CN111311042B - Electric power emergency plan evaluation method and device - Google Patents

Abstract

The invention discloses a method and a device for evaluating an electric power emergency plan, wherein the method comprises the following steps: constructing a first index evaluation system of the power emergency plan to be evaluated; carrying out assignment and standardization processing on each index in the first index evaluation system; calculating an index weight value of a first index evaluation system by adopting a weight division model; evaluating the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result; analyzing factors and principal component factors of the power emergency plan to be evaluated to obtain the cumulative variance contribution rate of each secondary evaluation index; screening main secondary evaluation indexes of a first index evaluation system; the first index evaluation system is reconstructed based on the main secondary indexes under each primary evaluation index to obtain a second index evaluation system, the reasonability of the second index evaluation system is judged, the evaluation indexes can be screened and optimized autonomously, and the calculation speed of the evaluation result can be effectively improved.

Description

Electric power emergency plan evaluation method and device

Technical Field

The invention relates to the technical field of power grid evaluation, in particular to a method and a device for evaluating an electric power emergency plan.

Background

At present, in an existing electric power emergency plan evaluation system, the influence of the importance of some indexes on an evaluation result can be very little, if the influence of the indexes is still considered, the evaluation system is too complicated and redundant, and meanwhile, the calculation speed of the result is also influenced.

Disclosure of Invention

The embodiment of the invention provides an electric power emergency plan evaluation method and device, which can autonomously screen and optimize evaluation indexes, effectively improve the calculation speed of an evaluation result and enable the evaluation result to be scientific.

An embodiment of the present invention provides a method for evaluating an electric power emergency plan, including:

acquiring electrical quantity data of a power grid, and constructing a first index evaluation system of an electric power emergency plan to be evaluated according to the electrical quantity data; the first index evaluation system comprises a plurality of first-level evaluation indexes and second-level evaluation indexes corresponding to the first-level evaluation indexes;

assigning and standardizing each index in the first index evaluation system to obtain an evaluation index value of the first index evaluation system;

calculating an index weight value of the first index evaluation system by adopting a preset weight division model;

evaluating the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result;

analyzing factors of the power emergency plan to be evaluated and principal component factors of the first index evaluation system to obtain the cumulative variance contribution rate of each secondary evaluation index;

screening main secondary evaluation indexes of the first index evaluation system according to the cumulative variance contribution rate of the secondary evaluation indexes;

reconstructing the first index evaluation system based on the main secondary indexes under each primary evaluation index to obtain a second index evaluation system, and judging the rationality of the second index evaluation system;

and when the second index evaluation system is detected to be reasonable, outputting the evaluation result of the plan.

As an improvement of the above scheme, the assigning and standardizing each index in the first index evaluation system to obtain an evaluation index value of the first index evaluation system specifically includes:

assigning the voltage grade conversion times in the secondary evaluation indexes by adopting a formula (1), and assigning qualitative indexes in the secondary evaluation indexes by adopting a preset triangular fuzzy number three-scale method to obtain the assignment of the secondary evaluation indexes:

wherein y is the assignment of the voltage level conversion frequency index, and x is the voltage level conversion frequency;

obtaining the assignment of the primary evaluation index according to the assignment of the secondary evaluation index, thereby obtaining the assignment of each index in the first index evaluation system;

carrying out standardization processing on the assignment of each index in the first index evaluation system according to a formula (2) to obtain an evaluation index value of the first index evaluation system:

wherein, a'_ijThe evaluation index value of the jth index in the ith to-be-evaluated electric power emergency plan, a_ijAssigning a value to the jth index in the ith power emergency plan to be evaluated,

the minimum value of j index assignment in m electric power emergency plans to be evaluated is represented,

and j is 1,2, …, n,i＝1,2,…,m。

as an improvement of the above scheme, the weight classification model includes an adaptive weight classification model, and the adaptive weight classification model is constructed by the following steps:

and (3) obtaining the similarity between each index and the similar index in the ideal optimal solution according to the formula:

wherein d is_ijTo the degree of similarity, a_jAssigning a j index;

and (3) constructing an adaptive weight division model according to the similarity between each index and the similar index in the ideal optimal solution and a formula (4):

wherein, ω is_jThe index weight value of each index of the first index evaluation system is obtained;

obtaining the constraint condition of the self-adaptive weight partitioning model according to a formula (5):

wherein ξ_ijCalculating a gray correlation coefficient for the jth index in the ith to-be-evaluated electric power emergency plan and the j-class index in the ideal optimal solution according to a formula (6):

where ρ is a resolution coefficient, 0< ρ < 1.

As an improvement of the above scheme, the weight classification model further includes a variable weight model, and the variable weight model is constructed by the following steps:

when the index weight value of the first index evaluation system is detected to exceed a preset normal weight threshold value, introducing a state variable weight factor based on the self-adaptive weight division model, and constructing the variable weight model;

obtaining the variable weight of the first index evaluation system according to a formula (7) by adopting the variable weight model:

wherein, ω is_eiIs the variable weight, ω_iAs an attribute weight value of each index, ω ═ ω (ω ═ ω)₁,ω₂,…,ω_n)，a_iFor the optimal solution of each index, S (a)_i) Obtaining the state variable weight factor of each index according to a formula (8):

where t is the penalty level, max (a)_i) Min (a) is the maximum value of the reasonable values of the indices_i) Is the minimum value of reasonable values of each index.

As an improvement of the above scheme, the evaluating the to-be-evaluated power emergency plan according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result specifically includes:

constructing a weighting decision matrix according to the evaluation index value of the first index evaluation system and the index weight value;

calculating a positive ideal solution and a negative ideal solution of the power emergency plan to be evaluated according to the weighting decision matrix;

calculating a first Euclidean distance between the power emergency plan to be evaluated and the positive ideal solution and a second Euclidean distance between the power emergency plan to be evaluated and the negative ideal solution according to the positive ideal solution and the negative ideal solution;

and calculating the relative closeness of the power emergency plan to be evaluated according to the first Euclidean distance and the second Euclidean distance, and taking the relative closeness as a plan evaluation result of the power emergency plan to be evaluated.

