New energy delivery system inertia evaluation method based on comprehensive benefit evaluation model

Abstract

The invention discloses a new energy delivery system inertia evaluation method based on a comprehensive benefit evaluation model, and belongs to the field of power grids, and the method comprises the steps: S1, constructing a system inertia comprehensive evaluation index system; s2, constructing a system inertia comprehensive benefit model; S21, constructing a subjective and objective index integrated weighting model, and performing evaluation index weight calculation; s22, using an ideal point method to improve the matter-element extension model, establishing an ideal matter-element extension evaluation model, and calculating a comprehensive close degree; and S3, evaluating the system inertia of the to-be-evaluated matter element according to the comprehensive close degree. According to the new energy delivery system inertia evaluation method based on the comprehensive benefit evaluation model, a system inertia comprehensive index evaluation system is established from multiple dimensions, the subjective and objective index integration weighting model is selected to carry out index weight calculation, and the ideal point method is used to improve the traditional matter element extension model, so that the system inertia evaluation accuracy is improved. And an accurate evaluation result can be obtained.

Description

Technical Field

[0001] The present invention relates to the field of power grid technology, and in particular to an inertia evaluation method for a new energy transmission system based on a comprehensive benefit evaluation model. Background Art

[0002] The current grid dispatching and operation process lacks a basis for accurately analyzing the degree of system inertia sufficiency. On the one hand, if the system inertia is maintained at a level far higher than the safety critical value, it may aggravate the phenomenon of wind and solar power abandonment; on the other hand, insufficient system inertia will bring the risk of frequency instability. In addition, the unclear inertia safety range also restricts the carrying capacity and frequency stability of power systems with a high proportion of converters. Therefore, a scientific system inertia assessment method is needed to enhance the real-time perception of the system inertia level by grid dispatchers, so that dispatchers no longer rely solely on experience to judge whether the system inertia and reserve need to be adjusted when clarifying the system inertia level.

[0003] At present, the inertia assessment methods of power systems can be summarized as assessment based on large disturbance events, assessment based on small disturbance events, and assessment based on quasi-steady-state operation, which are described in detail as follows:

[0004] 1. System inertia assessment method based on large disturbance events

[0005] This method is to estimate the inertia of the system by monitoring and analyzing the dynamic response of the power system when a large disturbance occurs. Usually, the change of system frequency is monitored, and combined with theoretical models such as rotor dynamics equations, the inertia characteristics of the system are inferred by analyzing the rate and amplitude of frequency change. This method requires dynamic modeling of the system and high-precision frequency measurement data. By analyzing the dynamic response of the system under large disturbance events, the inertia characteristics of the system can be inferred, thus providing an important reference for the frequency stability assessment of the power system.

[0006] At present, some related studies have performed polynomial fitting on the transient frequency of the system under large disturbance events, and then obtained a rough grid inertia based on the generator rotor motion equation; some studies have also proposed a system inertia estimation method based on the large disturbance rotor motion equation to make a difference between two regions.

[0007] This method has the following drawbacks:

[0008] 1) Difficulty in data acquisition: The frequency of large disturbance events is usually low, so it may be difficult to obtain sufficient large disturbance event data for evaluating system inertia;

[0009] 2) High experimental costs: Experiments on large disturbance events usually require a lot of time and resources, including equipment, manpower, and funds, so the cost is relatively high;

[0010] 3) High risk: Large disturbance events may cause serious damage to the system, so there is a certain risk in conducting related experiments and tests;

[0011] 4) Unable to fully evaluate the system: The inertia evaluation of the system needs to consider the response of the system under different working conditions, while large disturbance events can only provide data under specific conditions and cannot fully evaluate the inertia characteristics of the system.

[0012] 2. System inertia assessment method based on small disturbance events

[0013] This method estimates the inertia of the system by analyzing the dynamic response of the power system under small disturbances. The small disturbance usually refers to a slight change in the system frequency, such as frequency changes caused by load fluctuations or the connection of small power generators. The principle of this method is mainly to infer the inertia characteristics of the system by monitoring and analyzing the frequency response of the system under small disturbance events. And this method usually requires the use of the system's dynamic model and frequency response measurement data, and by analyzing these data, the inertia characteristics of the system can be inferred.

[0014] At present, some studies have proposed a power system inertia estimation method including wind power virtual inertia response to quantitatively analyze the impact of virtual inertia on the equivalent inertia of the power grid.

[0015] This method has the following drawbacks:

[0016] 1) Limited impact on system stability: small disturbance events may not fully reflect the system's response to large disturbances, so the assessment of system stability is not comprehensive enough;

[0017] 2) Ignoring the nonlinear characteristics of the system: small disturbance events can often only reflect the linear response of the system and fail to fully consider the nonlinear characteristics of the system, resulting in inaccurate evaluation results;

[0018] 3) Limitations on the dynamic characteristics of the system: Small disturbance events cannot fully reflect the dynamic characteristics of the system, especially for rapidly changing systems, the evaluation results may be biased;

[0019] 4) Sensitivity to changes in system parameters: Small disturbance events are highly sensitive to changes in system parameters. Once system parameters change, the evaluation results may no longer be applicable.

[0020] 5) Insufficient evaluation of the actual operation of the system: The evaluation method of small disturbance events often fails to consider the actual operation of the system and lacks an evaluation of the overall performance of the system.

[0021] 3. System inertia evaluation method based on quasi-steady-state operation

[0022] The inertia assessment based on quasi-steady-state operation is mainly based on a large amount of historical data and the operation status of the power grid. The inertia trend prediction of the power grid is realized through artificial intelligence, probability prediction and correlation analysis. The principle of this method is to evaluate the inertia of the system by analyzing the dynamic response of the system under quasi-steady-state conditions. Under quasi-steady-state operation conditions, the system parameters such as frequency and voltage are in a slowly changing state. At this time, the dynamic response of the system is affected by inertia. Therefore, by analyzing the dynamic response of the system under quasi-steady-state operation conditions, the inertia characteristics of the system can be inferred.

