Power distribution network investment strategy optimization method based on development demand and input-output evaluation

Abstract

The invention provides a power distribution network investment strategy optimization method based on development demands and input-output evaluation, and belongs to the technical field of power distribution network investment strategy optimization. In order to solve the technical problems that the existing power distribution network investment decision is difficult to comprehensively consider various factors, accurately evaluate investment benefits and reasonably plan an investment strategy, the investment optimization capability is evaluated by constructing an investment optimization evaluation model of a power distribution network; the method comprises the following steps: determining an index weight, calculating the weight based on actual data input and output of a region, carrying out index weight normalization, and finally determining an evaluation grade; on the basis of performing K-means clustering grid division on the latitude and longitude of a power grid of a single project, an input-output benefit evaluation index system taking economic benefits, social benefits and technical benefits as first-level indexes is constructed, and a multi-element membership model is constructed based on the correlation between investment and the indexes; the method is applied to power distribution network investment evaluation.

Description

Technical Field

[0001] This invention provides a method for optimizing distribution network investment strategies based on development needs and input-output evaluation, belonging to the field of distribution network investment strategy optimization technology. Background Technology

[0002] Based on the booming development of distributed energy and new loads in the new power system, the demand for accuracy and foresight in distribution network investment decision optimization is becoming increasingly prominent. At present, the integration of distributed energy consumption and the growth characteristics of new loads into the collaborative management of distribution network investment planning is carried out to achieve precise investment and improve the overall efficiency of the power grid. Distribution network investment projects are analyzed in detail according to three dimensions: economic benefits, social benefits, and technical benefits. Under the constraints of multi-objective fuzzy functions, a scientific and reasonable investment project screening and combination decision-making technology is constructed. The weights are calculated and graded by the decision experiment and evaluation laboratory comprehensive evaluation method (DEMATEL-AHP) to achieve the precise investment goals of the distribution network.

[0003] However, current research on distribution network investment strategies is mostly limited to the technical feasibility and economic rationality analysis of single projects, focusing solely on the local benefits of each project. In terms of project correlation analysis, traditional methods neglect the synergistic effects and mutual exclusion relationships between projects, leading to unreasonable investment allocation. Regarding weight determination, existing methods are highly subjective and lack objective quantitative basis. In multi-objective optimization, traditional methods struggle to balance indicators with different dimensions. When facing complex distribution network investment environments, this research approach has significant shortcomings. It fails to comprehensively consider the overall investment capacity of the region, nor does it adequately account for the differences and changes in investment needs across different regions and stages, neglecting limitations on investment scale and the rationality of fund allocation.

[0004] Furthermore, the analysis lacked in-depth exploration of the synergistic effects and mutual exclusion relationships between different projects, and failed to consider multiple projects as an organic whole. This made it difficult for investment decisions to adapt to the diversified and dynamic development trend of the power distribution network, to accurately reflect the actual needs of power distribution network investment, and to achieve optimal resource allocation. Summary of the Invention

[0005] To address the technical problems existing in the background art, the present invention adopts the following technical solution: a method for optimizing distribution network investment strategies based on development needs and input-output evaluation, comprising the following optimization steps:

[0006] Step 1: Construct an investment optimization evaluation model for the power distribution network to assess its investment optimization capabilities;

[0007] Step 2: Constructing a decision-making experiment and evaluation laboratory - When evaluating the investment optimization of power grid planning using the analytic hierarchy process (AHP) model, it is necessary to determine the weights of the indicators, calculate the weights based on the actual input and output data of the region, normalize the indicator weights, and finally determine the evaluation level.

[0008] Step 3: When evaluating the power grid planning investment optimization decision, based on the K-means clustering grid division of the power grid latitude and longitude of individual projects, an input-output benefit evaluation index system with economic benefits, social benefits, and technical benefits as primary indicators is constructed, and a multi-element membership model is constructed based on the correlation between investment and indicators.

[0009] Step 4: The input-output indicators for power grid investment optimization are weighted and solved using a decision experiment and evaluation laboratory-analytic hierarchy process (AHP) comprehensive evaluation method. The evaluation level of the power grid investment plan is then determined based on the simulation results.

[0010] The multi-element membership model constructed in step three specifically includes: constructing a correlation matrix of distribution network investment evaluation indicators, a multi-objective optimization model for distribution network investment, and a multi-objective fuzzy function for distribution network investment.

[0011] The specific method for grid division of the power grid latitude and longitude of a single project based on K-means clustering in step three is as follows:

[0012] In the application scenario of power grid investment planning, K-Means clustering is performed on the power grid longitude and latitude dataset. Each data object point is assigned to the cluster with the closest distance. Then, based on the coordinate mean of the members in the cluster, the cluster center of each cluster is recalculated. This process continues until the cluster center no longer changes, reaches the preset convergence range, or reaches the predetermined number of iterations.

[0013] The elbow algorithm was used to verify the optimal number of clusters:

[0014] The sum of squared intra-cluster errors is calculated iteratively for 1 to 10 different numbers of clusters. K-means clustering is used, and the sum of squared distances from the data point to its cluster center is calculated based on the cluster index and cluster center to obtain the sum of squared intra-cluster errors.

[0015] After completing the calculations, draw an elbow rule diagram, with the horizontal axis representing the number of clusters and the vertical axis representing the sum of squared errors within each cluster.

[0016] The optimal number of clusters is determined by calculating the decreasing slope of the sum of squared errors within adjacent clusters. When the slope is less than half of the average slope, the index of that point is incremented by 1 to obtain the cluster number corresponding to the elbow, which is the optimal number of clusters.

[0017] The specific method for constructing the input-output index system of the power distribution network in step four to determine the evaluation level of power grid investment planning is as follows:

[0018] By adapting to actual conditions and differences in power grid characteristics, an investment calculation model that conforms to the personalized development trend of the power grid is constructed. The primary indicators of the investment calculation model include indicators in three dimensions: economic benefits, social benefits, and technical benefits of the distribution network.

[0019] Economic benefit indicators include equipment purchase, site construction, and preliminary planning costs;

[0020] Social benefit indicators include the ability of users to obtain diverse and integrated energy sources to meet different needs;

[0021] Technical performance indicators include power supply reliability, overall line loss rate, average outage time, overall voltage qualification rate, and capacity-to-load ratio.

[0022] The secondary indicators of economic benefits in the primary indicators of the investment calculation model include:

[0023] The initial investment cost is calculated using the following formula:

[0024] Initial investment cost = electrical equipment purchase cost + material and construction cost + building construction cost + project planning cost + feasibility study cost + survey and design cost;

[0025] The annual average operation and maintenance cost is calculated using the following formula:

[0026] Annual average maintenance cost = Average maintenance cost over the past T years;

[0027] Net present value, calculated using the following formula:

[0028] Net present value = Annual cash inflows - Cash outflows;

[0029] Discount the present value using the discount rate, add up the discounted values ​​for each year, and then subtract the total cost of the project.