As an improvement of the above, the method further comprises:

obtaining the elements of the weighted decision matrix according to equation (9):

a_ij″＝ω_j×a_ij′(i＝1,2,…,m,j＝1,2,…,n) (9)

wherein, a_ij"is an element of the weighted decision matrix;

obtaining a positive ideal solution of the power emergency plan to be evaluated according to a formula (10):

wherein, (a')⁺Is the positive ideal solution;

obtaining a negative ideal solution of the power emergency plan to be evaluated according to a formula (11):

wherein, (a')^-Is the negative ideal solution;

obtaining a first Euclidean distance between the power emergency plan to be evaluated and the positive ideal solution according to a formula (12):

wherein,

the first Euclidean distance, a ', from the ith to-be-evaluated power emergency plan to the positive ideal solution'_iEvaluating index values of all indexes in the ith to-be-evaluated electric power emergency plan;

obtaining a second Euclidean distance from the power emergency plan to be evaluated to the negative ideal solution according to a formula (13):

wherein,

a second Euclidean distance from the ith power emergency plan to be evaluated to the negative ideal solution;

obtaining the relative closeness of the power emergency plan to be evaluated according to a formula (14):

wherein, C_iFor the improvement of the scheme, which is the relative closeness of the ith power emergency plan to be evaluated, analyzing the factor of the power emergency plan to be evaluated and the principal component factor of the first index evaluation system to obtain the cumulative variance contribution rate of each secondary evaluation index, specifically comprising:

obtaining an index data matrix of the first index evaluation system according to the evaluation index value of the first index evaluation system, and performing unification processing on the index data matrix of the first index evaluation system to obtain a unified index data matrix;

establishing a correlation matrix according to the unified index data matrix, and calculating a characteristic value and a characteristic vector of the correlation matrix;

carrying out standardized orthogonal transformation on the eigenvalue and the eigenvector of the correlation matrix to obtain an orthogonal matrix;

obtaining a principal component model of the first index evaluation system according to the index data matrix and the orthogonal matrix;

and calculating the variance contribution rate of each index in the first index evaluation system according to the principal component model, and obtaining the accumulated variance contribution rate of each secondary evaluation index according to the variance contribution rate.

As an improvement of the above scheme, the screening of the main secondary evaluation indexes of the first index evaluation system according to the cumulative variance contribution rate of the secondary evaluation indexes specifically includes:

sorting the cumulative variance contribution rate of the secondary evaluation indexes under each primary evaluation index from large to small, and screening the secondary evaluation indexes sorted to the front b bits as main secondary evaluation indexes; and b is not more than the total number of the secondary evaluation indexes under the corresponding primary evaluation indexes.

As an improvement of the above scheme, the judging the rationality of the second index evaluation system specifically includes:

constructing an index data matrix of the second index evaluation system;

calculating the information contribution rate of the main secondary evaluation indexes under each primary evaluation index to the secondary evaluation indexes according to the trace of a covariance matrix corresponding to the index data matrix of the first index evaluation system and the index data matrix of the second index evaluation system;

judging whether the information contribution rate exceeds a preset information contribution rate threshold value or not; if so, the second index evaluation system is considered to be reasonable, and the plan evaluation result is output; and if not, determining that the second index evaluation system is unreasonable, and assigning and standardizing each index in the second index evaluation system again.

Another embodiment of the present invention correspondingly provides an electric power emergency plan evaluation device, including:

the index evaluation system construction module is used for acquiring electrical quantity data of a power grid and constructing a first index evaluation system of the power emergency plan to be evaluated according to the electrical quantity data; the first index evaluation system comprises a plurality of first-level evaluation indexes and second-level evaluation indexes corresponding to the first-level evaluation indexes;

the index assignment module is used for assigning and standardizing each index in the first index evaluation system to obtain an evaluation index value of the first index evaluation system;

the weight calculation module is used for calculating an index weight value of the first index evaluation system by adopting a preset weight division model;

the plan evaluation module is used for evaluating the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result;

the factor analysis module is used for analyzing the factors of the power emergency plan to be evaluated and the principal component factors of the first index evaluation system to obtain the cumulative variance contribution rate of each secondary evaluation index;

the index screening module is used for screening main secondary evaluation indexes of the first index evaluation system according to the accumulated variance contribution rate of the secondary evaluation indexes;

the index evaluation system reconstruction module is used for reconstructing the first index evaluation system based on the main secondary indexes under each primary evaluation index to obtain a second index evaluation system and judging the rationality of the second index evaluation system;

and the evaluation result output module is used for outputting the plan evaluation result when the second index evaluation system is detected to be reasonable.

Compared with the prior art, the method and the device for evaluating the electric power emergency plan disclosed by the embodiment of the invention have the following beneficial effects:

1. the method adopts the self-adaptive weight division model, can effectively avoid the influence of artificial subjective factors on weight division, can effectively simplify the calculation method of the index weight, can realize the correction of the index weight by adopting the variable weight model, can effectively avoid the phenomenon that the evaluation index is excessively deviated from a normal value and the final evaluation result is not reflected, can effectively improve the evaluation accuracy and reduce the error.

2. The method can autonomously update and optimize the evaluation index system, remove redundant indexes in the index system through screening conditions, effectively improve the calculation speed of the evaluation result, realize the construction of a proper index system aiming at different electric power special plans, have stronger applicability, and have more scientific and reliable evaluation results.

Drawings

Fig. 1 is a schematic flow chart of a method for evaluating an electric power emergency plan according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of an electric power emergency plan evaluation device provided in the second embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Example one

Referring to fig. 1, a schematic flow chart of a method for evaluating an electric power emergency plan according to an embodiment of the present invention is shown, where the method includes steps S101 to S108.

S101, acquiring electrical quantity data of a power grid, and constructing a first index evaluation system of an electric power emergency plan to be evaluated according to the electrical quantity data; the first index evaluation system comprises a plurality of first-level evaluation indexes and second-level evaluation indexes corresponding to the first-level evaluation indexes.

It should be noted that the electrical quantity data of the power grid includes the number of voltage class conversion times, the number of switching operations of the main electrical equipment, the recovered load order, the important load recovery degree, the power failure time, the power loss, the node voltage constraint condition, the power constraint condition, the start power optimization of the black start power supply, and the like. The index evaluation system constructed by the invention comprises six functional departments in the power emergency process, namely a dispatching department, a transportation and inspection department, a communication department, a material department, a maintenance department and a safety supervision department. Furthermore, the evaluation index adopted by the invention is divided from the electric quantity data and the functional department and is used as a primary evaluation index for the power emergency plan evaluation. Furthermore, a plurality of secondary evaluation indexes are extracted under each primary evaluation index. Specifically, according to the designation of relevant power emergency guidance rules, the treatment processes of the past emergency plans and emergency accidents of the power grid enterprises are extracted and analyzed, and therefore each practical and effective secondary evaluation index is determined.

For example, the primary evaluation index may be power grid scheduling, condition maintenance, communication guarantee, material guarantee, system architecture, safety supervision, and the like. Furthermore, secondary evaluation indexes corresponding to power grid dispatching include, but are not limited to, voltage level conversion times, electrical switch operation times, recovered load order and power constraint conditions; the secondary evaluation indexes corresponding to the state overhaul include but are not limited to the number of accident emergency repair personnel, the time of the rescue department arriving at the site, the time of starting a disaster emergency system and the integrity of overhaul resources; the secondary evaluation indexes corresponding to the communication guarantee include but are not limited to the transmission time of emergency information, the external release time of an emergency, the capacity of an emergency database and the supplementary updating capacity of the database; the secondary evaluation indexes corresponding to the material guarantee include but are not limited to the number of emergency rescue vehicles, the number of emergency power generation lighting vehicles, the emergency road guarantee degree and the financial expense investment percentage; the secondary evaluation indexes corresponding to the system architecture include, but are not limited to, power grid overhaul capacity, electric main wiring reliability, reactive compensation capacity and load permeability; the secondary evaluation indexes corresponding to the safety supervision system include but are not limited to power grid disaster monitoring capacity, monitoring center facility completeness, gathering multi-aspect information capacity and auxiliary decision software function perfection.