[0023] This method has the following drawbacks:

[0024] 1) Ignoring the nonlinear characteristics of the system: The quasi-steady-state operation method usually assumes that the system is in a near-steady-state state and cannot fully consider the nonlinear characteristics of the system, resulting in inaccurate evaluation results;

[0025] 2) Sensitivity to changes in system parameters: The quasi-steady-state operation method is highly sensitive to changes in system parameters. Once the system parameters change, the evaluation results may no longer be applicable;

[0026] 3) Limitations on the dynamic characteristics of the system: The quasi-steady-state operation method cannot fully reflect the dynamic characteristics of the system, especially for rapidly changing systems, the evaluation results may be biased;

[0027] 4) Insufficient understanding of the actual operating conditions of the system: The quasi-steady-state operation method often fails to take into account the actual operating conditions of the system and lacks an evaluation of the overall performance of the system. Summary of the invention

[0028] In order to solve the above problems, the present invention provides an inertia evaluation method for a new energy delivery system based on a comprehensive benefit evaluation model, and establishes a system inertia comprehensive index evaluation system from multiple dimensions, so that when the dispatcher clarifies the system inertia level, he no longer relies solely on experience to judge whether the system inertia and standby need to be adjusted. The subjective and objective index integrated weighting model that takes into account the subjective and objective weighting method is selected to calculate the index weight, and the ideal point method is used to improve the traditional matter-element extension model, and an ideal matter-element extension evaluation model is established to calculate the comprehensive benefit of the system inertia, so that a relatively accurate evaluation result can be obtained.

[0029] To achieve the above object, the present invention provides a new energy delivery system inertia evaluation method based on a comprehensive benefit evaluation model, comprising the following steps:

[0030] S1. Construct a comprehensive evaluation index system for system inertia;

[0031] S2. Constructing the comprehensive benefit model of system inertia:

[0032] S21. Construct an integrated weighting model for subjective and objective indicators and calculate the weights of evaluation indicators;

[0033] S22. Use the ideal point method to improve the matter-element extension model, establish an ideal matter-element extension evaluation model, and calculate the comprehensive closeness;

[0034] S3. Evaluate the system inertia of the object-element to be evaluated based on the comprehensive closeness.

[0035] Preferably, the system inertia comprehensive evaluation index system described in step S1 includes three evaluation angles: safety benefit, economic benefit, and social benefit;

[0036] Among them, the evaluation of safety benefits is classified into system frequency safety benefits, which include the following evaluation indicators: non-synchronous power penetration rate, ratio of system inertia to maximum power loss during N-1 accident, and safe operation inertia;

[0037] Economic benefits include the following evaluation categories: power generation side benefits, grid side benefits and user side benefits. The classification indicators of power generation side benefits, grid side benefits and user side benefits are investment cost, operation and maintenance cost and economic present value respectively.

[0038] The evaluation category of social benefits is improving power supply reliability, and the evaluation indicator for improving power supply reliability is load stability.

[0039] Preferably, step S1 specifically includes the following steps:

[0040] S11. Use correlation test methods to analyze the overlap and inspection status between evaluation indicators and establish an indicator system with independent characteristics;

[0041] S12. Use normalization methods to make the evaluation indicators consistent and dimensionless.

[0042] Preferably, in step S11, correlation analysis is performed respectively according to the categories of the evaluation indicators, which specifically includes the following steps:

[0043] S111, Assume that there are n demand response implementation benefit evaluation indicators (I1, I2, ..., I n ), according to the classification when selecting the evaluation indicators, each type of evaluation indicators contains k indicators in total, and the correlation test is performed on each type of evaluation indicators according to the least squares method to construct the correlation test matrix R:

[0044]

[0045] In the formula, r αβ (α, β = 1, 2, ..., k) is the correlation coefficient between the αth indicator and the βth indicator, and the specific calculation is as follows:

[0046]

[0047] In the formula, I α are all observed values ​​of indicator α; is the average of all observed values ​​of indicator α; I β are all observed values ​​of the indicator β; is the average of all observed values ​​of indicator β;

[0048] And when α=β, r αβ =1, then r αβ =r βα ;

[0049] S112. Determine whether the relevance requirement is met according to the set relevance judgment principle.

[0050] Preferably, the evaluation index described in step S12 includes an extremely large index, an extremely small index and an intermediate index;

[0051] The method of using step S12 to respectively perform consistency processing on the extremely large index, the extremely small index and the intermediate index is as follows:

[0052] The consistent processing formula for extremely large indicators is as follows:

[0053]

[0054] In the formula, a and b are the lower and upper thresholds of the indicator respectively;

[0055] The consistent processing formula for very small indicators is as follows:

[0056]

[0057] The consistent processing formula for intermediate indicators is as follows:

[0058]

[0059] Preferably, step S21 specifically includes the following steps:

[0060] S211. Constructing the subjective empowerment model of ANP method:

[0061] S2111. Systematically analyze and combine decision problems to form elements and element sets to determine whether the elements are independent and whether there is a dependent feedback relationship;

[0062] S2112. Construct ANP structure: determine the control layer, and analyze the influence and influence relationship of all system elements according to the control criterion layer, and determine the criterion, elements and the relationship between elements;

[0063] S2113, construct ANP super matrix: establish probability as the main criterion, take one of the risk factors as the secondary criterion, compare the relative importance of other risk factors, and then calculate the characteristic vector of each risk category group as a unit, that is, the corresponding local weight vector, and establish the super matrix according to the formula after comparison, judgment and calculation with each element as the secondary criterion:

[0064]

[0065] Where W ij is a supermatrix, which is represented in the ANP model by the element set R i For the element set R j The impact of each The element is composed of R i The elements l and R in j After comparing and judging the element k in , the component of a eigenvector in the judgment matrix is ​​obtained;