[0030] The internal rate of return (IRR) is calculated using the following formula:

[0031] Internal rate of return (IRR) = the difference between cash inflows and cash outflows in each period from period 0 to period n, divided by (1 + number of periods of IRR) raised to the power of (1 + number of periods of IRR), then sum the results of each period and set the sum to 0.

[0032] The static payback period is calculated using the following formula:

[0033] Static payback period = Initial investment cost / Annual net cash flow. The static payback period is calculated as follows: (Number of years in which cumulative net cash flow first becomes positive) + (Number of years in which cumulative net cash flow first becomes positive) + (Number of years in which the investment has not yet been recovered ...

[0034] The dynamic payback period is calculated using the following formula:

[0035] Dynamic payback period = [(Number of years in which cumulative net cash flow first becomes positive - 1) + Present value of investment not yet recovered in the previous year] / Present value of net cash flow in that year;

[0036] In this process, the net cash flow for each year needs to be discounted using a discount rate.

[0037] The secondary indicators of social benefits in the primary indicators of the investment calculation model include:

[0038] The formula for reducing user spending costs is as follows:

[0039] Reduce user expenditure costs = (from year 1 to year N) annual reduction in user electricity consumption × electricity price / 1 + (discount rate raised to the power of the number of years).

[0040] The carbon emission reduction benefits of the power grid are calculated using the following formula:

[0041] The carbon emission reduction benefits of the power grid (i.e., the quantitative manifestation of the carbon emission reduction benefits of the power grid) = the difference in the amount of electricity generated by replacing high-carbon power sources with low-carbon or zero-carbon power sources due to measures such as power grid optimization and dispatch * emission factor.

[0042] The formula for calculating the regional tax contribution rate is as follows:

[0043] Regional tax contribution rate = (total amount of various taxes actually paid by power grid companies in the region / total tax revenue of the region in the same period) × 100%.

[0044] The secondary indicators of technical benefits in the primary indicators of the investment calculation model include:

[0045] The power supply reliability rate is calculated using the following formula:

[0046] Power supply reliability rate = 1 minus the sum of the products of the number of users experiencing power outages for each type of outage event and the average outage time for that type of outage event, divided by the total number of hours in a year, then multiplied by 100%;

[0047] The comprehensive line loss rate is calculated using the following formula:

[0048] Overall line loss rate = (Power supply - Power sales) / Power supply;

[0049] The average power outage time is calculated using the following formula:

[0050] Average power outage time = (Number of users experiencing each type of power outage event + product of average power outage time for that type of power outage event) / Total number of users.

[0051] The specific method for constructing the correlation matrix of power distribution network investment evaluation indicators is as follows:

[0052] The correlation analysis method was used to simplify the investment benefit evaluation index system of power grid enterprises, and the investment benefit evaluation index was defined. Divided into The class, expressed as:

[0053] ;

[0054] Each row above represents a category of indicators. A correlation matrix analysis is performed on the indicators in each row. The closer the absolute value of the correlation coefficient between two indicator data is to 1, the stronger the correlation. The correlation coefficient is used to... The range of change of two indicators is used to determine the degree of correlation between them, and the expression is:

[0055] ;

[0056] in, , These are the averages of the two exponents, satisfying:

[0057] .

[0058] The specific method for constructing a multi-objective optimization model for power distribution network investment is as follows:

[0059] The weights of each evaluation indicator were determined using the comprehensive evaluation method of decision-making experiments and evaluation laboratories, including the following steps:

[0060] Define the universe of discourse of the elements of the evaluated object as follows: ;

[0061] Define the universe of evaluation gradation as ;

[0062] Establish a fuzzy relation matrix R for single-element evaluation, and evaluate the j-th evaluation object in the membership model. Membership values ​​of each indicator The calculation yields the fuzzy relation matrix, expressed as follows:

[0063] ;

[0064] The DEMATEL-AHP model is used to obtain the weight values ​​of each evaluation index. The entropy weight of each indicator is calculated based on the empirical entropy weight method. The calculation formula is:

[0065] ;

[0066] By calculating the proportionality coefficient and To obtain the final weight values, in the formula... Indicates the first The first evaluation object One indicator;

[0067] The fuzzy relation matrix was obtained through calculation. and index weight vector Through the combined operation of these two, the final result is obtained. The fuzzy comprehensive evaluation result vector of each evaluation object The calculation formula is:

[0068] ;

[0069] in It is by and The The column operation yields a value representing the degree of membership of the rated object to the fuzzy subset as a whole.

[0070] Assuming that the weights of each evaluation indicator in the investment efficiency evaluation system have been determined as follows: Based on this, the evaluation object A fuzzy assessment is made of the overall efficiency level, and the assessment expression is:

[0071] ;

[0072] in, As an indicator The weight value, As an indicator The evaluation value.

[0073] The specific method for constructing a multi-objective fuzzy function for power distribution network investment is as follows:

[0074] A fuzzy multi-objective optimization method is adopted, and trapezoidal curves are used to fuzzify each sub-objective function, as expressed in the following expression:

[0075] ;

[0076] In the formula: This represents the optimal value for a single objective indicator, such as operational efficiency. The difference between the optimal value for a single objective;

[0077] For positive indices related to space frame structures, the membership function expression is:

[0078] ;

[0079] In the formula: This represents the optimal value of a single-objective optimization for a positive indicator. The difference between the optimal value for a single objective;

[0080] Investor satisfaction is used To indicate, to make It is the minimum value among all values ​​after blurring all sub-targets, that is, it satisfies:

[0081] ;

[0082] To transform a multi-objective optimization problem into a single-objective nonlinear optimization problem, it is only necessary to satisfy the satisfaction index under all constraints. To obtain the maximum value, the expression of the fuzzy multi-objective optimization model becomes:

[0083] ;

[0084] Then, a fuzzy model conversion is performed, including the following steps:

[0085] Input historical data of the distribution network, and solve the distribution network problem. This problem involves a single-objective optimization problem aimed at maximizing the improvement effect of each development direction. The goal is to find the optimal expected value for all development directions within this single-objective optimization problem. and minimum improvement expectation ;

[0086] Set the scaling factor The difference between various development directions After adjustment, its value range becomes:

[0087] ;

[0088] in, , The value of reflects the power grid company's subjective intention to invest. The smaller the value, the stronger the power grid company's willingness to address this weakness, and the lower its tolerance for the indicator failing to reach its optimal value.

[0089] The present invention has the following advantages over the prior art:

[0090] I. Scientific and reasonable grid division of the power distribution network based on K-means clustering. By analyzing the distribution characteristics of power grid coordinate data, the power grid is clustered into 5 clusters. This division takes into account data characteristics, distribution patterns, and practical application needs, and is highly effective in regional division and management, as well as improving the accuracy of analysis. It can provide an effective data division basis for subsequent power grid analysis and planning.