And S102, carrying out assignment and standardization processing on each index in the first index evaluation system to obtain an evaluation index value of the first index evaluation system.

In an alternative embodiment, the secondary evaluation index is assigned by:

assigning the voltage grade conversion times in the secondary evaluation indexes by adopting a formula (1), and assigning qualitative indexes in the secondary evaluation indexes by adopting a preset triangular fuzzy number three-scale method to obtain the assignment of the secondary evaluation indexes:

wherein y is the assignment of the voltage level conversion frequency index, and x is the voltage level conversion frequency.

In the embodiment, a value is assigned to a secondary evaluation index in the power emergency plan. For the quantitative indexes in the secondary evaluation indexes, assignment is carried out on the quantitative indexes according to assignment rules in an event library, for example, the voltage level conversion times are assigned, the voltage changes once when the voltage passes through one transformer substation while the voltage passes through the transformer substation in the electric power emergency plan, the voltage level conversion times are used as the secondary evaluation indexes under the power grid dispatching, and the purpose is to ensure the stability of the voltage of the power grid nodes.

For qualitative indexes in the secondary evaluation indexes, assignment of the qualitative indexes is represented by a grade description method, and then the qualitative indexes are quantized by adopting a triangular fuzzy number, and the method specifically comprises the following steps:

definition 1: if N ═ x, y, z, where x, y, z are the three scales of quantitative indices, 0< x ≦ y ≦ z, N is a triangular ambiguity number, α ═ y-x, β ═ z-y, and α ═ β ═ 1;

definition 2: the sharpness value of the triangular blur number is N' ═ x +2y + z)/4.

The embodiment of the invention adopts a three-scale method of the triangular fuzzy number to quantize the qualitative index into the triangular fuzzy number, and then converts the triangular fuzzy number into a clear value according to the formula, thereby facilitating data calculation.

Further, according to the assignment of the secondary evaluation index, the assignment of the primary evaluation index is obtained, and therefore the assignment of each index in the first index evaluation system is obtained.

Further, the evaluation index value of the first index evaluation system is obtained by normalizing the assignment of each index in the first index evaluation system according to formula (2):

wherein, a_i′_jThe evaluation index value of the jth index in the ith to-be-evaluated electric power emergency plan, a_ijAssigning a value to the jth index in the ith power emergency plan to be evaluated,

the minimum value of the j-th index in the m power emergency plans to be evaluated is represented,

and j is 1,2, …, and n, i is 1,2, …, m.

It can be understood that the data standardization processing is performed on the assignment of the secondary evaluation index, and the problem that the reverse order often occurs when the traditional solution is applied can be effectively avoided.

S103, calculating an index weight value of the first index evaluation system by adopting a preset weight division model.

In an optional embodiment, the method for determining the self-adaptive optimal weight is used for dividing the weight of the indexes of the power emergency plan to be evaluated. Firstly, assume that there are m electric power emergency plans to be evaluated, and each plan has n indexes.

The ideal optimal solution of each evaluation index is as follows:

the ideal optimal solution can be the maximum value of each index in the power emergency plan to be evaluated, and is also the positive-end emergency plan.

Further, according to the formula (3), the similarity d between each index and the similar index in the ideal optimal solution is calculated_ij：

Wherein d is_ijD, representing the similarity between the jth index in the ith to-be-evaluated electric power emergency plan and the j indexes in the ideal optimal solution_ijThe higher the index value is, the better the index value is, otherwise, the opposite is true.

Further, the similarity between the ith power emergency plan to be evaluated and the ideal optimal solution is obtained through weighted sum:

further, according to the similarity between each index and the similar index in the ideal optimal solution, a single-target linear programming model, namely an adaptive weight division model, is constructed according to a formula (4):

wherein, ω is_jAnd evaluating the weight value of each index of the system for the first index.

Meanwhile, the weight of each index should have a certain limit range, i.e. a constraint condition of the index weight. Firstly, an ideal optimal solution of the power emergency plan is constructed, and the correlation coefficient of each plan to be evaluated and the ideal optimal solution under each evaluation index is obtained.

Using xi_ijRepresenting the gray correlation coefficient of the jth index in the ith to-be-evaluated electric power emergency plan and the j indexes in the ideal optimal solution, and calculating the gray correlation coefficient according to a formula (6):

where ρ is a resolution coefficient, 0< ρ < 1.

The proportion of the correlation coefficient of each evaluation index under each emergency plan and the ideal optimal solution to the gray correlation coefficient of the total emergency plan reflects the inherent significance of fluctuation of different evaluation indexes between emergency plans and reflects the importance degree of each index to a certain extent, so that the weight interval of each evaluation index is determined according to the maximum value and the minimum value of the proportion. Meanwhile, the variance value interval of each index weight is determined by the maximum variance value and the minimum variance value of the ratio of the grey correlation coefficient of each index and the grey correlation coefficient of the ideal optimal solution in each plan to the grey correlation coefficient of the total plan.

Therefore, the constraint condition of the adaptive weight-binning model is obtained according to equation (5):

in another preferred embodiment, on the basis of the above embodiment, the weight classification model further includes a variable weight model, and the variable weight model is constructed by:

when the index weight value of the first index evaluation system is detected to exceed a preset normal weight threshold value, introducing a state variable weight factor based on the self-adaptive weight division model, and constructing the variable weight model;

obtaining the variable weight of the first index evaluation system according to a formula (7) by adopting the variable weight model:

wherein, ω is_eiIs the variable weight, ω_iAs an attribute weight value of each index, ω ═ ω (ω ═ ω)₁,ω₂,…,ω_n)，a_iFor the optimal solution of each index, S (a)_i) Obtaining the state variable weight factor of each index according to a formula (8):

where t is the penalty level, max (a)_i) Min (a) is the maximum value of the reasonable values of the indices_i) Is the minimum value of the reasonable values of the indexes.

Currently, in an evaluation system, some indexes are not invariable to the final evaluation result, and the index values of the indexes have the influence of qualitative change on the final evaluation result after the 'quantitative change' exceeds a certain standard. Taking a permanent emergency commanding means as an example, if the index value is reduced to a certain level, the final evaluation result should be rapidly deteriorated even if the other index values in the power system emergency index are high, and it is difficult to reflect the phenomenon of "quality deterioration due to quantity change" described above by using a permanent weight with a fixed weight. Therefore, the method adopts the variable weight model, can effectively correct the final weight, and can effectively avoid the phenomenon that the evaluation index is excessively deviated from a normal value and the final evaluation result is not reflected.