[0066] Combine the eigenvectors of the interactions of all network layer elements to obtain a supermatrix under the control elements:

[0067]

[0068] Where n 1,2,...,N represents the normalization factor; W N1,N2,...,NN is the influence weight of the Nth element or criterion in the Nth row of the matrix on all other elements or criteria;

[0069] Normalize the supermatrix, that is, weight the elements of the supermatrix to obtain the weighted supermatrix a ij is the weighting factor, i,j=1,2,…,n;

[0070] S2114. Calculate the limit supermatrix:

[0071] Compute the limiting relative ranking vector for each supermatrix:

[0072]

[0073] Formula (8) expresses the limit of the average value of the z-th power of the supermatrix W;

[0074] If the above limit converges and is unique, the value of the corresponding row of the original supermatrix is ​​judged to be the stable weight of each evaluation index;

[0075] S212. Construct an objective weighting model using entropy weight method:

[0076] Assuming that the system inertia participates in the comprehensive benefits of the new energy delivery system, different integration methods t={t1,t2,…,tn}, the comprehensive evaluation index system of each integration method is u={u1,u2,…,u n}, Integration method i (i=1,2,…,m) in index u j The attribute value under (j=1,2,…,n) is t ij , then the decision matrix is ​​A=(t ij ) m×n , let M = (1, 2, ..., m), N = (1, 2, ..., n);

[0077] The indicator types include benefit type and cost type, and the benefit type and cost type are dimensionless respectively:

[0078] For benefit-type attributes, let:

[0079]

[0080] For cost attributes, let:

[0081]

[0082] In the formula, r ij is the standardized value, representing the elements in each attribute evaluation index; t ij is the attribute value, which represents the original value in the evaluation matrix or decision matrix, representing the actual measurement value or score of the jth evaluation indicator in the i-th option or situation; is the maximum value among all options of the jth evaluation index; is the minimum value of all options for the jth evaluation index;

[0083] Get the normalized decision matrix R = (r ij ) m×n ;

[0084] Calculate the entropy E of attribute index j j :

[0085]

[0086] Where: j = 1, 2, ..., m, k = 1 / ln (n) is a constant related to the number of samples, the purpose is to make E j ∈[0,1]; r j Satisfying 0<r ij <1 and And when r ij = 0, r ij ln(r ij )=0;

[0087] Calculate the information deviation:

[0088] d j =1-E j (12)

[0089] Calculate indicator weights:

[0090]

[0091] S213. Construct a subjective and objective integrated weighting model, and determine the weights of evaluation indicators with the subjective and objective integrated weighting model that minimizes the deviation between subjective and objective weights:

[0092] S2131. Assume that the subjective weight vector of the operation and maintenance cost evaluation index obtained by the subjective weighting model of the ANP method is w'=(w'1,w'2,…,w' n ) T , and satisfies w' j ∈[0,1],∑ j＝1 w' j =1, and the objective weight vector of operation and maintenance cost evaluation index is calculated by using the entropy weight method w”=(w”1,w”2,…,w” n ) T , and satisfies w" j ∈[0,1],∑ j＝1 w" j =1; add the subjective weight and the objective weight to get the final weight vector:

[0093] w=αw'+βw" (14)

[0094] Among them, α, β satisfy α, β > 0, and α + β = 1;

[0095] S2132. Considering that the weighted attribute value determined by the subjective weight tends to be consistent with the weighted attribute value determined by the objective weight, a distance function is constructed to represent the weighted value of the subjective and objective weights:

[0096]

[0097] Where s(w',w") represents the similarity or difference between the subjective weight vector w' obtained by the ANP method and the objective weight vector w" obtained by the entropy weight method.

[0098] Preferably, the ideal matter-element extension evaluation model described in step S22 is based on the matter-element extension model, by setting positive and negative ideal matter-elements to replace the classical domain, and using proximity to replace the maximum membership criterion to determine the distance between the matter-element to be evaluated and the positive and negative ideal matter-elements.

[0099] Preferably, the positive and negative ideal matter elements are determined based on expert opinions;

[0100] Alternatively, select the best and worst data from each evaluation index in the matter-element to be evaluated as the positive ideal matter-element and negative ideal matter-element respectively. If the evaluation index c i is a positive indicator, then For the index c i The optimal data; if the evaluation index c i is a reverse indicator, then If the evaluation index c i is a moderate indicator, then If the indicator c i is an interval indicator, then Take interval value represents the optimal interval; thus we obtain the positive ideal element R + and negative ideal element R - .

[0101] Preferably, asymmetric closeness is used as a closeness function, and the closeness function value is calculated and sorted to obtain:

[0102]

[0103] Where N is the closeness; D is the distance, which represents the difference measure of a single evaluation index; w i is the weight of index i, then the closeness of the object-element to be evaluated to each level is:

[0104]

[0105] In the formula, w i (X) represents the weight function of evaluation index i, which depends on the decision variable set X; D j (v' i ) represents the closeness between the matter-element R0 to be evaluated and the normalized ideal matter-element classical domain, which includes the positive closeness and negative proximity The calculation formula is as follows:

[0106]

[0107]

[0108] In the formula, v' i is the actual value under index i;

[0109] The closeness of the matter-element to be evaluated to the positive ideal matter-element and the negative ideal matter-element are:

[0110]

[0111]

[0112] The comprehensive closeness is:

[0113]

[0114] The present invention has the following beneficial effects:

[0115] 1. Evaluate the inertia of the new energy delivery system from multiple dimensions, so that the dispatcher no longer relies solely on experience to determine whether the system inertia and reserve need to be adjusted when determining the system inertia level;

[0116] 2. Considering the different grid connection conditions in different regions, the priority of the system inertia indicators concerned is also different.

