CN104021300A - Comprehensive assessment method based on effect of distribution type electrical connection on power distribution network - Google Patents

Abstract

The invention discloses a comprehensive assessment method based on the effect of a distribution type electrical connection on a power distribution network. The comprehensive assessment method is characterized in that a comprehensive assessment index system of the effect of the distribution type electrical connection on the power distribution network, three-level index values under different distribution type electrical connection schemes are worked out by analyzing the power distribution network including a distribution type power source, a standard decision matrix of an assessment index is established, the weight coefficients of a one-level index, a two-level index and a three-level index are respectively obtained, accordingly, the comprehensive weight coefficient is determined, the comprehensive assessment satisfaction degrees of different distribution type electrical connection schemes are finally worked out, and the scheme with the largest comprehensive assessment satisfaction degree serves as the optimal distribution type electrical connection scheme. The comprehensive assessment method avoids the insufficient data information amount probably occurring through the single utilization of the object analysis method and also avoids the subject randomness of direct utilization the subject analysis method, and complex links of detecting and judging the matrix consistency are reduced while the judging fuzziness of people is taken into consideration.

Description

A kind of comprehensive estimation method that access affects power distribution network based on distributed power source

Technical field

The present invention relates to distribution network planning evaluation areas, be specifically related to a kind of comprehensive estimation method that access affects power distribution network based on distributed power source.

Background technology

Along with environmental protection pressure become increasingly conspicuous and energy demand increasing, the development of distributed power source has become the focus that various circles of society pay close attention to.The access of distributed power source makes traditional power distribution network become a power distribution network containing many power supplys that spreads all over middle-size and small-size power supply and load, the raising of distributed power source permeability makes the stable state of power distribution network and the variation of dynamic perfromance generation matter containing distributed power source, and this can bring many impacts to the aspects such as economy, reliability and the quality of power supply of the power distribution network containing distributed power source.Along with the difference of access point and the access capacity of distributed power source, its impact bringing is also different.Therefore, assessment distributed power source access on the impact of power distribution network to promoting greatly developing and apply and thering is important theory and realistic meaning of distributed power source.

Distributed power source access is one to the impact of power distribution network and involves a wide range of knowledge and complicated problem, existing research is mostly from distributed power source access assessing in a certain respect power distribution network impact, on distributed power source, access is not carried out comprehensive assessment to power distribution network impact (comprising economy, reliability and the quality of power supply etc.), lack the comprehensive estimation method of distributed power source access on power distribution network impact simultaneously, in evaluation process, seldom consider people's judgement ambiguity, the structure of judgment matrix is more loaded down with trivial details, and determining of the weight coefficient of different evaluation indexes is also perfect not.

Summary of the invention

The present invention is for avoiding the existing weak point of above-mentioned prior art, a kind of comprehensive estimation method that access affects power distribution network based on distributed power source is provided, consider the impact of distributed power source access on power distribution network, by definite weight coefficient that the coefficient of variation of three grades of indexs and improved Triangular Fuzzy Number analytical hierarchy process are combined, avoid the data message quantity not sufficient of using merely objective analysis method to run into, also avoid using merely the subjectivity of subjective analysis method random, in the judgement ambiguity of considering people, reduce the conforming loaded down with trivial details link of test and judge matrix.

The present invention is that technical solution problem adopts following technical scheme:

The present invention is based on distributed power source access is to carry out as follows on the feature of the comprehensive estimation method of power distribution network impact:

Step 1, set up the distributed power source access comprehensive assessment index system on power distribution network impact:

According to the membership between the character of each evaluation index and each evaluation index, evaluation index is divided into different levels and group as follows: basic evaluation index is designated as to three grades of indexs, and to incorporate into by attribute be separately six two-level index, described six two-level index are respectively cost of investment, benefit, system reliability, load reliability, busbar voltage and load point voltage; It is three first class index that described six two-level index are incorporated into by attribute separately, and described three first class index are respectively economy, reliability and the quality of power supply, obtains the comprehensive assessment index system with four levels; That described four levels are respectively is top, the second layer, the 3rd layer and the bottom, wherein:

Top is the comprehensive assessment satisfaction of distributed power source access on power distribution network impact; The second layer is first class index; The 3rd layer is two-level index; The bottom is three grades of indexs;

Step 2, by analyzing containing the distribution network of distributed power source, calculate three grades of index value x under different distributions formula power supply access scheme _ij, i=1 wherein, 2 ..., n, n is total number of three grades of indexs; J=1,2 ..., m, m is total number of distributed power source access scheme;

Step 3, set up as follows the standardization decision matrix of evaluation index:

1. according to three grades of index value x under different distributions formula power supply access scheme _ij, by formula (1) and formula (2), draw respectively the arithmetic mean x of each three grades of indexs _iwith standard deviation s _i,

$x_i=\frac{\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{x}_{\mathrm{ij}}}{m}---\left(1\right)$