And S104, evaluating the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result.

In a preferred embodiment, step S104 includes:

constructing a weighting decision matrix according to the evaluation index value of the first index evaluation system and the index weight value;

calculating a positive ideal solution and a negative ideal solution of the power emergency plan to be evaluated according to the weighting decision matrix;

calculating a first Euclidean distance between the electric power emergency plan to be evaluated and the positive ideal solution and a second Euclidean distance between the electric power emergency plan to be evaluated and the negative ideal solution according to the positive ideal solution and the negative ideal solution;

and calculating the relative closeness of the power emergency plan to be evaluated according to the first Euclidean distance and the second Euclidean distance, and taking the relative closeness as a plan evaluation result of the power emergency plan to be evaluated.

Preferably, the elements of the weighted decision matrix are obtained according to equation (9):

a_ij″＝ω_j×a_ij′(i＝1,2,…,m,j＝1,2,…,n) (9)

wherein, a_ijIs an element of the weighted decision matrix

Thus, the weighted decision matrix is a ″ ═ a_ij″)_m×n。

Preferably, a positive ideal solution of the power emergency plan to be evaluated, namely a positive terminal scheme, is obtained according to the formula (10):

wherein, (a')⁺Is the positive ideal solution.

Preferably, a negative ideal solution of the power emergency plan to be evaluated, namely a negative terminal scheme, is obtained according to the formula (11):

wherein, (a')^-Is the negative ideal solution.

Preferably, a first euclidean distance between the power emergency plan to be evaluated and the ideal solution is obtained according to formula (12):

wherein,

the first Euclidean distance, a ', from the ith to-be-evaluated power emergency plan to the positive ideal solution'_iAnd the evaluation index value is the evaluation index value of each index in the ith power emergency plan to be evaluated.

Preferably, a second euclidean distance between the power emergency plan to be evaluated and the negative ideal solution is obtained according to formula (13):

wherein,

and the second Euclidean distance from the ith power emergency plan to be evaluated to the negative ideal solution.

Preferably, the relative closeness of the power emergency plan to be evaluated is obtained according to the formula (14):

wherein, C_iAnd the relative closeness of the ith power emergency plan to be evaluated. And sequencing all the electric power emergency plans to be evaluated according to the relative closeness. When the relative closeness is larger, the power emergency plan is more ideal, and vice versa, the power emergency plan is less ideal.

And S105, analyzing the factor of the power emergency plan to be evaluated and the main component factor of the first index evaluation system to obtain the cumulative variance contribution rate of each secondary evaluation index.

In a preferred embodiment, step S105 includes:

obtaining an index data matrix of the first index evaluation system according to the evaluation index value of the first index evaluation system, and performing unification processing on the index data matrix of the first index evaluation system to obtain a unified index data matrix;

establishing a correlation matrix according to the unified index data matrix, and calculating a characteristic value and a characteristic vector of the correlation matrix;

carrying out standardized orthogonal transformation on the eigenvalue and the eigenvector of the correlation matrix to obtain an orthogonal matrix;

obtaining a principal component model of the first index evaluation system according to the index data matrix and the orthogonal matrix;

and calculating the variance contribution rate of each index in the first index evaluation system according to the principal component model, and obtaining the accumulated variance contribution rate of each secondary evaluation index according to the variance contribution rate.

Specifically, the index data matrix of the first index evaluation system is

Further, the index data matrix of the first index evaluation system is unified to obtain a unified index data matrix Z ═ zi_j)_m×nThe elements are

Wherein,

further, a correlation matrix R is established, and its eigenvalues and eigenvectors are calculated. Wherein R ═ Z^TZ/(N-1), N is the dimension of the correlation matrix. Thus, the eigenvalue λ of the correlation matrix is found₁≥λ₂≥...≥λ_nAnd subjecting the corresponding feature vector to normalized orthogonal transformation to obtain e_i＝(α_i1,α_i2,...,α_in)^T。

The principal component model of the first index evaluation system is F_i＝α_i1A₁+α_i2A₂+...+α_inA_n。

Wherein A is_iIs the ith index and is the column vector of the matrix A; f_iIs the ith principal component;a_ikfor corresponding characteristic value of lambda_iThe kth component of the feature vector of (1).

Further, the variance contribution rate of the principal component to the original index data is calculated

Thereby, the characteristic value λ is set_iThe accumulated variance contribution rate is calculated according to the sequence from big to small

Preferably, a critical cumulative contribution ratio β is set_{Critical point of}When the variance contribution ratio beta (i) is accumulated>β_{Critical point of}Then, selecting principal components corresponding to the first k characteristic values to obtain the factor load of the ith index on the jth principal component

Finally, the factor loads are rotated by the coordinate axis, so that all the points fall in the first quadrant, and an ideal situation is achieved, namely, each variable has larger load on only one factor, and the load on other factors is smaller. Therefore, the multi-index system is converted into a main component system with lower dimensionality, so that the evaluation index system is simplified.

And S106, screening the main secondary evaluation indexes of the first index evaluation system according to the accumulated variance contribution rate of the secondary evaluation indexes.

In a preferred embodiment, step S106 includes:

sorting the cumulative variance contribution rate of the secondary evaluation indexes under each primary evaluation index from large to small, and screening the secondary evaluation indexes sorted to the front b bits as main secondary evaluation indexes; and b is not more than the total number of the secondary evaluation indexes under the corresponding primary evaluation indexes.

S107, reconstructing the first index evaluation system based on the main secondary indexes under each primary evaluation index to obtain a second index evaluation system, and judging the rationality of the second index evaluation system.

And S108, outputting the plan evaluation result when the second index evaluation system is detected to be reasonable.

Preferably, the judging the rationality of the second index evaluation system specifically includes:

constructing an index data matrix of the second index evaluation system;

calculating the information contribution rate of the main secondary evaluation indexes under each primary evaluation index to the secondary evaluation indexes according to the trace of a covariance matrix corresponding to the index data matrix of the first index evaluation system and the index data matrix of the second index evaluation system;

judging whether the information contribution rate exceeds a preset information contribution rate threshold value or not; if so, the second index evaluation system is considered to be reasonable, and the plan evaluation result is output; and if not, determining that the second index evaluation system is unreasonable, and assigning and standardizing each index in the second index evaluation system again.

