] Through the above steps, a comprehensive and accurate portrait of regional power grid capabilities can be established, providing strong support for improving the performance and service quality of the regional power grid.

[0128] S4: Evaluate the grid capacity in the area to be analyzed based on the regional portrait established in S3 using the DEA-ANN method. The details are as follows:

[0129] The grid capacity is regarded as a production process, and the input and output indicators of the area to be analyzed are defined. For each determined indicator, its fuzzy set and membership function are defined; the fuzzified indicators are divided into input variables and output variables: the DMUs corresponding to the 30 sub-areas in the area to be analyzed are set, and the i-th DMU is defined as DMU i , which consumes 30×4 input variables a ij , generating 30×8 output variables b ik , the relative efficiency value of the grid in each area is calculated based on the DEA model:

[0130]

[0131] The constraints are as follows:

[0132]

[0133] Among them, α k , β j are the weights of input variables and output variables respectively;

[0134] Based on the calculated efficiency values, the Figure 3 The grid is shown. Feature extraction is also performed based on the efficiency values ​​of different DMUs and input into a multi-layer perceptron. The structure of the multi-layer perceptron can be referred to Figure 4 , the grid capacity is evaluated, where the weight parameters in DEA and the parameters in the multi-layer perceptron are optimized based on the improved cuckoo algorithm.

[0135] Combining the line connection rate, complex wiring rate, standard wiring rate in S2 and the evaluation process in S4, 30 sub-areas were evaluated. The specific results can be found in Table 1:

[0136] Table 1 Evaluation results of 30 sub-areas

[0137]

[0138]

[0139] Based on the above evaluation results, we can determine which areas have relatively strong grid capacity and which areas have relatively weak grid capacity based on the evaluation scores given by the multi-layer perceptron. At the same time, we can compare other evaluation indicators such as line connection rate, complex wiring rate, standard wiring rate, etc. to find out the problems or areas that can be improved.

[0140] Based on the analysis of the evaluation results, the specific improvement goals are clarified, for example, whether to improve the grid power supply reliability in a specific area, reduce the rate of complex wiring, or improve the overall grid operation efficiency.

[0141] For areas with low grid capacity assessment scores, further investigate the details of their line layout, equipment conditions, load distribution, etc. Analyze possible causes of the problem, such as line overload, unreasonable connection, complex wiring, etc.

[0142] If the line connection rate is low, you can consider adding new connection lines or optimizing existing connection methods to improve the power supply reliability and flexibility of the power grid. Ensure that the planning of new connection lines is reasonable and will not cause new problems or risks.

[0143] For areas with high complex wiring rates, re-examine the wiring method, remove unnecessary detours and cross-connections, and simplify the wiring structure. This helps reduce the possibility of failures and improves operation and maintenance efficiency.

[0144] According to the requirements of the standard wiring method, the wiring that does not meet the standard is modified or adjusted. The standard wiring method is usually verified by practice and has good performance and reliability.

[0145] Check the operating status of the equipment and replace aging or faulty equipment in a timely manner. At the same time, consider using more advanced and better-performing equipment to improve the overall capacity of the grid.

[0146] If conditions permit, distributed power sources can be reasonably connected to enhance the power supply capacity and stability of the regional power grid.

[0147] Before implementing specific improvement measures, power system simulation software can be used to simulate and analyze the improvement plan, predict the effect after the improvement, and verify the feasibility and effectiveness of the plan.

[0148] According to the determined improvement plan, various measures shall be implemented step by step. During the implementation process, relevant standards and specifications shall be strictly followed to ensure the quality and safety of construction.

[0149] After the improvement measures are implemented, the operation of the regional grid is continuously monitored, and the indicators before and after the improvement are compared to evaluate whether the improvement effect has achieved the expected goal. If the effect is not ideal, further analysis of the reasons is required to adjust the improvement plan.

[0150] The experience and lessons learned in the improvement process are summarized and fed back into the subsequent grid planning, design and operation management to form a closed-loop management and continuously improve the capacity and level of the regional grid. The above is only a preferred specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field within the technical scope disclosed by the present invention, according to the technical solution and its inventive concept of the present invention, shall be covered within the protection scope of the present invention.

Claims

1. A grid capacity evaluation system based on multi-dimensional regional portraits, characterized by: include: Data collection module (100): used to collect power grid related data in the area to be analyzed and perform pre-processing operations; A preliminary evaluation module (200): used to calculate the line connection rate, complex wiring rate, and standard wiring rate in the area to be analyzed, and to perform a preliminary evaluation on the area to be analyzed; Multi-dimensional portrait module (300): used to create regional portraits from several dimensions; A capability evaluation module (400): used to evaluate the grid capability in the area to be analyzed based on the area portrait established by the multi-dimensional portrait module (300); The multi-dimensional portrait module (300) includes a grid power supply capacity unit (310), a power quality unit (320), a grid structure unit (330), an equipment level unit (340), an intelligent level unit (350) and an economic operation unit (360); and is used to establish a regional portrait from six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligent level and economic operation.