[0117] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0118] Figure 1 The present invention is a flow chart of a new energy delivery system inertia evaluation method based on a comprehensive benefit evaluation model. DETAILED DESCRIPTION

[0119] In order to make the purpose, technical scheme and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention are further described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not intended to limit the embodiments of the present invention. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions.

[0120] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products or devices.

[0121] Like reference numerals and letters denote similar items in the following drawings, and thus, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0122] A new energy delivery system inertia evaluation method based on a comprehensive benefit evaluation model comprises the following steps:

[0123] S1. Construct a comprehensive evaluation index system for system inertia;

[0124] As the scale of clean energy power generation connected to the grid gradually increases, the phenomenon of a high proportion of new energy power generation connected to the grid leading to a significant reduction in system inertia is becoming more and more frequent. Low inertia systems are prone to voltage and frequency stability problems, among which frequency problems are particularly prominent, and also indirectly lead to problems such as excessively high investment costs for clean energy power generation and environmental pollution, thereby affecting the economic benefits of the operation of new power systems. Therefore, in order to comprehensively evaluate the benefits of system inertia, it is necessary to establish a multi-dimensional evaluation index system.

[0125] The system inertia comprehensive evaluation index system described in step S1 includes three evaluation angles: safety benefit, economic benefit, and social benefit;

[0126] Among them, the evaluation of safety benefits is classified into system frequency safety benefits, which include the following evaluation indicators: non-synchronous power penetration rate, ratio of system inertia to maximum power loss during N-1 accident, and safe operation inertia;

[0127] Economic benefits include the following evaluation categories: power generation side benefits, grid side benefits and user side benefits. The classification indicators of power generation side benefits, grid side benefits and user side benefits are investment cost, operation and maintenance cost and economic present value respectively.

[0128] The evaluation category of social benefits is improving power supply reliability, and the evaluation indicator for improving power supply reliability is load stability.

[0129] That is, the above-mentioned safety benefits, economic benefits, and social benefits are first-level indicators, system frequency safety benefits, power generation side benefits, grid side benefits, user side benefits and improved power supply reliability are second-level indicators; asynchronous power supply penetration rate, system inertia and the ratio of maximum power generation missing power during N-1 accidents, safe operation inertia, investment cost, operation and maintenance cost, economic present value and load stability are third-level indicators.

[0130] Table 1 System inertia comprehensive benefit evaluation index system

[0131]

[0132] After the preliminary establishment of the system inertia comprehensive benefit evaluation index system, the evaluation index system needs to be tested for correlation and standardized. The correlation test is mainly used to analyze the overlap and inspection between indicators to establish an index system with independent characteristics. Standardization is mainly used to make the index system consistent and dimensionless, so as to do a good job in the preliminary work of carrying out the system inertia comprehensive benefit evaluation.

[0133] Step S1 specifically includes the following steps:

[0134] S11. Use correlation test methods to analyze the overlap and inspection status between evaluation indicators and establish an indicator system with independent characteristics;

[0135] In step S11, correlation analysis is performed according to the types of evaluation indicators, which specifically includes the following steps:

[0136] S111. Considering that when selecting indicators, the scope of each type of indicators inevitably overlaps and crosses, there will be a high degree of correlation between the indicators. Therefore, correlation analysis can be performed separately according to the indicator category. Assume that there are n demand response implementation benefit evaluation indicators (I1, I2, …, I n ), according to the classification when selecting the evaluation indicators, each type of evaluation indicators contains k indicators in total, and the correlation test is performed on each type of evaluation indicators according to the least squares method to construct the correlation test matrix R:

[0137]

[0138] In the formula, r αβ (α, β = 1, 2, ..., k) is the correlation coefficient between the αth indicator and the βth indicator, and the specific calculation is as follows:

[0139]

[0140] In the formula, I α are all observed values ​​of indicator α; is the average of all observed values ​​of indicator α; I β are all observed values ​​of the indicator β; is the average of all observed values ​​of indicator β;

[0141] And when α=β, r αβ =1, then r αβ =r βα ;

[0142] S112. Determine whether the relevance requirement is met according to the set relevance judgment principle.

[0143] The correlation judgment principle set in this embodiment meets the content of Table 2. It can be seen that if the correlation coefficient of two indicators is less than 0.3, the two indicator data are considered to be independent of each other; if the correlation coefficient of two indicator data is greater than 0.3, and one indicator is also highly correlated with the other indicator, then the indicator is removed. The k indicators are simplified in turn, and finally the indicators that meet the correlation requirements are obtained.

[0144] Table 2 Principles of correlation judgment

[0145] Correlation coefficient Relevance 0.3＜r＜0.5 Low correlation 0.5＜r＜0.8 Significant correlation 0.8＜r＜1 High relevance r＝1 Completely relevant

[0146] S12. Use normalization methods to make the evaluation indicators consistent and dimensionless.

[0147] The system inertia comprehensive benefit evaluation index system established in this embodiment has the characteristics of multi-level and multi-index. In order to facilitate comparative analysis, it is necessary to eliminate the differentiation of evaluation index unit dimensions. The evaluation index described in step S12 includes extremely large index, extremely small index and intermediate index;

[0148] The method of using step S12 to respectively perform consistency processing on the extremely large index, the extremely small index and the intermediate index is as follows:

[0149] The consistent processing formula for extremely large indicators is as follows:

[0150]

[0151] Where a and b are the lower and upper thresholds of the indicator respectively, and an extremely large indicator is one in which the larger the better.

[0152] The consistent processing formula for very small indicators is as follows:

[0153]

[0154] An extremely small indicator is one where the smaller the better.

[0155] The consistent processing formula for intermediate indicators is as follows:

[0156]

[0157] An intermediate indicator is one that has the best value within a certain interval. The larger the value is, the better it is in the interval from a to b, and the smaller the value is, the better it is in the interval from c to d. That is, a, b, c, and d define the optimal interval and the acceptable interval.