[0091] II. Construct a comprehensive and systematic input-output benefit evaluation index system. This system covers multiple levels, including economic, social, and technological benefits, and defines the meaning and calculation methods of each index in detail. It comprehensively assesses the benefits of power grid investment and provides multi-faceted references for investment decisions.

[0092] Third, correlation analysis was used to simplify the evaluation index system for the investment benefits of power grid enterprises. Redundant information in highly similar index clusters was identified and eliminated, making the evaluation indicators more concise and effective, and providing a more reliable basis for accurately assessing power grid investment needs.

[0093] Fourth, the DEMATEL-AHP method is used to determine the weights of indicators and combined with the fuzzy comprehensive evaluation method for investment benefit assessment. This approach can not only deeply analyze the interactions and coupling relationships between indicators, but also accurately determine the weights of each indicator, and comprehensively evaluate investment benefits, thereby improving the accuracy and scientific rigor of investment benefit assessment.

[0094] Fifth, the multi-objective optimization problem of distribution network investment allocation is transformed into a single-objective optimization problem through multi-objective function fuzzification. Trapezoidal curves are used to fuzzify each sub-objective function, and scaling coefficients are set to adjust the difference. Under the constraint conditions, investor satisfaction is maximized to determine the reasonable allocation of distribution network project investment. Attached Figure Description

[0095] The present invention will be further described below with reference to the accompanying drawings:

[0096] Figure 1 This is a flowchart illustrating the steps of the power distribution network construction planning method of the present invention;

[0097] Figure 2 This is a flowchart illustrating the steps of using the K-Means clustering algorithm in this embodiment of the invention.

[0098] Figure 3 This is a schematic diagram of the evaluation index system used in an embodiment of the present invention;

[0099] Figure 4 This is a flowchart illustrating the steps involved in the evaluation using the DEMATEL-AHP model in an embodiment of the present invention. Detailed Implementation

[0100] like Figures 1 to 4As shown, this invention addresses the technical problem of existing distribution network investment decisions struggling to comprehensively consider multiple factors, accurately assess investment benefits, and rationally plan investment strategies. It proposes a distribution network investment strategy optimization method based on development needs and input-output evaluation. This method includes: constructing an investment capacity prediction model to comprehensively analyze internal financial conditions, external factors, and constraints to predict investment capacity; constructing an evaluation model based on K-means clustering, the DEMATEL-AHP model, and a multi-objective function fuzzy evaluation model to determine investment needs; when investment capacity cannot meet the needs, constructing an optimization evaluation system for power grid project planning investment strategies from three dimensions: economic benefits, social benefits, and technical benefits; constructing an objective function to maximize the investment benefits of different project sets and setting constraints such as investment amount, investment needs, and project construction; and using an improved algorithm to solve the objective function, constructing a multi-objective optimization model for distribution network investment. Simultaneously, it achieves quantitative evaluation of power grid investment planning, providing a scientific basis for distribution network investment decisions.

[0101] The preferred embodiments of this study are described in detail below with reference to the accompanying drawings. The drawings constitute a part of this study and are used together with the embodiments to illustrate the principles of this study, but are not intended to limit the scope of this study.

[0102] An embodiment of the present invention discloses a method for power grid investment planning and analysis in a certain area, comprising:

[0103] In step S1, the power grid of a certain area is divided into grids based on K-means clustering;

[0104] In step S2, an input-output efficiency evaluation index system is established;

[0105] In step S3, a correlation analysis is performed between investment and indicators;

[0106] In step S4, a multi-objective optimization model for power distribution network investment is constructed and evaluated using the fuzzy comprehensive evaluation method. At the same time, the multi-objective function is fuzzified to determine the investment allocation scheme.

[0107] Compared with existing technologies, the power grid investment planning and analysis method provided in this embodiment firstly uses K-means clustering to rationally divide the power grid into grids, laying the foundation for subsequent accurate analysis. Secondly, it establishes a comprehensive input-output benefit evaluation index system to assess investment benefits from multiple perspectives. Thirdly, it uses correlation analysis to simplify the index system and improve evaluation efficiency. Finally, it constructs a multi-objective optimization model and combines fuzzy comprehensive evaluation and multi-objective function fuzzification to scientifically determine the investment allocation scheme, enhancing the scientific nature and rationality of power grid investment decisions.

[0108] The following text, with reference to relevant diagrams (such as the flowchart of the K-Means algorithm and the flowchart of the DEMATEL-AHP evaluation model), details each step of the power grid investment planning and analysis method in a certain region according to the embodiment of this study.

[0109] In step S1, the power grid of a certain region is divided into grids based on K-means clustering. The K-Means algorithm is a classic data mining algorithm. It calculates the distance between each data object in the dataset and each cluster center, assigns the power grid coordinate points to the nearest cluster, and recalculates the cluster centers based on the mean of the coordinate points within each cluster, iterating until convergence is achieved. Five points are randomly selected from the power grid coordinate data as initial cluster centers. Euclidean distance is used to measure the distance, ultimately clustering the power grid coordinates of a certain region into five clusters, thus achieving grid division. The elbow algorithm is used to verify the rationality of the number of clusters. By iteratively calculating the sum of squared errors (SSE) within clusters under different numbers of clusters and plotting the elbow rule diagram, it is determined from the elbow algorithm perspective that clustering the power grid of a certain region into five clusters achieves a good balance between intra-cluster compactness and inter-cluster separation, conforms to the data characteristics and distribution patterns, meets practical application needs, is beneficial for regional division and management, and improves the accuracy of analysis.

[0110] Based on the above method, K-means clustering is used to divide the latitude and longitude coordinates of the power grid into a grid. Given a parameter k, it divides the dataset into k clusters, ensuring high similarity between data objects within each cluster and low similarity between data objects in different clusters. In the clustering calculation, it calculates the distance between each data object and its cluster center, assigning each data object to the nearest cluster. Then, it recalculates the cluster center based on the mean coordinates of the cluster members. Generally, this process iterates until the objective function is minimized. Let p represent a data object; the objective function is typically a squared error criterion function.

[0111] (1);

[0112] Among them, c i Indicates cluster C i The mean, that is:

[0113] (2);

[0114] The distance metric in this objective function is Euclidean distance, but other distance metrics can also be used, such as Mahalanobis distance, to analyze elliptical clusters.

[0115] The K-Means algorithm can be described as follows:

[0116] Based on K-means clustering grid partitioning, the elbow algorithm is used to verify the optimal number of clusters to validate the rationality of the cluster number. The sum of squared intra-cluster errors (SSE) is calculated iteratively for different cluster numbers from 1 to 10. During the calculation, the K-means algorithm is used to cluster the data, obtaining the cluster index of each data point and the center of each cluster. The sum of squared distances from each data point to its cluster center is then calculated as the SSE. Compared with other cluster numbers, according to the elbow algorithm diagram, if a good balance is achieved between intra-cluster compactness and inter-cluster separation, the clustering result is considered reasonable and can provide a more effective data partitioning basis for subsequent power grid analysis and planning.