It can be understood that the information contribution rate In of the selected index to the sea election index is trS_s/trS_h. Wherein, S is a covariance matrix corresponding to the index data matrix A, trS is a trace of the covariance matrix, and represents the sum of each index variance on the main diagonal of the covariance matrix; s is the number of indexes after screening, and h is the number of sea election indexes. The information contribution rate represents trS sum of variances of the filtered s indexes_strS sum of variances of h indexes in sea election_hThe ratio of (a) to (b) can also be understood as information of h hailing indexes reflected by the s screened indexes. A judgment standard of the reasonability of the index system construction is provided according to the idea of factor analysis, and the problem that the judgment of the reasonability of the index system construction lacks quantitative standards is solved. For example, if the original information of 95% or more is reflected by the sea election index of 30% or less, it is considered that the index evaluation system is reasonably constructed.

In a specific embodiment, on the basis of the above embodiments, the power emergency plan evaluation method is applied to an actual power grid. Specifically, 5 sets of power emergency plans for typhoon disasters were evaluated in a certain area (5 sets of power emergency plans are sequentially represented by Y1 to Y5). The 5 sets of power emergency plans are emphasized, the Y1 has the strongest capacity in the aspect of material guarantee, and the other aspects are relatively balanced; y2 has the most outstanding capability in two aspects of power grid dispatching and safety supervision; y3 has stronger emergency capacity in the aspect of power grid system architecture; y4 is not the strongest in the aspect of six emergency functions, but the overall capacity is balanced; y5 is particularly advantageous in terms of health maintenance and system architecture.

First, an index evaluation system is constructed. In the construction of an index evaluation system of the power emergency plan, six primary indexes are constructed first, and 12-15 secondary indexes are established under each primary index. For example, under the dispatching of the power grid, 13 secondary indexes are shared, namely the voltage grade conversion times, the electrical switch operation times, the recovered load order, the power constraint conditions, the number of permanent emergency command mechanisms, the emergency linkage departments, the response time after information receiving, the completeness of the basic environment of the command center, the capability of collecting multi-aspect information, the function perfection of auxiliary decision software, the coverage range of the dispatching of the central power grid, the vulnerability analysis of important power facilities, the upgrading research of an emergency system and the management of the emergency system.

Further, each index in the index evaluation system is assigned. Illustratively, the assignment method of the first-level index of the power grid scheduling of the emergency plan 1 is as follows:

a₁₁＝0.077×(1.0+1.0+0.9+0.9+0.9+0.8+0.8+0.9+1.0+1.0+0.8+0.8+0.9)＝0.9

the values in brackets are respectively the assignment sizes of secondary indexes (voltage grade conversion times, electrical switch operation times, recovered load orders, power constraint conditions and the like) under the primary indexes of the power grid dispatching, and 0.077 is the weight size of each secondary index, and the weight system such as the secondary indexes is adopted. Therefore, the assignment of the primary indexes of each power emergency plan is shown in table 1, and B1-B6 sequentially represent six primary indexes of power grid dispatching, state overhaul, communication guarantee, material guarantee, system architecture and safety supervision.

TABLE 1

Further, the evaluation index values of the index evaluation system obtained by normalizing the assignment of each index in the index evaluation system are shown in table 2.

TABLE 2

Further, the weight division model in step S103 is used to calculate the index weight value of the index evaluation system. Thus, the index weight value of each primary index is ω ═ 0.270,0.066,0.143,0.121,0.135, 0.265. Further, a weighting decision matrix shown in table 3 is obtained according to the index weight value of the primary index.

TABLE 3

The positive and negative ideal solutions are calculated according to the weighted decision matrix of table 3. A positive ideal solution can be obtained as:

(a″)⁺＝(0.270,0.066,0.143,0.121,0.135,0.265)；

negative ideal solution is (a')^-＝(0.135,0.033,0.072,0.061,0.068,0.133)。

Further, the euclidean distance and the relative closeness from each plan to the positive and negative ideal solutions are calculated, and plan sorting is performed according to the relative closeness, and the plan evaluation result of each plan is shown in table 4.

TABLE 4

Sorting according to the relative penetration degree of each power emergency plan, and obtaining Y2, Y1, Y5, Y4 and Y3, so that the quality sorting of the power emergency plans can be Y2, Y1, Y5, Y4 and Y3.

Further, the correlation factor and principal component of each index are calculated. And analyzing the secondary indexes under each primary index of all the schemes, and respectively calculating the correlation factor and the principal component of each scheme. In this embodiment, six calculations are needed, and only the second-level index corresponding to the first-level index is taken as an example to illustrate the main method of the present invention. The association degree and the principal component of the 13 secondary indexes under the power grid dispatching level are calculated, and the cumulative variance contribution rate of the secondary indexes shown in the table 5 can be obtained.

TABLE 5

If the critical cumulative variance contribution rate is set to 70% in advance, it can be seen from table 5 that the second-level indexes ranked as the first 4 are used as the main second-level indexes under the power grid dispatching. For other five primary indexes, the cumulative variance contribution rate can be obtained according to the method, and finally, the main secondary indexes under each primary index are selected to serve as the reconstructed evaluation index system. Therefore, the evaluation index system after reconstitution is shown in table 6.

TABLE 6

Further, the rationality of the reconstructed evaluation index system is determined. The reasonability of the reconstructed evaluation index system is judged, and the reasonability of the secondary indexes under the six primary indexes is judged respectively to obtain the information contribution rate under each primary index as shown in table 7:

TABLE 7

As can be seen from table 7 above, in the reconstructed evaluation index system, the new secondary indexes under each primary index all reflect more than 95% of the original secondary index information, so the reconstructed evaluation index system is reasonable.

The method and the device for evaluating the power emergency plan provided by the embodiment of the invention have the following beneficial effects:

1. the embodiment of the invention adopts the self-adaptive weight division model, can effectively avoid the influence of artificial subjective factors on weight division, can effectively simplify the calculation method of the index weight, can realize the correction of the index weight by adopting the variable weight model, can effectively avoid the phenomenon that the evaluation index is excessively deviated from a normal value and the final evaluation result is not reflected, can effectively improve the evaluation accuracy and reduce the error.

2. The embodiment of the invention can autonomously update and optimize the evaluation index system, remove redundant indexes in the index system by screening conditions, effectively improve the calculation speed of the evaluation result, realize the construction of a proper index system aiming at different electric power special plans, have stronger applicability and have more scientific and reliable evaluation results.

3. The embodiment of the invention fully considers all functional departments of the power emergency capacity, mainly comprises six main departments during power emergency, and the six main departments are dispatching, operation and inspection, communication, material, maintenance and safety supervision departments respectively, so that the preplan evaluation has higher reliability and integrity.