[0158] S2. Constructing the comprehensive benefit model of system inertia:

[0159] S21. As far as the weight determination method is concerned, the existing methods are mainly divided into subjective weighting method and objective weighting method. The subjective weighting method mainly relies on expert knowledge and experience to judge the subjective importance of each indicator. The main methods include eigenvalue method, order relationship method and scheme preference weighting method, etc.; the objective weighting method is based on the original data and uses relatively complete mathematical theories and methods to determine the indicator weights. The main methods include entropy weight method, rough set method, etc.; the subjective weighting method reflects the subjective judgment and intuition of the evaluator, but the evaluation results may produce a certain degree of subjective arbitrariness; the objective weighting method avoids the subjectivity of the evaluation results, but ignores the subjective information necessary for safety evaluation. Therefore, this implementation constructs an integrated weighting model for subjective and objective indicators, and calculates the weights of evaluation indicators;

[0160] Step S21 specifically includes the following steps:

[0161] S211. Constructing the subjective empowerment model of ANP method:

[0162] The Analytic Network Progress (ANP) is a decision analysis method that is adapted to non-independent hierarchical structures and was proposed by Professor Thomas L. Saaty of the University of Pittsburgh in 1996. The typical ANP hierarchy consists of a control layer and a network layer. The control layer includes the problem objectives and decision criteria; the network layer consists of all elements controlled by the control layer, and the elements influence and control each other. The basic principle of ANP is that the weight of each criterion in the control layer can be obtained by the AHP method; the weight of each element in the network layer is determined by the limit supermatrix. It specifically includes the following steps:

[0163] S2111. Systematically analyze and combine decision-making problems to form elements and element sets to determine whether the elements are independent and whether there is a dependent feedback relationship. In the present invention, the goal is to evaluate the inertia of the new energy delivery system. The elements are the three primary indicators of safety benefit, economic benefit, and social benefit, as well as the derived secondary and tertiary indicators.

[0164] S2112. Construct ANP structure: determine the control layer, and analyze the influence and influence relationship of all system elements according to the control criterion layer, and determine the criterion, elements and the relationship between elements;

[0165] S2113, construct ANP super matrix: establish the probability as the main criterion, take one of the risk factors as the secondary criterion, compare the relative importance of other risk factors, that is, compare the importance of other risk factors on the probability of occurrence of this risk factor. Since not every other risk factor has an impact on it, not all elements need to be compared under this criterion, and other risk factors that affect this risk factor can be found in the impact relationship table; then calculate the characteristic vector of each risk category group as a unit, that is, the corresponding local weight vector, and after comparison, judgment and calculation with each element as the secondary criterion, establish the super matrix according to the formula:

[0166]

[0167] Where W ij is a supermatrix, which is represented in the ANP model by the element set R i For the element set R j The impact of each The element is actually composed of R i The elements l and R in jAfter comparing and judging the element k in , the component of a eigenvector in the judgment matrix is ​​obtained;

[0168] Combine the eigenvectors of the interactions of all network layer elements to obtain a supermatrix under the control elements:

[0169]

[0170] Where n 1,2,...,N represents the normalization factor, which is used to ensure that the influence of each element is properly considered; W N1,N2,...,NN is the Nth row of the matrix, indicating the influence weight of the Nth element or criterion on all other elements or criteria;

[0171] Each element of the matrix is ​​a matrix and the sum is 1, but it is not a normalized matrix. For the convenience of calculation, the super matrix is ​​normalized, that is, the elements of the super matrix are weighted to obtain the weighted super matrix a ij is the weighting factor, i,j=1,2,…,n;

[0172] S2114. Calculate the limit supermatrix:

[0173] In order to reflect the dependencies between elements, it is necessary to make a stable output of the supermatrix, that is, to calculate the limit relative sorting vector of each supermatrix:

[0174]

[0175] Formula (8) expresses the limit of the average value of the z-th power of the supermatrix W;

[0176] If the above limit converges and is unique, the value of the corresponding row of the original supermatrix is ​​judged to be the stable weight of each evaluation index;

[0177] S212. Construct an objective weighting model using entropy weight method:

[0178] The entropy weight method is an objective weighting method. It mainly uses information entropy to calculate the entropy weight of each indicator based on the degree of variation of each indicator, and then uses the entropy weight to correct the weight of each indicator to obtain a more objective indicator weight.

[0179] Assuming that the system inertia participates in the comprehensive benefits of the new energy delivery system, different integration methods t={t1,t2,…,t n}, the comprehensive evaluation index system of each integration method is u={u1,u2,…,u n}, Integration method i (i=1,2,…,m) in index u j The attribute value under (j=1,2,…,n) is t ij, then the decision matrix is ​​A=(t ij ) m×n , let M = (1, 2, ..., m), N = (1, 2, ..., n);

[0180] The indicator types include benefit type and cost type, and the benefit type and cost type are dimensionless respectively:

[0181] For benefit-type attributes, let:

[0182]

[0183] For cost attributes, let:

[0184]

[0185] In the formula, r ij is the standardized value, representing the elements in each attribute evaluation index; t ij is the attribute value, which represents the original value in the evaluation matrix or decision matrix, representing the actual measurement value or score of the jth evaluation indicator in the i-th option or situation; is the maximum value among all options of the jth evaluation index; is the minimum value of all options for the jth evaluation index;

[0186] Get the normalized decision matrix R = (r ij ) m×n ;

[0187] Calculate the entropy E of attribute index j j :

[0188]

[0189] Where: j = 1, 2, ..., m, k = 1 / ln (n) is a constant related to the number of samples, the purpose is to make E j ∈[0,1]; r j Satisfying 0<r ij <1 and And when r ij = 0, r ij ln(r ij )=0;

[0190] Calculate the information deviation:

[0191] d j =1-E j (12)

[0192] Calculate indicator weights:

[0193]