[0117] Further expanding upon the aforementioned methods, a comprehensive evaluation system for distribution network investment benefits is constructed to fully assess investment benefits. This includes considering that distribution network investment benefits are influenced by multiple dimensions of factors, including economic, social, and technical benefits. Among these, social and technical benefits indirectly affect economic benefits by influencing the operational stability of the power grid and user satisfaction. The comprehensive evaluation system for power grid investment planning is divided into three dimensions: economic, social, and technical benefits. Simultaneously, considering the interrelationships and weight differences among various indicators, and combining the principles of the Analytic Hierarchy Process (AHP) and the DemATEL (Decision Experiment and Evaluation Laboratory) approach, the weights of each indicator are determined. Through the dynamic transmission mechanism of causal relationships among various indicators, a comprehensive, objective, and accurate evaluation of investment benefits is achieved, enabling the evaluation results to more accurately reflect the actual benefits of distribution network investment.

[0118] Based on further improvements to the above methods, a comprehensive evaluation model for the investment benefits of power distribution networks is constructed to assess investment benefits. This includes constructing a comprehensive evaluation index system for investment benefits in the following way: Comprehensive benefit score = Economic benefit score × Economic benefit weight + Social benefit score × Social benefit weight + Technical benefit score × Technical benefit weight.

[0119] The economic efficiency score is determined by the following secondary indicators:

[0120] Initial investment cost is typically the sum of all expenses, expressed by the following formula:

[0121] (3);

[0122] Among them, C i This represents the specific initial investment cost for the i-th item, where n is the total number of initial investment items. This includes the cost of purchasing various electrical equipment, such as transformers, switch cabinets, and power cables; the material and construction costs for laying power transmission and distribution lines; the construction costs for infrastructure such as substations and switch stations, covering site leveling and building construction; and the costs of preliminary project planning, feasibility studies, surveying and design.

[0123] Annual average operation and maintenance cost reflects the average cost incurred each year by the power grid to maintain its safe, stable, and efficient operation during normal operation. It is expressed by the following formula:

[0124] (4);

[0125] Where O1, O2, ..., Ot represent the annual operation and maintenance costs within year T. These mainly include equipment maintenance costs (regular inspection and maintenance of transformers, lines, and other equipment, and replacement of aging or damaged parts), operation and management costs (salaries, bonuses, and benefits paid to operation and maintenance personnel, as well as costs for power grid operation monitoring and dispatch management), and energy consumption costs (electricity consumed to ensure the normal operation of power grid equipment).

[0126] Net present value (NPV) typically takes into account the time value of money and can comprehensively assess the economic value of a project. NPV is expressed by the following formula:

[0127] (5);

[0128] Where t represents the year, n is the project calculation period, CI is the cash inflow in year t, CO is the cash outflow in year t, and i is the discount rate.

[0129] The internal rate of return (IRR) is the discount rate that makes the net present value (NPV) of a project equal to zero. The IRR is expressed by the following formula:

[0130] (6);

[0131] Where n represents the number of the project calculation period, t represents a specific point in time within the calculation period, CI represents the financial income of the project at a specific point in time, CO represents the financial expenditure of the project at a specific point in time, such as construction costs and operation and maintenance costs, and IRR is the internal rate of return, which is used to measure the actual rate of return on investment that the project can achieve.

[0132] The payback period refers to the time required to recover the initial investment cost through the project's net cash flow. It is expressed by the following formula:

[0133] (When the net cash flow is equal each year) (7);

[0134] (The net cash flow is not equal each year) (8);

[0135] Where T represents the year in which the cumulative net cash flow first becomes positive.

[0136] The dynamic payback period is calculated by first discounting the net cash flows for each year using a line graph. The dynamic payback period is expressed by the following formula:

[0137] (9);

[0138] The social benefit indicator score is determined by the following secondary indicators:

[0139] Reducing user expenses refers to the amount of electricity saved by users, which is expressed by the following formula:

[0140] (10);

[0141] Where M(k) represents the amount of electricity saved by the user, Q(k) represents the reduction in electricity consumption by the user in year k, and c(k) represents the electricity price.

[0142] The carbon emission reduction benefits of power grids refer to the comprehensive benefits achieved by reducing greenhouse gas emissions such as carbon dioxide through a series of measures during the operation, construction, and management of the power grid. The carbon emission reduction benefits of power grids are expressed by the following formula:

[0143] (11);

[0144] Among them, C r ΔE represents the amount of carbon emission reduction (unit: tons of CO2), which is a quantitative manifestation of the carbon emission reduction benefits of the power grid; ΔE represents the difference in electricity generated by the replacement of high-carbon power sources with low-carbon or zero-carbon power sources due to measures such as power grid optimization and dispatch (unit: kilowatt-hours kWh).

[0145] Regional tax contribution rate is a key indicator for measuring the extent to which a power grid company contributes to the tax revenue of its region. The reduction in regional tax contribution rate is expressed by the following formula:

[0146] (12);

[0147] Among them, R tax Indicates the region's tax contribution rate (%); T paid This represents the total amount of various taxes actually paid by the power grid company in the region (unit: yuan), including value-added tax, corporate income tax, urban maintenance and construction tax, etc.; T total This represents the total tax revenue of the region during the same period (unit: yuan).

[0148] The score for the technical efficiency indicator is determined by the following secondary indicators:

[0149] Power supply reliability (RS-1) refers to the ratio of the total number of hours of effective power supply to users within a statistical period to the total number of hours in the statistical period, reflecting the power supply system's ability to continuously supply power to users. The power supply reliability is expressed by the following formula:

[0150] (13)

[0151] Where: n is the number of types of power outage events; N i T is the number of users experiencing power outages due to the i-th type of power outage event; i is the average outage time (in hours) for the i-th type of power outage event; N is the total number of users during the statistical period; 8760N is the number of hours in a year.

[0152] The comprehensive line loss rate refers to the percentage of power loss relative to the power supply in a power grid over a certain period. It is an important indicator for measuring the economic operation of a power grid and is expressed by the following formula:

[0153] (14)

[0154] Where: ΔP% is the overall line loss rate. Psupply is the amount of electricity supplied, that is, the amount of electricity supplied by the power grid to users, usually measured in kilowatt-hours (kWh); P sale It refers to the electricity sold, which is the amount of electricity that power companies sell to users, and the unit is also kilowatt-hour (kWh).

[0155] Average power outage time refers to the average number of hours of power outage for each user during the statistical period. The average power outage time is expressed by the following formula:

[0156] (15)

[0157] Where AIT is the average outage time (hours / household); n, N i T i The meanings of N and N are the same as those in the power supply reliability calculation formula.