Example two

Referring to fig. 2, a schematic structural diagram of an electric power emergency plan evaluation device provided in the second embodiment of the present invention includes:

the index evaluation system construction module 201 is configured to obtain electrical quantity data of a power grid, and construct a first index evaluation system of an electric power emergency plan to be evaluated according to the electrical quantity data; the first index evaluation system comprises a plurality of first-level evaluation indexes and second-level evaluation indexes corresponding to the first-level evaluation indexes;

an index assignment module 202, configured to perform assignment and standardization on each index in the first index evaluation system to obtain an evaluation index value of the first index evaluation system;

the weight calculation module 203 is configured to calculate an index weight value of the first index evaluation system by using a preset weight division model;

the plan evaluation module 204 is configured to evaluate the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system, so as to obtain a plan evaluation result;

the factor analysis module 205 is configured to analyze a factor of the power emergency plan to be evaluated and a principal component factor of the first index evaluation system to obtain a cumulative variance contribution rate of each secondary evaluation index;

an index screening module 206, configured to screen a main secondary evaluation index of the first index evaluation system according to the cumulative variance contribution rate of the secondary evaluation index;

the index evaluation system reconstruction module 207 is used for reconstructing the first index evaluation system based on the main secondary indexes under each primary evaluation index to obtain a second index evaluation system and judging the rationality of the second index evaluation system;

and the evaluation result output module 208 is configured to output the plan evaluation result when the second index evaluation system is detected to be reasonable.

Preferably, the index assignment module 202 includes:

the secondary evaluation index assignment unit is used for assigning the voltage level conversion times in the secondary evaluation indexes by adopting a formula (1), assigning qualitative indexes in the secondary evaluation indexes by adopting a preset triangular fuzzy number three-scale method, and obtaining the assignment of the secondary evaluation indexes:

wherein y is the assignment of the voltage level conversion frequency index, and x is the voltage level conversion frequency;

the first-level evaluation index assignment unit is used for obtaining assignment of the first-level evaluation index according to assignment of the second-level evaluation index so as to obtain assignment of each index in the first index evaluation system;

the standardization processing unit is used for carrying out standardization processing on the assignment of each index in the first index evaluation system according to a formula (2) to obtain an evaluation index value of the first index evaluation system:

wherein, a'_ijThe evaluation index value of the jth index in the ith to-be-evaluated electric power emergency plan, a_ijAssigning a value to the jth index in the ith power emergency plan to be evaluated,

the minimum value of j index assignment in m electric power emergency plans to be evaluated is represented,

and j is 1,2, …, and n, i is 1,2, …, m.

Preferably, the weight calculating module 203 includes:

and the similarity calculation unit is used for obtaining the similarity between each index and the similar index in the ideal optimal solution according to a formula (3):

wherein d is_ijIs a stand forThe similarity, a_jAssigning a j index;

the adaptive weight division model building unit is used for building an adaptive weight division model according to the similarity between each index and the similar index in the ideal optimal solution; wherein the weight partitioning model comprises the adaptive weight partitioning model;

the adaptive weight partitioning model building unit is used for building an adaptive weight partitioning model according to the similarity between each index and the similar index in the ideal optimal solution and the formula (4):

wherein, ω is_jThe index weight value of each index of the first index evaluation system is obtained;

a constraint condition construction unit of the adaptive weight partitioning model, configured to obtain a constraint condition of the adaptive weight partitioning model according to formula (5):

wherein ξ_ijCalculating the gray correlation coefficient of the jth index in the ith power emergency plan to be evaluated and the j indexes in the ideal optimal solution according to a formula (6):

where ρ is a resolution coefficient, 0< ρ < 1.

Preferably, the weight calculating module 203 further comprises:

the variable weight model building unit is used for building a variable weight model by introducing a state variable weight factor based on the adaptive weight partitioning model when the index weight value of the first index evaluation system is detected to exceed a preset normal weight threshold value;

the variable weight calculation unit is used for obtaining the variable weight of the first index evaluation system according to a formula (7) by adopting the variable weight model:

wherein, ω is_eiIs the variable weight, ω_iAs an attribute weight value of each index, ω ═ ω (ω ═ ω)₁,ω₂,…,ω_n)，a_iFor the optimal solution of each index, S (a)_i) Obtaining the state variable weight factor of each index according to a formula (8):

where t is the penalty level, max (a)_i) Min (a) is the maximum value of the reasonable values of the indices_i) Is the minimum value of reasonable values of each index.

Preferably, the plan evaluation module 204 includes:

a weighting decision matrix construction unit, configured to construct a weighting decision matrix according to the evaluation index value of the first index evaluation system and the index weight value;

the positive ideal solution and negative ideal solution calculating unit is used for calculating a positive ideal solution and a negative ideal solution of the power emergency plan to be evaluated according to the weighting decision matrix;

the Euclidean distance calculation unit is used for calculating a first Euclidean distance between the electric power emergency plan to be evaluated and the positive ideal solution and a second Euclidean distance between the electric power emergency plan to be evaluated and the negative ideal solution according to the positive ideal solution and the negative ideal solution;

and the plan evaluation result calculation unit is used for calculating the relative closeness of the power emergency plan to be evaluated according to the first Euclidean distance and the second Euclidean distance, and the relative closeness is used as a plan evaluation result of the power emergency plan to be evaluated.

Preferably, the plan evaluation module 204 further includes:

a weighted decision matrix element calculation unit, configured to obtain an element of the weighted decision matrix according to formula (9):

a_ij″＝ω_j×a_ij′(i＝1,2,…,m,j＝1,2,…,n) (9)

wherein, a_ij"is an element of the weighted decision matrix;

the positive ideal solution calculating unit is used for obtaining a positive ideal solution of the power emergency plan to be evaluated according to a formula (10):

wherein, (a')⁺Is the positive ideal solution;

the negative ideal solution calculating unit is used for obtaining a negative ideal solution of the power emergency plan to be evaluated according to a formula (11):

wherein, (a')^-Is the negative ideal solution;

the first Euclidean distance calculating unit is used for obtaining a first Euclidean distance between the power emergency plan to be evaluated and the ideal solution according to a formula (12):

wherein,

the first Euclidean distance, a ', from the ith to-be-evaluated power emergency plan to the positive ideal solution'_iEmergency prediction for ith power to be evaluatedEvaluation index values of each index in the case;

the second Euclidean distance calculating unit is used for obtaining a second Euclidean distance between the power emergency plan to be evaluated and the negative ideal solution according to a formula (13):

wherein,

a second Euclidean distance from the ith power emergency plan to be evaluated to the negative ideal solution;

the relative closeness calculation unit is used for obtaining the relative closeness of the power emergency plan to be evaluated according to a formula (14):

wherein, C_iAnd the relative closeness of the ith power emergency plan to be evaluated is obtained.