[0194] S213. In order to overcome the shortcomings of the subjective and objective weighting method, the weight of the indicator takes into account both subjective and objective factors, and constructs a subjective and objective integrated weighting model. The subjective and objective integrated weighting model with the goal of minimizing the deviation of subjective and objective weights is used to determine the weight of the evaluation indicator:

[0195] S2131. Assume that the subjective weight vector of the operation and maintenance cost evaluation index obtained by the subjective weighting model of the ANP method is w'=(w'1,w'2,…,w' n ) T , and satisfies w' j ∈[0,1],∑ j＝1 w' j =1, and the objective weight vector of operation and maintenance cost evaluation index is calculated by using the entropy weight method w”=(w”,w”2,…,w” n ) T , and satisfies w" j ∈[0,1],∑ j＝1 w" j =1; add the subjective weight and the objective weight to get the final weight vector:

[0196] w=αw'+βw" (14)

[0197] Among them, α, β satisfy α, β > 0, and α + β = 1;

[0198] S2132. In order to fully reflect subjective information and objective information in the scheme ranking, this embodiment starts from the weighted attribute value, considers that the weighted attribute value determined by the subjective weight and the weighted attribute value determined by the objective weight tend to be consistent, and constructs a distance function to represent the weighted value of the subjective and objective weights:

[0199]

[0200] Where s(w',w") represents the similarity or difference between the subjective weight vector w' obtained by the ANP method and the objective weight vector w" obtained by the entropy weight method.

[0201] The distance function described in this embodiment is a squared Euclidean distance function, which is a function that represents the distance between numerical points. The distance function weighting method can be used to organically combine the calculation results of the subjective weighting method and the objective weighting method, thereby retaining the subjectivity of the prior judgment and ensuring the objective characteristics of the data itself.

[0202] S22. Use the ideal point method to improve the matter-element extension model, establish an ideal matter-element extension evaluation model, and calculate the comprehensive closeness;

[0203] The idea of ​​the traditional object-element evaluation method is: first, the level of the evaluation object is divided into several levels according to the existing data, and the data range of each level is given by the database or expert opinion. Then, the indicators of the evaluation object are substituted into the set of each level for multi-indicator evaluation. The evaluation results are compared according to the degree of correlation between it and the set of each level. The greater the correlation, the better the degree of conformity with its level set. The reliability of the traditional object-element evaluation results mainly depends on the rationality of the setting of the classical domain and the section domain. When the setting of the classical domain range is unreasonable, the evaluation objects may be at the same evaluation level, and only a partial ranking of the evaluation scheme can be obtained. In order to overcome this problem, this embodiment adopts the following steps:

[0204] The ideal matter-element extension evaluation model described in step S22 is based on the matter-element extension model, by setting positive and negative ideal matter-elements to replace the classical domain, and using closeness to replace the maximum membership criterion to determine the distance between the matter-element to be evaluated and the positive and negative ideal matter-elements.

[0205] Positive and negative ideal matter elements are determined based on expert opinions;

[0206] Alternatively, select the best and worst data (the maximum (optimal) and minimum (worst) values ​​in the possible value set of the indicator) from each evaluation indicator in the object-element to be evaluated as the positive ideal object-element and negative ideal object-element respectively. i is a positive indicator, then For the index c i The optimal data; if the evaluation index c i is a reverse indicator, then If the evaluation index c i is a moderate indicator, then If the indicator c i is an interval indicator, then Take interval value [u 1i ,u 2i ] represents the optimal interval. Thus, we obtain the positive ideal element R + and negative ideal element R - .

[0207] Closeness refers to the degree of closeness between the evaluated sample and the standard sample. The larger the value, the closer the two are, and vice versa. Therefore, the quality of each evaluation object can be directly judged based on the degree of closeness to the positive and negative ideal elements. The positive and negative ideal elements can also be divided into different levels, and the closeness to each level is calculated to determine the quality of the site selection. This embodiment uses asymmetric closeness as the closeness function, calculates the closeness function value, and sorts it out to get:

[0208]

[0209] Where N is the closeness; D is the distance, which represents the difference measure of a single evaluation index; w i is the weight of index i, then the closeness of the object-element to be evaluated to each level is:

[0210]

[0211] In the formula, w i (X) represents the weight function of evaluation index i, which depends on the decision variable set X; D j (v' i ) represents the closeness between the matter-element R0 to be evaluated and the normalized ideal matter-element classical domain, which includes the positive closeness and negative proximity The calculation formula is as follows:

[0212]

[0213]

[0214] In the formula, v' i is the actual value under index i;

[0215] The closeness of the matter-element to be evaluated to the positive ideal matter-element and the negative ideal matter-element are:

[0216]

[0217]

[0218] The comprehensive closeness is:

[0219]

[0220] S3. Evaluate the system inertia of the object-element to be evaluated based on the comprehensive closeness.

[0221] Simulation Experiment Design

[0222] In order to analyze the comprehensive benefits of system inertia when different power sources participate in the microgrid, the indicators listed in Table 1 are used as benefit evaluation indicators, and the qualitative indicators are assigned values ​​through questionnaires to obtain the initial assignment results of different indicators.

[0223] Table 3 Initial values ​​of benefit evaluation indicators for comprehensive benefits of system inertia

[0224] index unit type C4 Yuan Very small C5 Yuan Very small C6 Yuan Very small C7 - Very large C1 - Very large C2 - Very large C3 - Very large

[0225] Based on the initial values ​​of the benefit evaluation indicators in Table 3, in order to facilitate analysis and evaluation, it is necessary to test the correlation of the evaluation indicators, delete the evaluation indicators with higher correlation, and select independent and representative benefit evaluation indicators (Table 4). Then, according to the simplified benefit evaluation indicators, the ANP method and the entropy weight method are used to obtain the subjective weight and objective weight of the comprehensive benefit evaluation index of the inertia of the new energy delivery system as a comparison ratio. At the same time, since the subjective weight is too dependent on the subjective experience of experts and lacks consideration of objective data, the objective weight is completely dependent on objective data. Therefore, in order to optimize the weight result, the minimum deviation degree of the weight coefficient obtained by the subjective weighting method and the objective weighting method described in the present invention is used as the optimization goal to calculate the comprehensive weight of the comprehensive benefit evaluation index of the inertia of the new energy delivery system.