$s_i=\sqrt{\frac{1}{m}\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{(x_{\mathrm{ij}}-x_i)}^2}---\left(2\right)$

2. by formula (3), obtain three grades of index value z of standardization _ijfor:

$z_{\mathrm{ij}}=\frac{(x_{\mathrm{ij}}-x_i)}{s_i}---\left(3\right)$

3. the sign before reverse index is exchanged, have standardization decision matrix Z=(z _ij) _{n * m};

Step 4, obtain as follows the weight coefficient of three grades of indexs:

By formula (4), obtained the coefficient of variation V of i three grades of indexs _i:

$V_i=\frac{s_i}{x_i}---\left(4\right)$

Have: the weight coefficient ω of i three grades of indexs _ifor

${\mathrm{\ω}}_i=\frac{V_i}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{V}_i}---\left(5\right)$

Step 5, obtain respectively as follows the weight coefficient of each first class index and two-level index:

1. adopt analytical hierarchy process structure consistance judgment matrix

N index is labeled as respectively to a by significance level ₁, a ₂... a _n, and be a according to the sequence of significance level ₁>=a ₂>=... a _n, β index and β+1 index are compared, corresponding scale value is designated as t _β, β=1 wherein, 2 ... N-1, then calculates other element value in judgment matrix according to the transitivity of index significance level, thereby constructs consistance judgment matrix R=(r _sv) _{n * N}, wherein, r _svfor judgment matrix element, s=1,2 ... N, v=1,2 ... N;

2. by consistance judgment matrix element r _svbe expressed as Triangular Fuzzy Number form and obtain just reciprocal Triangular Fuzzy Number consistance judgment matrix, consistance judgment matrix element r _svcorresponding to wherein: the value of the possibility maximum of Triangular Fuzzy Number $r_{\mathrm{sv}}^m=r_{\mathrm{sv}};$ The lower limit of Triangular Fuzzy Number $r_{\mathrm{sv}}^l=r_{\mathrm{sv}}^m-0.5;$ The higher limit of Triangular Fuzzy Number $r_{\mathrm{sv}}^u=r_{\mathrm{sv}}^m+0.5;$

3. build respectively lower limit matrix A ^l, possibility maximum value matrix A ^mwith higher limit matrix A ^ufor:

$A^l={\left(r_{\mathrm{sv}}^l\right)}_{N\×N};A^m={\left(r_{\mathrm{sv}}^m\right)}_{N\×N};A^u={\left(r_{\mathrm{sv}}^u\right)}_{N\×N};$

4. calculate according to the following procedure weight coefficient:

Calculate respectively lower limit matrix A ^l, possibility maximum value matrix A ^mwith higher limit matrix A ^ulower limit normalization proper vector y corresponding to eigenvalue of maximum ^l, possibility maximum value normalization proper vector y ^m, higher limit normalization proper vector y ^u;

Wherein, $y^l=\left(\begin{array}{c}{y}_1^l\\ y_2^l\\ .\\ .\\ .\\ y_N^l\end{array}\right),y^m=\left(\begin{array}{c}{y}_1^m\\ y_2^m\\ .\\ .\\ .\\ y_N^m\end{array}\right),y^u=\left(\begin{array}{c}{y}_1^u\\ y_2^u\\ .\\ .\\ .\\ y_N^u\end{array}\right)$

Calculate respectively intermediate variable k, h, g;

$k=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\Σ}}}{r}_{\mathrm{sv}}^u}},h=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\Σ}}}{r}_{\mathrm{sv}}^m}},g=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\Σ}}}{r}_{\mathrm{sv}}^l}}$

The Triangular Fuzzy Number weight of s index $w_s=(w_s^l,w_s^m,w_s^u),$ And have: $w_s^l=k{y}_s^l;w_s^m=h{y}_s^m;$ wherein, be the lower limit of the Triangular Fuzzy Number weight of s index, be the value of possibility maximum of the Triangular Fuzzy Number weight of s index, be the higher limit of the Triangular Fuzzy Number weight of s index, for s the element that index is corresponding in lower limit normalization proper vector, for s the element that index is corresponding in the value normalization proper vector of possibility maximum, for s the element that index is corresponding in higher limit normalization proper vector;

Determine the weight coefficient of each index

$w_s^{*}=\frac{1}{2}[(1-\mathrm{\α})w_s^l+w_s^m+\mathrm{\α}{w}_s^u],\mathrm{\α}\∈\left[\mathrm{0,1}\right]---\left(6\right)$

In formula (6), α is risk factor;

When a described N index is taken as first class index, weight coefficient be first class index weight coefficient, be designated as λ _p, p=1 wherein, 2 ... P, P is first class index number;

When a described N index is taken as two-level index, weight coefficient be two-level index weight coefficient, be designated as η _q, q=1 wherein, 2 ... Q, Q is two-level index number;