Preferably, the factor analyzing module 205 includes:

the index data matrix construction unit of the first index evaluation system is used for obtaining an index data matrix of the first index evaluation system according to the evaluation index value of the first index evaluation system, and performing unification processing on the index data matrix of the first index evaluation system to obtain a unified index data matrix;

the correlation matrix calculation unit is used for establishing a correlation matrix according to the unified index data matrix and calculating the eigenvalue and the eigenvector of the correlation matrix;

the orthogonal transformation unit is used for carrying out standardized orthogonal transformation on the eigenvalue and the eigenvector of the correlation matrix to obtain an orthogonal matrix;

the principal component model construction unit is used for obtaining a principal component model of the first index evaluation system according to the index data matrix and the orthogonal matrix;

and the accumulated variance contribution rate calculating unit is used for calculating the variance contribution rate of each index in the first index evaluation system according to the principal component model and obtaining the accumulated variance contribution rate of each secondary evaluation index according to the variance contribution rate.

Preferably, the index screening module 206 includes:

the accumulated variance contribution rate sorting unit is used for sorting the accumulated variance contribution rates of the secondary evaluation indexes under each primary evaluation index from large to small, and screening the secondary evaluation indexes sorted to the front b bits as main secondary evaluation indexes; and b is not more than the total number of the secondary evaluation indexes under the corresponding primary evaluation indexes.

Preferably, the index evaluation system reconstruction module 207 includes:

the index data matrix construction unit of the second index evaluation system is used for constructing an index data matrix of the second index evaluation system;

the information contribution rate calculating unit is used for calculating the information contribution rate of the main secondary evaluation indexes under each primary evaluation index to the secondary evaluation indexes according to the trace of a covariance matrix corresponding to the index data matrix of the first index evaluation system and the index data matrix of the second index evaluation system;

the reasonableness judging unit is used for judging whether the information contribution rate exceeds a preset information contribution rate threshold value or not; if so, the second index evaluation system is considered to be reasonable, and the plan evaluation result is output; and if not, determining that the second index evaluation system is unreasonable, and assigning and standardizing each index in the second index evaluation system again.

The electric power emergency plan evaluation device provided in the second embodiment is used to execute the steps of the electric power emergency plan evaluation method in any one of the above embodiments, and the working principles and beneficial effects of the two are in one-to-one correspondence, so that details are not repeated.

It should be noted that the above-described device embodiments are merely illustrative, where the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.

Claims (7)

1. A power emergency plan evaluation method is characterized by comprising the following steps:

acquiring electrical quantity data of a power grid, and constructing a first index evaluation system of an electric power emergency plan to be evaluated according to the electrical quantity data; the first index evaluation system comprises a plurality of first-level evaluation indexes and second-level evaluation indexes corresponding to the first-level evaluation indexes;

assigning and standardizing each index in the first index evaluation system to obtain an evaluation index value of the first index evaluation system;

calculating an index weight value of the first index evaluation system by adopting a preset weight division model;

evaluating the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result;

analyzing factors of the power emergency plan to be evaluated and principal component factors of the first index evaluation system to obtain the cumulative variance contribution rate of each secondary evaluation index;

screening main secondary evaluation indexes of the first index evaluation system according to the cumulative variance contribution rate of the secondary evaluation indexes;

reconstructing the first index evaluation system based on the main secondary indexes under each primary evaluation index to obtain a second index evaluation system, and judging the rationality of the second index evaluation system;

when the second index evaluation system is detected to be reasonable, outputting the evaluation result of the plan;

the weight classification model comprises an adaptive weight classification model, and the adaptive weight classification model is constructed by the following steps:

and (3) obtaining the similarity between each index and the similar index in the ideal optimal solution according to the formula:

wherein d is_ijTo the degree of similarity, a_jAssigning a j index;

and (3) constructing an adaptive weight division model according to the similarity between each index and the similar index in the ideal optimal solution and a formula (4):

wherein, ω is_jThe index weight value of each index of the first index evaluation system is obtained;

obtaining the constraint condition of the self-adaptive weight partitioning model according to a formula (5):

wherein ξ_ijCalculating a gray correlation coefficient for the jth index in the ith to-be-evaluated electric power emergency plan and the j-class index in the ideal optimal solution according to a formula (6):

d(a_ij,a_0j)＝|a_ij-a_j| (6)

wherein rho is a resolution coefficient, 0< rho < 1;

the weight division model further comprises a variable weight model, and the variable weight model is constructed by the following steps:

when the index weight value of the first index evaluation system is detected to exceed a preset normal weight threshold value, introducing a state variable weight factor based on the self-adaptive weight division model, and constructing the variable weight model;

obtaining the variable weight of the first index evaluation system according to a formula (7) by adopting the variable weight model:

wherein, ω is_eiIs the variable weight, ω_iAs an attribute weight value of each index, ω ═ ω (ω ═ ω)₁,ω₂,…,ω_n)，a_iFor the optimal solution of each index, S (a)_i) Obtaining the state variable weight factor of each index according to a formula (8):

where t is the penalty level, max (a)_i) Min (a) is the maximum value of the reasonable values of the indices_i) Is the minimum value of the reasonable values of all indexes;

the screening of the main secondary evaluation indexes of the first index evaluation system according to the cumulative variance contribution rate of the secondary evaluation indexes specifically comprises the following steps:

sorting the cumulative variance contribution rate of the secondary evaluation indexes under each primary evaluation index from large to small, and screening the secondary evaluation indexes sorted to the front b bits as main secondary evaluation indexes; and b is not more than the total number of the secondary evaluation indexes under the corresponding primary evaluation indexes.

2. The method for evaluating the power emergency plan according to claim 1, wherein the assigning and standardizing the indexes in the first index evaluation system to obtain the evaluation index values of the first index evaluation system specifically comprises:

assigning the voltage grade conversion times in the secondary evaluation indexes by adopting a formula (1), and assigning qualitative indexes in the secondary evaluation indexes by adopting a preset triangular fuzzy number three-scale method to obtain the assignment of the secondary evaluation indexes:

wherein y is the assignment of the voltage level conversion frequency index, and x is the voltage level conversion frequency;

obtaining the assignment of the primary evaluation index according to the assignment of the secondary evaluation index, thereby obtaining the assignment of each index in the first index evaluation system;

carrying out standardization processing on the assignment of each index in the first index evaluation system according to a formula (2) to obtain an evaluation index value of the first index evaluation system:

wherein, a'_ijIn the ith electric power emergency plan to be evaluatedEvaluation index value of jth index, a_ijAssigning a value to the jth index in the ith power emergency plan to be evaluated,

the minimum value of j index assignment in m electric power emergency plans to be evaluated is represented,

and j is 1,2, …, and n, i is 1,2, …, m.

3. The method for evaluating the power emergency plan according to claim 1, wherein the evaluating the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result specifically comprises:

constructing a weighting decision matrix according to the evaluation index value of the first index evaluation system and the index weight value;

calculating a positive ideal solution and a negative ideal solution of the power emergency plan to be evaluated according to the weighting decision matrix;

calculating a first Euclidean distance between the electric power emergency plan to be evaluated and the positive ideal solution and a second Euclidean distance between the electric power emergency plan to be evaluated and the negative ideal solution according to the positive ideal solution and the negative ideal solution;

and calculating the relative closeness of the power emergency plan to be evaluated according to the first Euclidean distance and the second Euclidean distance, and taking the relative closeness as a plan evaluation result of the power emergency plan to be evaluated.