2. The grid capacity evaluation method based on multi-dimensional regional portrait according to claim 1 is characterized in that: The steps include: S1: Collect relevant data of the power grid in the area to be analyzed and perform preprocessing operations; S2: Calculate the line connection rate, complex wiring rate, and standard wiring rate in the area to be analyzed, and make a preliminary assessment of the area to be analyzed; S3: Evaluate the grid capacity in the area to be analyzed based on six dimensions: grid power supply capacity, power quality, grid structure, equipment level, intelligence level and economic operation, and establish a regional profile; S4: Evaluate the grid capacity in the area to be analyzed based on the DEA-ANN method according to the regional portrait established in S3.

3. The grid capacity evaluation method based on multi-dimensional regional portrait according to claim 2 is characterized in that: The analysis indicators of the grid power supply capacity include the maximum power supply load, load margin and power supply reliability; The analysis indicators of the power quality include voltage deviation, frequency deviation and harmonic content; The analysis indicators in the grid structure include wiring mode, line corridor layout and network topology; The analysis indicators in the equipment level include transformer performance, switch reliability and transmission line material and insulation level; The analytical indicators in the intelligence level include automated monitoring system, intelligent scheduling capability and big data analysis application; The analysis indicators in the economic operation include network loss rate, return on investment and operation and maintenance level.

4. The grid capacity evaluation method based on multi-dimensional regional portrait according to claim 3 is characterized in that: In S3, the six dimensional indicators of the evaluation are comprehensively analyzed and a regional portrait is drawn, as follows: Quantify each dimension indicator and set corresponding scoring indicators to calculate the score of each dimension. Calculate the weight value of different indicators in each dimension based on the improved cuckoo algorithm; The scores for each of the six dimensions were calculated and plotted; According to the development of the power grid and the updating of data, the assessment data and regional portraits are updated regularly, and the assessment indicators are optimized.

5. The grid capacity evaluation method based on multi-dimensional regional portrait according to claim 4 is characterized in that: The S4 is specifically as follows: Take the grid capacity as a production process, define the input index and output index of the area to be analyzed, and define the fuzzy set and membership function for each determined index; divide the fuzzified index into input variables and output variables: set n indivual DMU , among which i indivual DMU Defined as DMU i , its consumption m Input variables a ij , generating output variables b ik , the relative efficiency value of the grid in each area is calculated based on the DEA model: The constraints are as follows: in, α k , β j are the weights of input variables and output variables respectively; Calculated different DMU The efficiency value is obtained and then feature extraction is performed and input into the multi-layer perceptron to evaluate the grid capacity. The weight parameters in DEA and the parameters in the multi-layer perceptron are optimized based on the improved cuckoo algorithm.

6. The grid capacity evaluation method based on multi-dimensional regional portrait according to claim 5 is characterized in that: The improved cuckoo algorithm is specifically as follows: Take the inverse of the evaluation error as the fitness value, randomly generate NP individuals, and set the maximum number of iterations T , execute the cuckoo algorithm to generate a new solution for the parameters: S x,t+1 = S x,t + ξ t + L (x)+( S x,t- S best )+ θ ( S x,t- S y,t )( μ x,t - μ y,t ) ξ t = ξ max -( ξ max - ξ min ) / T × t μ x =| f best - f x | / | f best - f worst | in, S x,t , S x,t+1 For the x Parameter t Parameter values ​​before and after iterations; ξ t is the iteration factor; ξ max , ξ min are the maximum and minimum values ​​of the iteration factor respectively; t is the number of iterations; L (χ) is the Levy flight formula; S best In the iteration process t The optimal parameter value in the iteration; θ is a random number between [0,1]; S y,t For the y Parameter t The parameter value for the iteration; μ x,t , μ y,t For the x Parameters, y Parameter t The weight factor of the iteration; f best , f worst is the upper and lower limits of the parameter fitness; f x For the x The fitness value of each parameter; Based on the particle swarm algorithm, a new solution for the individual is generated: W x,t+1 = W x,t +g x ( W x,t - W y,t ) in, W x,t+1 , W x,t In the particle swarm algorithm x Parameter t The parameter values ​​before and after the iteration; γ x Represents a random number between the interval [-1,1], W y,t In the particle swarm algorithm y Parameter t The parameter value for the iteration; Q x,t+1 =ω× S x,t+1 +(1-ω)× W x,t+1 in, Q x,t+1 is the parameter value of the mixed iteration; ω is the weight coefficient; The optimization selection is performed according to the fitness value, and the merge sorting is performed by mixing the previous generation population and the next generation population. The fitness values ​​are sorted from best to worst, and the NP parameter values ​​with the best fitness values ​​are selected to continue the subsequent iterations, and whether to end the iteration is determined according to the termination condition.

Images