[0226] Table 4 Calculation of system inertia comprehensive benefit evaluation weights

[0227] index ANP method Entropy Weight Method The subjective and objective index integrated weighting method C1 0.201 0.123 0.169 C6 0.098 0.107 0.102 C2 0.129 0.211 0.162 C3 0.221 0.084 0.165 C5 0.108 0.107 0.108 C4 0.168 0.134 0.154 C7 0.075 0.234 0.140

[0228] After normalizing each indicator, the ideal matter-element extension model is used to evaluate the comprehensive benefits of system inertia, and the following positive and negative ideal matter-element R is constructed: + , R - :

[0229]

[0230] Then, the closeness and comprehensive closeness between the object to be evaluated and the positive and negative ideal matter-elements are calculated to obtain the comprehensive benefit evaluation result of the inertia of the new energy delivery system.

[0231] Table 5 Comprehensive closeness of the matter-element to be evaluated

[0232] Inertia Control Technology Posting progress Negative progress Comprehensive progress Ranking Synchronous generator 0.928 1.236 0.308 1 Asynchronous motors 1.391 1.008 0.226 3 Units with virtual inertia 1.279 1.158 0.256 2 Synchronous condenser 1.393 0.891 0.210 4

[0233] According to the above evaluation results, considering the safety, economic and social benefits, the traditional inertia of synchronous generators is still the main part of the system inertia, which resists the frequency change in the system. With the large-scale application of new energy power generation technologies represented by wind and light, virtual inertia technology, that is, through specific equipment, new energy, energy storage or DC transmission can provide inertia that can replace synchronous generators to meet the requirements of the system, has shown an upward trend in the proportion of system inertia.

[0234] In general, the comprehensive inertia benefits of the new energy transmission system are closely related to the operation of the synchronous generator sets in the system, and are also closely related to the control of the units with virtual inertia. To maximize the comprehensive benefits, it is necessary to maintain the stability of the system frequency, and consider both the frequency regulation requirements of the system inertia level and the inertia requirements of the system frequency regulation capability.

[0235] Therefore, the present invention adopts the above-mentioned new energy transmission system inertia evaluation method based on the comprehensive benefit evaluation model, establishes a system inertia comprehensive index evaluation system from multiple dimensions, selects the subjective and objective index integrated weighting model to calculate the index weight, and uses the ideal point method to improve the traditional material-element extension model, so as to obtain a relatively accurate evaluation result.

[0236] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A new energy delivery system inertia assessment method based on a comprehensive benefit evaluation model, characterized by: The following steps are involved: S1. Construct a comprehensive evaluation index system for system inertia; S2. Constructing the comprehensive benefit model of system inertia: S21. Construct an integrated weighting model for subjective and objective indicators and calculate the weights of evaluation indicators; S22. Use the ideal point method to improve the matter-element extension model, establish an ideal matter-element extension evaluation model, and calculate the comprehensive closeness; S3. Evaluate the system inertia of the object-element to be evaluated based on the comprehensive closeness.

2. The inertia evaluation method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 1 is characterized in that: The system inertia comprehensive evaluation index system described in step S1 includes three evaluation angles: safety benefit, economic benefit, and social benefit; Among them, the evaluation of safety benefits is classified into system frequency safety benefits, which include the following evaluation indicators: non-synchronous power penetration rate, ratio of system inertia to maximum power loss during N-1 accident, and safe operation inertia; Economic benefits include the following evaluation categories: power generation side benefits, grid side benefits and user side benefits. The classification indicators of power generation side benefits, grid side benefits and user side benefits are investment cost, operation and maintenance cost and economic present value respectively. The evaluation category of social benefits is improving power supply reliability, and the evaluation indicator for improving power supply reliability is load stability.

3. The inertia assessment method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 2 is characterized in that: Step S1 specifically includes the following steps: S11. Use correlation test methods to analyze the overlap and inspection status between evaluation indicators and establish an indicator system with independent characteristics; S12. Use normalization methods to make the evaluation indicators consistent and dimensionless.

4. The inertia assessment method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 3 is characterized in that: In step S11, correlation analysis is performed according to the types of evaluation indicators, which specifically includes the following steps: S111. Assume that there are n demand response implementation benefit evaluation indicators (I1, I2, ..., I n ), according to the classification when selecting the evaluation indicators, each type of evaluation indicators contains k indicators in total, and the correlation test is performed on each type of evaluation indicators according to the least squares method to construct the correlation test matrix R: In the formula, r αβ (α, β = 1, 2, ..., k) is the correlation coefficient between the αth index and the βth index, and the specific calculation is as follows: In the formula, I α are all observed values ​​of indicator α; is the average of all observed values ​​of indicator α; I β are all observed values ​​of the indicator β; is the average of all observed values ​​of indicator β; And when α=β, r αβ =1, then r αβ =r βα ; S112. Determine whether the relevance requirement is met according to the set relevance judgment principle.