4. The power emergency protocol evaluation method according to claim 3, further comprising:

obtaining the elements of the weighted decision matrix according to equation (9):

a_ij″＝ω_j×a_ij′(i＝1,2,…,m,j＝1,2,…,n) (9)

wherein, a_ij"is an element of the weighted decision matrix;

obtaining a positive ideal solution of the power emergency plan to be evaluated according to a formula (10):

wherein, (a')⁺Is the positive ideal solution;

obtaining a negative ideal solution of the power emergency plan to be evaluated according to a formula (11):

wherein, (a')^-Is the negative ideal solution;

obtaining a first Euclidean distance between the power emergency plan to be evaluated and the positive ideal solution according to a formula (12):

wherein,

the first Euclidean distance, a ', from the ith to-be-evaluated power emergency plan to the positive ideal solution'_iEvaluating index values of all indexes in the ith to-be-evaluated electric power emergency plan;

obtaining a second Euclidean distance between the power emergency plan to be evaluated and the negative ideal solution according to a formula (13):

wherein,

a second Euclidean distance from the ith power emergency plan to be evaluated to the negative ideal solution;

obtaining the relative closeness of the power emergency plan to be evaluated according to a formula (14):

wherein, C_iAnd the relative closeness of the ith power emergency plan to be evaluated.

5. The method for evaluating the power emergency plan according to claim 1, wherein the analyzing the factor of the power emergency plan to be evaluated and the principal component factor of the first index evaluation system to obtain the cumulative variance contribution rate of each secondary evaluation index specifically comprises:

obtaining an index data matrix of the first index evaluation system according to the evaluation index value of the first index evaluation system, and performing unification processing on the index data matrix of the first index evaluation system to obtain a unified index data matrix;

establishing a correlation matrix according to the unified index data matrix, and calculating a characteristic value and a characteristic vector of the correlation matrix;

carrying out standardized orthogonal transformation on the eigenvalue and the eigenvector of the correlation matrix to obtain an orthogonal matrix;

obtaining a principal component model of the first index evaluation system according to the index data matrix and the orthogonal matrix;

and calculating the variance contribution rate of each index in the first index evaluation system according to the principal component model, and obtaining the accumulated variance contribution rate of each secondary evaluation index according to the variance contribution rate.

6. The method for evaluating the power emergency plan according to claim 1, wherein the judging the rationality of the second index evaluation system specifically comprises:

constructing an index data matrix of the second index evaluation system;

calculating the information contribution rate of the main secondary evaluation indexes under each primary evaluation index to the secondary evaluation indexes according to the trace of a covariance matrix corresponding to the index data matrix of the first index evaluation system and the index data matrix of the second index evaluation system;

judging whether the information contribution rate exceeds a preset information contribution rate threshold value or not; if so, the second index evaluation system is considered to be reasonable, and the plan evaluation result is output; and if not, determining that the second index evaluation system is unreasonable, and assigning and standardizing each index in the second index evaluation system again.

7. An electric power emergency plan evaluation device, characterized by comprising:

the index evaluation system construction module is used for acquiring the electric quantity data of the power grid and constructing a first index evaluation system of the power emergency plan to be evaluated according to the electric quantity data; the first index evaluation system comprises a plurality of first-level evaluation indexes and second-level evaluation indexes corresponding to the first-level evaluation indexes;

the index assignment module is used for assigning and standardizing each index in the first index evaluation system to obtain an evaluation index value of the first index evaluation system;

the weight calculation module is used for calculating an index weight value of the first index evaluation system by adopting a preset weight division model;

the plan evaluation module is used for evaluating the power emergency plan to be evaluated according to the evaluation index value and the index weight value of the first index evaluation system to obtain a plan evaluation result;

the factor analysis module is used for analyzing the factors of the power emergency plan to be evaluated and the principal component factors of the first index evaluation system to obtain the cumulative variance contribution rate of each secondary evaluation index;

the index screening module is used for screening main secondary evaluation indexes of the first index evaluation system according to the accumulated variance contribution rate of the secondary evaluation indexes;

the index evaluation system reconstruction module is used for reconstructing the first index evaluation system based on the main secondary indexes under each primary evaluation index to obtain a second index evaluation system and judging the rationality of the second index evaluation system;

the evaluation result output module is used for outputting the plan evaluation result when the second index evaluation system is detected to be reasonable;

wherein the weight calculation module comprises:

and the similarity calculation unit is used for obtaining the similarity between each index and the similar index in the ideal optimal solution according to a formula (3):

wherein d is_ijTo the degree of similarity, a_jAssigning a j index;

the adaptive weight division model building unit is used for building an adaptive weight division model according to the similarity between each index and the similar index in the ideal optimal solution; wherein the weight compartmentalization model comprises the adaptive weight compartmentalization model;

the adaptive weight partitioning model building unit is used for building an adaptive weight partitioning model according to the similarity between each index and the similar index in the ideal optimal solution and a formula (4):

wherein, ω is_jIs the first fingerEvaluating the index weight value of each index of the system;

a constraint condition construction unit of the adaptive weight partitioning model, configured to obtain a constraint condition of the adaptive weight partitioning model according to formula (5):

wherein xi is_ijCalculating a gray correlation coefficient for the jth index in the ith to-be-evaluated electric power emergency plan and the j-class index in the ideal optimal solution according to a formula (6):

d(a_ij,a_0j)＝|a_ij-a_j| (6)

wherein rho is a resolution coefficient, 0< rho < 1;

the weight calculation module further comprises:

the variable weight model building unit is used for building a variable weight model by introducing a state variable weight factor based on the adaptive weight partitioning model when the index weight value of the first index evaluation system is detected to exceed a preset normal weight threshold value;

the variable weight calculation unit is used for obtaining the variable weight of the first index evaluation system according to a formula (7) by adopting the variable weight model:

wherein, ω is_eiIs the variable weight, ω_iAs an attribute weight value of each index, ω ═ ω (ω ═ ω)₁,ω₂,…,ω_n)，a_iFor the optimal solution of each index, S (a)_i) For each index state weight factor, according to the formula (8) Obtaining the state variable weight factor:

where t is the penalty level, max (a)_i) Min (a) is the maximum value of the reasonable values of the indices_i) Is the minimum value of the reasonable values of all indexes;

the index screening module comprises:

the accumulated variance contribution rate sorting unit is used for sorting the accumulated variance contribution rates of the secondary evaluation indexes under each primary evaluation index from large to small, and screening the secondary evaluation indexes sorted to the front b bits as main secondary evaluation indexes; and b is not more than the total number of the secondary evaluation indexes under the corresponding primary evaluation indexes.

Images