[0158] The overall voltage qualification rate refers to the percentage of cumulative operating time within a specified voltage deviation range to the total operating time within a statistical period, reflecting the quality of the power supply voltage. The overall voltage qualification rate is expressed by the following formula:

[0159] (16)

[0160] Among them, V qual It represents the overall voltage qualification rate; m is the number of monitoring points; t i T is the cumulative time (in hours) during which the voltage at the i-th monitoring point is within the specified deviation range; T is the statistical time (in hours).

[0161] The capacity-to-load ratio is an important technical and economic indicator reflecting the power supply capacity of a power grid. It is expressed by the following formula:

[0162] (17)

[0163] Where K is the capacity-to-load ratio; S is the total capacity of the transformer (kVA); and P is the corresponding power supply load (kW).

[0164] Based on the above indicators, an investment calculation model that conforms to the personalized development trend of the power grid is constructed. Finally, the power grid investment calculation indicator system for this region is established as shown in Table 1 below;

[0165] Table 1. Indicator System for Power Grid Investment Calculation in this Region

[0166]

[0167] Based on the above power grid investment evaluation index system, the power grid enterprise investment benefit evaluation index system is simplified using correlation analysis. Assume there are investment benefit evaluation indices S1, S2, ..., S... m They are divided into k categories:

[0168] (18)

[0169] Each row represents a category of indicators, and a correlation analysis matrix is ​​performed on the indicators in each row. The closer the absolute value of the correlation coefficient between two indicator data is to 1, the stronger the correlation. The degree of correlation between two indicators is judged by the range of variation of the correlation coefficient r. The formula for calculating the correlation coefficient r is as follows:

[0170] (19)

[0171] in, , These are the averages of the two exponents, satisfying:

[0172] (20)

[0173] The correlation matrix is ​​as follows:

[0174] ;(twenty one)

[0175] Among them, two indicators that fail the correlation test can be used as independent indicators. For two indicators with high correlation, one is selected according to the above principle. The k types of indicators are reduced in this way. Finally, the remaining indicators are those obtained through correlation analysis. The determination of correlation is shown in Table 2 below:

[0176] Table 2. Correlation Determination Rules

[0177]

[0178] After conducting correlation analysis, a techno-economic analysis of power grid investment was performed:

[0179] Annual fee calculation formula:

[0180] ;(twenty two)

[0181] Where Ca is the total annual cost of medium-voltage lines distributed evenly over n years; CI is the investment in the construction of medium-voltage lines; CO is the annual operating cost of medium-voltage lines; CF is the annual power outage loss cost of medium-voltage lines; n is the economic service life of the line; and r0 is the discount rate.

[0182] High-voltage line investment HI:

[0183] The investment in high-voltage lines can generally be estimated using the following formula:

[0184] ;(twenty three)

[0185] Among them, L h For high-voltage lines (above 35kV); C 0h The investment per unit length of high-voltage lines varies significantly depending on the voltage level (as shown in Table 3 below). C f Investment in ring main units (load switches or ring main units) (ten thousand yuan / unit); num h This is the sum of the number of segments and connections of the line; the specific investment figures are shown in Table 3 below:

[0186] Table 3. Correspondence between High Voltage Line Investment and Voltage Level

[0187]

[0188] The investment for medium-voltage lines can generally be estimated using the following formula:

[0189] ;(twenty four)

[0190] Where L is the length of the medium-voltage line (1-35kV); C0 is the investment per unit length of the line, C f This represents the investment in ring main units (load switches or ring main units); num represents the total number of line segments and connections.

[0191] Annual operating costs include line loss costs and line inspection and maintenance costs, calculated using the following formula:

[0192] (25)

[0193] U1 is the cost of line inspection and maintenance; The comprehensive electricity price for 10kV power (yuan / kWh); the total annual power loss of the line (kWh); and the following:

[0194] (26)

[0195] Where, ΔP i τ represents the maximum load loss of the line; τ represents the number of hours (h) of maximum load loss.

[0196] There are two scenarios to consider when calculating power outage losses: one is to analyze only the company's costs, i.e., only considering direct losses; the other is to consider social benefits in addition to the company's costs, i.e., to consider both direct and indirect losses (i.e., the economic losses caused to the power supply company by the power outage). Direct losses are estimated by multiplying the electricity purchase and sale revenue by the outage duration. The formula for calculating the annual power outage loss CF is as follows:

[0197] (27)

[0198] Among them, LOSS1 is the direct economic loss to the power supply company caused by the power outage (ten thousand yuan); the average power outage duration of the AIHC system (hours); and P1 is the average load of a single line (kW). The difference between the purchase and sale price of electricity (RMB / kWh).

[0199] Indirect losses refer to the losses caused to society and users by power outages. Indirect economic losses can be calculated using the value per kilowatt-hour (kWh). The value per kWh is the ratio (yuan / kWh) of the Gross Domestic Product (GDP) to the electricity consumed in a specific region during a particular period (year). The formula for calculating indirect losses is as follows:

[0200] (28)

[0201] Where LOSS2 is the indirect economic loss caused by the power outage (ten thousand yuan); AIHC is the average power outage duration of the system (hours); P1 is the average load of a single line (kW); and k is the value of electricity generated per kilowatt-hour (yuan / kWh).

[0202] The power outage loss value calculated using the above formula is the power outage loss for a single 10kV outgoing line. To calculate the total power outage loss for the entire power supply area, simply multiply the power outage loss for a single outgoing line by the number of outgoing lines at the substation.

[0203] When calculating annual costs, the cost breakdown structure model does not need to cover all detailed costs throughout the life cycle. Based on the principle of comparison, costs that are the same or very similar before and after the modification are ignored. Therefore, decommissioning costs can be disregarded, and operating costs can be simplified to line loss costs and line maintenance and repair costs.

[0204] In step S4, the investment allocation scheme for the distribution network project is determined using the fuzzy comprehensive evaluation results and the fuzzy multi-objective optimization model. Based on the determined index weights, the investment benefits are evaluated using the fuzzy comprehensive evaluation method, resulting in five levels of evaluation results: "poor," "fairly poor," "medium," "good," and "excellent," presented as a percentage. Then, using the fuzzy multi-objective optimization method, the sub-objective functions are fuzzified, and a scaling coefficient is set to adjust the difference, transforming the multi-objective optimization problem into a single-objective optimization problem. This maximizes investor satisfaction under the constraints, thereby determining the allocation of investment in the distribution network project.