5. The inertia assessment method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 3 is characterized in that: The evaluation index described in step S12 includes an extremely large index, an extremely small index, and an intermediate index; The method of using step S12 to respectively perform consistency processing on the extremely large index, the extremely small index and the intermediate index is as follows: The consistent processing formula for extremely large indicators is as follows: In the formula, a and b are the lower and upper thresholds of the indicator respectively; The consistent processing formula for very small indicators is as follows: The consistent processing formula for intermediate indicators is as follows:

6. The inertia assessment method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 1 is characterized in that: Step S21 specifically includes the following steps: S211. Constructing the subjective empowerment model of ANP method: S2111. Systematically analyze and combine decision problems to form elements and element sets to determine whether the elements are independent and whether there is a dependent feedback relationship; S2112. Construct ANP structure: determine the control layer, and analyze the influence and influence relationship of all system elements according to the control criterion layer, and determine the criterion, elements and the relationship between elements; S2113, construct ANP super matrix: establish probability as the main criterion, take one of the risk factors as the secondary criterion, compare the relative importance of other risk factors, and then calculate the characteristic vector of each risk category group as a unit, that is, the corresponding local weight vector, and establish the super matrix according to the formula after comparison, judgment and calculation with each element as the secondary criterion: Where W ij is a supermatrix, which is represented in the ANP model by the element set R i For the element set R j The impact of each The element is composed of R i The elements l and R in j After comparing and judging the element k in , the component of an eigenvector in the supermatrix is ​​obtained; Combine the eigenvectors of the interactions of all network layer elements to obtain a supermatrix under the control elements: Where n 1，2，...，N represents the normalization factor; W N1，N2，...，NN is the influence weight of the Nth element or criterion in the Nth row of the matrix on all other elements or criteria; Normalize the supermatrix, that is, weight the elements of the supermatrix to obtain the weighted supermatrix a ij is a weighting factor, i, j = 1, 2, ..., n; S2114. Calculate the limit supermatrix: Compute the limiting relative ranking vector for each supermatrix: Formula (8) expresses the limit of the average value of the z-th power of the supermatrix W; If the above limit converges and is unique, the value of the corresponding row of the original supermatrix is ​​judged to be the stable weight of each evaluation index; S212. Construct an objective weighting model using entropy weight method: Assuming that the system inertia participates in the comprehensive benefits of the new energy delivery system, different integration methods t = {t1, t2, ..., t n }, the comprehensive evaluation index system of each integration method is u={u1,u2,...,u n }, Integration method i (i=1,2,...,m) in index u j The attribute value under (j=1, 2, ..., n) is t ij , then the decision matrix is ​​A=(t ij ) m×n , let M = (1, 2, ..., m), N = (1, 2, ..., n); The indicator types include benefit type and cost type, and the benefit type and cost type are dimensionless respectively: For benefit-type attributes, let: For cost attributes, let: In the formula, r ij is the standardized value, representing the elements in each attribute evaluation index; t ij is the attribute value, which represents the original value in the evaluation matrix or decision matrix, representing the actual measurement value or score of the jth evaluation indicator in the i-th option or situation; is the maximum value among all options of the jth evaluation index; is the minimum value of all options for the jth evaluation index; Get the normalized decision matrix R = (r ij ) m×n ; Calculate the entropy E of attribute index j j : Where: j = 1, 2, ..., m, k = 1 / ln(n) is a constant related to the number of samples, the purpose is to make E j ∈[0,1]; r j Satisfying 0<r ij <1 and And when r ij = 0, r ij ln(r ij )=0; Calculate the information deviation: d j =1-E j (12) Calculate indicator weights: S213. Construct a subjective and objective integrated weighting model, and determine the weights of evaluation indicators with the subjective and objective integrated weighting model that minimizes the deviation between subjective and objective weights: S2131. Assume that the subjective weight vector of the operation and maintenance cost evaluation index obtained by the subjective weighting model of the ANP method is w′=(w′1, w′2, ..., w′ n ) T , and satisfies w′ j ∈[0,1],∑ j＝1 w′ j =1, and the objective weight vector of operation and maintenance cost evaluation index is calculated by using the entropy weight method w″=(w″1, w″2, ..., w″ n ) T , and satisfies w″ j ∈[0,1],∑ j＝1 w″ j =1; add the subjective weight and the objective weight to get the final weight vector: w=αw'+βw" (14) Among them, α, β satisfy α, β > 0, and α + β = 1; S2132. Considering that the weighted attribute value determined by the subjective weight tends to be consistent with the weighted attribute value determined by the objective weight, a distance function is constructed to represent the weighted value of the subjective and objective weights: Where s(w',w") represents the similarity or difference between the subjective weight vector w' obtained by the ANP method and the objective weight vector w" obtained by the entropy weight method.

7. The inertia assessment method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 6 is characterized in that: The ideal matter-element extension evaluation model described in step S22 is based on the matter-element extension model, by setting positive and negative ideal matter-elements to replace the classical domain, and using closeness to replace the maximum membership criterion to determine the distance between the matter-element to be evaluated and the positive and negative ideal matter-elements.

8. The inertia assessment method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 7 is characterized in that: Positive and negative ideal matter elements are determined based on expert opinions; Alternatively, select the best and worst data from each evaluation index in the matter-element to be evaluated as the positive ideal matter-element and negative ideal matter-element respectively. If the evaluation index c i is a positive indicator, then For the index c i The optimal data; if the evaluation index c i is a reverse indicator, then If the evaluation index c i is a moderate indicator, then If the indicator c i is an interval indicator, then Take interval value [u 1i ,u 2i ] represents the optimal interval; Thus we get the positive ideal element R + and negative ideal element R - .

9. The inertia assessment method for a new energy delivery system based on a comprehensive benefit evaluation model according to claim 8 is characterized in that: Using asymmetric proximity as the proximity function, calculate the proximity function value and get: Where N is the closeness; D is the distance, which represents the difference measure of a single evaluation index; w i is the weight of index i, then the closeness of the object-element to be evaluated to each level is: In the formula, w i (X) represents the weight function of evaluation index i, which depends on the decision variable set X; D j (v′ i ) represents the closeness between the matter-element R0 to be evaluated and the normalized ideal matter-element classical domain, which includes the positive closeness and negative proximity The calculation formula is as follows: In the formula, v′ i is the actual value under index i; The closeness of the matter-element to be evaluated to the positive ideal matter-element and the negative ideal matter-element are: The comprehensive closeness is:

Images