[0205] Although the fuzzy comprehensive evaluation method can comprehensively consider multiple benefits, it is difficult to accurately weigh the weights of each objective when dealing with complex multi-objective problems, and the construction of the fuzzy relation matrix is ​​subjective. The fuzzy multi-objective optimization method is easily affected by the initial parameters and data fluctuations in the process of determining the optimal solution. The proposed optimization strategy combines expert experience and data mining technology to optimize the construction of the fuzzy relation matrix. An adaptive parameter adjustment mechanism is introduced in the fuzzy multi-objective optimization to dynamically adjust parameters such as the scaling factor according to the optimization process. The main steps are as follows: (1) Determine the membership function for the benefit-type indicators. (2) Determine the indicator weights. According to the power grid development stage and actual demand changes, use the improved DEMATEL-AHP method or other weight determination methods to update the indicator weights to ensure that the investment allocation scheme always conforms to the actual situation. (3) In the process of fuzzy multi-objective optimization, monitor the optimization effect in real time. When the optimization process tends to be flat or falls into a local optimum, adjust the scaling factor adaptively. If the optimization of the operation efficiency index is difficult in a certain stage, appropriately reduce the corresponding scaling factor to enhance the optimization strength of the index.

[0206] (1) Determination of Membership Function: After determining the weights of the indicators, the fuzzy comprehensive evaluation method was applied to the investment benefit assessment, and the final evaluation result was obtained. First, the membership function of the benefit-type indicators was set. The upper limit of the evaluation indicator value is c, and the lower limit is d. Three standard reference points were selected in the interval [c,d], and the interval was evenly divided to solve for the membership function of the benefit-type indicators, as shown in the following formula:

[0207] (29)

[0208] In the formula, Q is the evaluation index value: .

[0209] (2) The Dematel-AHP method is used to determine the weights of the evaluation indicators. The Dematel method (DEMATEL) establishes an influence matrix based on the logical relationships between elements to deeply analyze the interactions and coupling relationships between elements within a complex system. Although the Dematel method excels in revealing causal relationships between elements, it is insufficient in accurately quantifying the weights of each element. Therefore, based on the influence matrix determined by the Dematel method, the Analytic Hierarchy Process (AHP) is further combined to accurately determine the weight values ​​of each element. This combined strategy not only reflects the closeness of the mutual influence between various indicators but also fully considers the weights. The specific steps for determining the weights based on the Dematel-AHP method will be detailed below.

[0210] When using the Dematel method for systematic rating and quantitative scoring of indicators, m experts and scholars are invited to participate in the evaluation of each indicator, conducting pairwise comparisons to accurately reveal the degree of mutual influence between the indicators. After determining the rating of the indicators, the relationships between the indicators should be clarified, thus forming an influence matrix as shown below:

[0211] (30)

[0212] The preliminary direct influence matrix A is normalized to obtain the normalized direct influence matrix S:

[0213] (31)

[0214] Based on S, further calculations and processing are performed to obtain the comprehensive influence matrix C:

[0215] (32)

[0216] In the formula, E is the identity matrix.

[0217] Based on the above steps, construct the overall influence matrix D:

[0218] (33)

[0219] By using the overall impact matrix D, indicators with weaker impacts are filtered out, thereby constructing a suitable comprehensive evaluation structure model for investment planning benefits.

[0220] First, a scale of 1 to 9 is used to compare and analyze each element, and a judgment matrix is ​​established based on this. Next, the eigenvalue method is used to obtain the weight vector. For elements that have no impact, their corresponding eigenvectors are set to zero during the calculation process, thereby further deducing the specific weight value G of each indicator.

[0221] Using the DEMATEL method, a mixed weight relationship matrix H of the evaluation indicators is constructed:

[0222] (34)

[0223] Solving for subjective mixed weights:

[0224] (35)

[0225] In fuzzy multi-objective optimization, the evaluation indicators for various weak links have different dimensions, and their weights are difficult to determine. Therefore, a fuzzy multi-objective optimization method is adopted, using trapezoidal curves to fuzzify the sub-objective functions.

[0226] (36)

[0227] In the formula, This represents the optimal value for a single objective indicator, such as operational efficiency. The difference between the optimal value for a single objective.

[0228] For positive indices such as space frame structures, the membership function is:

[0229] (37)

[0230] In the formula: This represents the optimal value of a single objective indicative of a positive indicator. The difference between the optimal values ​​for a single objective and the target. Investor satisfaction is represented by μ, and let μ be the minimum value among all values ​​after fuzzification of all sub-objectives, i.e.:

[0231] ;

[0232] Thus, the original multi-objective optimization problem, in which the sub-objectives had different dimensions and their weights were difficult to determine, is transformed into a single-objective nonlinear optimization problem. The solution is simply to maximize the satisfaction index μ while satisfying all constraints. The fuzzy multi-objective optimization model is then:

[0233] (38)

[0234] Based on the above, the steps for fuzzy model conversion are determined as follows:

[0235] Step 1: Input historical data of the distribution network and solve the distribution network problem. This is a single-objective optimization problem with the goal of maximizing the improvement effect in the development direction of a class. Taking the operational efficiency class as an example, we obtain the optimal expected value f of the improvement in operational efficiency. imax The corresponding improvement value f' of other classes n .

[0236] Step 2: Similarly, the optimal expected value f for all development directions in the single-objective optimization problem can be obtained. imax and the minimum expected improvement f imin .

[0237] Step 3: Based on the above steps, we set the scaling factor ε. i The difference between various development directions After adjustment, its value range becomes:

[0238] (39)

[0239] In the formula, , ε i The value of ε reflects the power grid company's subjective intention to invest. i The smaller the value, the stronger the power grid company's willingness to address this weakness, and the lower its tolerance for the indicator failing to reach its optimal value.

[0240] Fuzzy evaluation is a comprehensive assessment of the economic, social, and safety benefits of the evaluated object, and it usually follows the basic steps below for evaluation.

[0241] Step 1: Define the domain of discourse of the elements being evaluated as follows .

[0242] Step 2: Define the evaluation level domain as .

[0243] Step 3: When performing single-element evaluation, it is necessary to establish a fuzzy relation matrix R, and the membership value r of the i-th index of the j-th evaluation object in the membership function. i,j The calculations are performed to obtain the fuzzy relation matrix, as shown in the equation.

[0244] (40)

[0245] Step 4: Use the DEMATEL-AHP model to obtain the weight values ​​of each evaluation indicator. Then, based on the empirical entropy weight method, the entropy weight of each indicator is calculated separately. .

[0246] (41)

[0247] By calculating the proportionality coefficient and To obtain the final weight value, W is given in the formula. i,j This represents the i-th indicator of the j-th evaluation object.

[0248] Step 5: After obtaining the fuzzy relation matrix R and the index weight vector W, the fuzzy comprehensive evaluation result vector B of the j-th evaluation object is finally obtained through the synthesis operation of these two.

[0249] (42)

[0250] Where b=(j=1,2,……,m) is obtained by the operation of the j-th column of W and R, representing the degree of membership of the rated object to the fuzzy subset as a whole.

[0251] Assume that the weights of each evaluation indicator in the investment efficiency evaluation system are already determined as W. i Based on this, a fuzzy assessment of the overall benefit level of the evaluation object j is made.

[0252] (43)

[0253] In the formula, W i (k) represents the weight value of indicator i. i (k) is the evaluation value of indicator i.

[0254] Step 6: Divide the results into 5 levels: "Poor," "Fairly Poor," "Average," "Good," and "Excellent." For ease of understanding and comparison, the fuzzy evaluation results are expressed on a percentage basis, that is, by calculating the weighted vector of the evaluation results. The fuzzy comprehensive evaluation results are analyzed as shown in the formula.

[0255] (44)

[0256] In the formula, y1≤35 is the poor grade; 35<y2≤50 is the poor grade; 50<y3≤65 is the average grade; 65<y4≤80 is the good grade; and 80<y5≤100 is the excellent grade.

[0257] The comprehensive evaluation value is obtained through calculation:

[0258] Z=YB; (45)

[0259] Those skilled in the art will understand that all or part of the above-described methods for power grid investment planning and analysis can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. This computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0260] This invention, based on the grid division of the power grid's latitude and longitude coordinates, analyzes the influencing factors of distribution network investment strategies, assesses the power grid's carrying capacity for new energy access and the impact of new energy access on the power grid, and analyzes factors such as investment costs and benefits. It constructs an evaluation index system for distribution network input and output, selecting input indicators reflecting investment scale, structure, and efficiency, as well as output indicators reflecting power supply capacity, security, and reliability. Using methods such as the Analytic Hierarchy Process (AHP) and the DemATEL (Decision Experiment and Evaluation Laboratory) method, it evaluates the effectiveness of distribution network input and output and conducts sensitivity analysis, identifies key input factors, and proposes improvement suggestions. Based on the power grid development stage and load growth trend, it constructs a fuzzy objective function model and establishes a dynamic adjustment mechanism for investment projects. Through fuzzy comprehensive evaluation, it determines the index weights, conducts multi-element fuzzy evaluation, analyzes the evaluation results, and transforms the multi-objective optimization problem of distribution network investment allocation into a single-objective optimization problem to determine the allocation of distribution network project investment to maximize satisfaction.

[0261] This invention, through in-depth analysis of various data characteristics of the power grid, including longitude, latitude, node geographical distribution, load distribution, and grid structure, uses the K-means clustering algorithm combined with the elbow algorithm to determine the optimal number of clusters, dividing the power grid into several grids. This division method comprehensively considers data characteristics, distribution patterns, and practical application needs. Compared to traditional methods of arbitrary or single-criteria division, it is more scientific and reasonable, providing a more effective data foundation and division basis for subsequent power grid investment benefit assessment, operation and maintenance, regional management, and performance index analysis. This invention constructs an input-output benefit evaluation index system and combines methods such as the Analytic Hierarchy Process (AHP) and the DemATEL (Decision Experiment and Evaluation Laboratory) for effectiveness evaluation and sensitivity analysis. This process deeply explores the correlation characteristics between different investment projects, achieving a comprehensive consideration of the mutual exclusion and synergistic relationships between projects, making up for the shortcomings of traditional planning models in this regard, and enabling investment decisions to better balance the relationships between various projects. By constructing a multi-objective optimization model for distribution network investment and using the fuzzy comprehensive evaluation method to transform the multi-objective optimization problem into a single-objective optimization problem, the investment allocation is determined based on maximizing investor satisfaction. At the same time, various factors are fully considered in the model construction and solution process, such as adjusting the difference between different development directions by setting a scaling factor to meet the different investment intentions of the power grid company, thereby improving the scientificity and rationality of investment decisions and maximizing the benefits of distribution network investment.

[0262] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention. For example, in practical applications, the number of clusters in K-means clustering can be adjusted according to the data characteristics and actual needs of power grids in different regions, or the investment benefit evaluation index system can be optimized and supplemented. In multi-objective optimization models, improvements can be made to the fuzzification processing method, the scaling factor setting, etc. These are all reasonable changes within the scope of protection of this study.

Claims

1. A power distribution network investment strategy optimization method based on development needs and input-output evaluation, characterized in that: The optimization steps include the following: Step 1: Construct an investment optimization evaluation model of the power distribution network to evaluate the investment optimization capability; Step 2: Construct a decision test and evaluation laboratory-AHP model to evaluate the investment optimization of the power grid planning, determine the index weight, calculate the weight based on the actual data of regional input and output, normalize the index weight, and finally determine the evaluation grade; Step 3: When evaluating the investment optimization decision of the power grid planning, based on the K-means clustering grid division of the longitude and latitude of a single project, an input-output benefit evaluation index system is constructed with economic benefit, social benefit and technical benefit as the first-level indicators, and a multi-element membership degree model is constructed based on the correlation between investment and indicators; Step 4: The weight of the input-output indicators of the power grid investment optimization is solved by the comprehensive evaluation method of the decision test and evaluation laboratory-AHP, and the evaluation grade of the power grid investment planning is determined according to the simulation results. 2.The power distribution network investment strategy optimization method based on development demand and input-output evaluation of claim 1, wherein: The multi-element membership degree model constructed in step 3 specifically includes: constructing a correlation matrix of power distribution network investment evaluation indicators, a power distribution network investment multi-objective optimization model, and a power distribution network investment multi-objective fuzzy function.

3. The power distribution network investment strategy optimization method based on development demand and input-output evaluation according to claim 2, characterized in that: The specific method of grid division of the longitude and latitude of a single project based on K-means clustering in step 3 is as follows: In the application scenario of power grid investment planning, the K-Means clustering iteration is performed on the longitude and latitude data set of the power grid, each data object point is assigned to the cluster closest to it, and then the cluster center of each cluster is recalculated according to the coordinate mean of the members in the cluster, and the iteration is continued until the cluster center no longer changes or the preset convergence range is reached or the predetermined number of iterations is reached; The elbow algorithm is used to verify the optimal cluster number: The intra-cluster error sum of squares under 1 to 10 different cluster numbers is calculated, the data points are clustered by the K-means algorithm, the intra-cluster error sum of squares is calculated according to the cluster index and cluster center of the data points, and the elbow rule graph is drawn after the calculation is completed, with the horizontal coordinate representing the cluster number and the vertical coordinate representing the intra-cluster error sum of squares; The optimal cluster number is determined by calculating the descending slope of the intra-cluster error sum of squares of adjacent points, and when the slope is less than half of the average slope, the index of the point is added by 1 as the cluster number corresponding to the elbow, that is, the optimal cluster number. The specific method of constructing the input-output index system of the power distribution network in step 4 to determine the evaluation grade of the power grid investment planning is as follows:

4. The power distribution network investment strategy optimization method based on development demand and input-output evaluation according to claim 3, characterized in that: An investment calculation model is constructed according to the actual situation and the differences in power grid characteristics, which conforms to the individualized development trend of the power grid, the first-level indicators of the investment calculation model include economic benefit, social benefit and technical benefit, and the economic benefit indicators include equipment purchase, site construction and preliminary planning cost; The social benefit indicators include the ability of users to obtain multiple comprehensive energy and meet different needs; The technical benefit indicators include power supply reliability rate, comprehensive line loss rate, average power outage time, comprehensive voltage qualification rate and load ratio. The second-level indicators of the economic benefit in the first-level indicators of the investment calculation model include: The initial investment cost is calculated by the formula:

5. The method for power distribution network investment strategy optimization based on development demand and input-output evaluation according to claim 4, characterized in that: ​ ​ Initial investment cost = cost of electrical equipment + cost of materials and construction + cost of building engineering + cost of project planning + cost of feasibility study + cost of survey and design; Annual average operation and maintenance cost, calculation formula is: Annual average operation and maintenance cost = average operation and maintenance cost in recent T years; Net present value, calculation formula is: Net present value = annual cash inflow - cash outflow; According to the discount rate, add up the discounted results of each year, and finally subtract the total cost of the project; Internal rate of return, calculation formula is: Internal rate of return = the difference between cash inflow and cash outflow from the 0th period to the nth period, divided by (1 + the number of periods of internal rate of return) times, then add up the calculation results of each period, and let the total result equal to 0; Static investment recovery period, calculation formula is: Static investment recovery period = initial investment cost / annual net cash flow to get the static investment recovery period (when the annual net cash flow is equal); The number of years when the cumulative net cash flow first appears positive minus 1 + the investment value of the previous year that has not been recovered divided by the net cash flow of that year; Dynamic investment recovery period, calculation formula is: Dynamic investment recovery period = [(the number of years when the cumulative net cash flow first appears positive - 1) + the investment value of the previous year that has not been recovered] / the present value of the net cash flow of that year; Among them, each year's net cash flow needs to be calculated according to the discount rate.

6. The power distribution network investment strategy optimization method based on development demand and input-output evaluation according to claim 5, characterized in that: The second-level indicators of social benefits in the construction of the investment estimation model of the first-level indicators include: Reducing user expenditure cost, calculation formula is: Reducing user expenditure cost = (user reduced power consumption from the 1st year to the Nth year) * electricity price / 1 + (the number of years of discount rate times); Power grid carbon emission reduction benefit, calculation formula is: Power grid carbon emission reduction benefit (i.e. the quantitative embodiment of power grid carbon emission reduction benefit) = the difference between the power generated by low-carbon or zero-carbon power sources and high-carbon power sources due to power grid optimization scheduling and other measures * emission factor; Regional tax contribution rate, calculation formula is: Regional tax contribution rate = (the total amount of taxes actually paid by the power grid enterprise in the region / the total tax revenue in the region during the same period) * 100%.

7. The method for power distribution network investment strategy optimization based on development demand and input-output evaluation according to claim 6, characterized in that: The second-level indicators of technical benefits in the construction of the investment estimation model of the first-level indicators include: Power supply reliability rate, calculation formula is: Power supply reliability rate = 1 minus the sum of the product of each type of outage user number and the average outage time of the type of outage event divided by the total number of hours in a year, and then multiplied by 100%; Comprehensive line loss rate, calculation formula is: Comprehensive line loss rate = (power supply quantity - power supply quantity) / power supply quantity; Average outage time, calculation formula is: Average outage time = (the product of each type of outage user number and the average outage time of the type of outage event) / total user number.

8. The power distribution network investment strategy optimization method based on development demand and input-output evaluation according to claim 7, characterized in that: The specific method for constructing the correlation matrix of the distribution network investment evaluation index is: The investment benefit evaluation index system of power grid enterprises is reduced by using correlation analysis method, and the investment benefit evaluation index is defined is divided into category, and the expression is: ； Each of the above lines represents a class of indicators. Correlation matrix analysis is performed on the indicators in each line. The closer the absolute value of the correlation coefficient of the data of two indicators is to 1, the stronger the correlation is. The correlation coefficient is used to determine the correlation degree between two indicators by the range of the correlation coefficient The expression is: ； wherein , are the average of two exponential values, respectively, satisfying: 。 9. The power distribution network investment strategy optimization method based on development demand and input-output evaluation according to claim 8, characterized in that: The specific method for constructing the distribution network investment multi-objective optimization model is: Using the decision-making trial and evaluation laboratory comprehensive evaluation method to determine the weight of each evaluation index, including the following steps: The element domain of the object to be evaluated is defined as ; The assessment rating domain is defined as ; The fuzzy relation matrix R is established to perform single-element evaluation, and the membership degree value of the jth evaluation object of the ith index of the membership degree model is calculated The fuzzy relation matrix is calculated, and the expression is as follows:​ ； The DEMATEL-AHP model is used to obtain weight values of various evaluation indexes , and the entropy weight of each index is calculated based on the empirical entropy weight method , and the calculation formula is ； By calculating the proportional coefficient and , the final weight value is obtained, in which represents the first index of the first evaluation object; and represents the first index of the first evaluation object. ​ The fuzzy relation matrix is calculated and the index weight vector , through the synthetic operation of the two, the final fuzzy comprehensive evaluation result vector of the first evaluation object is obtained , and the calculation formula is: ； wherein is obtained by and the first column operation, representing the membership degree of the rated object to the fuzzy subset as a whole; Assume that the weight of each evaluation index in the investment benefit evaluation system is determined as , and based on this, the overall benefit level of the evaluation object is judged, and the judgment expression is: ； wherein, is a weight value of the index , is an evaluation value of the index .

10. The power distribution network investment strategy optimization method based on development demand and input-output evaluation according to claim 9, characterized in that: The specific method for constructing the distribution network investment multi-objective fuzzy function is: Using the fuzzy multi-objective optimization method, using the trapezoidal curve to fuzzy the sub-objective functions, the expression is: ； In the formula: represents the optimal value of single-objective optimization of negative indicators such as operating efficiency; is the difference of the optimal value of single-objective optimization. For the positive index of the grid structure, the membership function expression is: ； In the formula: denotes the optimal value of the single-objective optimization of the positive indicator, is the difference of the optimal values of the single-objective optimization. Investor satisfaction is used To indicate, to make It is the minimum value among all values ​​after blurring all sub-targets, that is, it satisfies: ； The multi-objective optimization problem is changed into a single-objective nonlinear optimization problem, and only the satisfaction index needs to be maximized under the condition of meeting all the constraint conditions The expression of the fuzzy multi-objective optimization model is changed into: ； Then the fuzzy model conversion is performed, including the following steps: Input the distribution network historical data, solve the distribution network with the first Single-objective optimization problem with the optimal improvement effect of the development direction as the target, and obtain the optimal expected value of all development directions in the single-objective optimization problem And the minimum improvement expectation ; Setting the telescopic coefficient The difference between the various development directions Adjustment is made, and the value range becomes: ； wherein, , The value of the index represents the subjective intention of the power grid company to invest, The smaller the value, the stronger the intention of the power grid company to develop this type of weak link, and the lower the tolerance for not achieving the optimal value of this type of index.

Images