Step 6, by formula (7), determine comprehensive weight coefficient W _i ^*:

${W_i}^{*}=\frac{\underset{p=1}{\overset{P}{\mathrm{\Σ}}}\underset{q=1}{\overset{Q}{\mathrm{\Σ}}}\mathrm{\ζ}{\mathrm{\ω}}_i{\mathrm{\η}}_q{\mathrm{\λ}}_p}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{p=1}{\overset{P}{\mathrm{\Σ}}}\underset{q=1}{\overset{Q}{\mathrm{\Σ}}}\mathrm{\ζ}{\mathrm{\ω}}_i{\mathrm{\η}}_q{\mathrm{\λ}}_p}---\left(7\right)$

Wherein, ζ is for considering coefficient, ζ=1 when i three grades of indexs are under the jurisdiction of q two-level index and q two-level index and are under the jurisdiction of p first class index, otherwise ζ=0, W _i ^*it is the comprehensive weight coefficient of i three grades of indexs;

The calculation expression of step 7, comprehensive assessment satisfaction is suc as formula (8):

$S_j=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{W_i}^{*}{z}_{\mathrm{ij}}---\left(8\right)$

Wherein, S _jit is the comprehensive assessment satisfaction of j distributed power source access scheme;

According to formula (8), calculate respectively the comprehensive assessment satisfaction of m distributed power source access scheme, wherein the scheme of comprehensive assessment Maximum Satisfaction is best distribution formula power supply access scheme.

Compared with the prior art, beneficial effect of the present invention is embodied in:

The present invention in conjunction with distributed power source access power distribution network after operation characteristic, analyze the character of different evaluation indexes and the membership between different evaluation index, set up perfect comprehensive assessment index system, the impact of reaction profile formula power supply access on power distribution network more all sidedly; By the coefficient of variation of three grades of indexs and improved Triangular Fuzzy Number analytical hierarchy process definite weight coefficient that combines, the data message quantity not sufficient of having avoided simple use objective analysis method to run into, also avoid the subjectivity of simple use subjective analysis method random, in the judgement ambiguity of considering people, reduced the conforming loaded down with trivial details link of test and judge matrix; By setting up the hierarchical structure of comprehensive assessment index system, can calculate the comprehensive assessment satisfaction of distributed power source access on power distribution network impact, assessment result is more scientific, practical.

Accompanying drawing explanation

Fig. 1 is the comprehensive estimation method process flow diagram of distributed power source access on power distribution network impact.

Embodiment

Referring to Fig. 1, in the present embodiment, based on distributed power source access, on the comprehensive estimation method of power distribution network impact, be to carry out as follows:

Step 1, set up the distributed power source access comprehensive assessment index system on power distribution network impact:

According to the membership between the character of each evaluation index and each evaluation index, evaluation index is divided into different levels and group as follows: basic evaluation index is designated as to three grades of indexs, and to incorporate into by attribute be separately six two-level index, described six two-level index are respectively cost of investment, benefit, system reliability, load reliability, busbar voltage and load point voltage; It is three first class index that described six two-level index are incorporated into by attribute separately, and described three first class index are respectively economy, reliability and the quality of power supply, obtains the comprehensive assessment index system with four levels; That described four levels are respectively is top, the second layer, the 3rd layer and the bottom, wherein:

Top is the comprehensive assessment satisfaction of distributed power source access on power distribution network impact; The second layer is first class index; The 3rd layer is two-level index; The bottom is three grades of indexs;

In the process of comprehensive assessment index system grouping, evaluation index attribute is on the same group identical, the same index that is under the jurisdiction of last layer, evaluation index attribute is not different on the same group, guaranteeing to guarantee that the problem of different evaluation index reflections is relatively independent when the access of the comprehensive reaction profile formula of evaluation index system power supply is on power distribution network impact, can not have too much overlappingly, otherwise just in a disguised form increase the weight of this evaluation index.

Step 2, by analyzing containing the distribution network of distributed power source, calculate three grades of index value x under different distributions formula power supply access scheme _ij, i=1 wherein, 2 ..., n, n is total number of three grades of indexs; J=1,2 ..., m, m is total number of distributed power source access scheme;

By the contents such as the calculating of distribution network trend, reliability assessment, economic evaluation and the quality of power supply containing distributed power source are analyzed, calculate three grades of index values under different distributions formula power supply access scheme.

Step 3, set up as follows the standardization decision matrix of evaluation index:

1. according to three grades of index value x under different distributions formula power supply access scheme _ij, by formula (1) and formula (2), draw respectively the arithmetic mean x of each three grades of indexs _iwith standard deviation s _i,

$x_i=\frac{\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{x}_{\mathrm{ij}}}{m}---\left(1\right)$

$s_i=\sqrt{\frac{1}{m}\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{(x_{\mathrm{ij}}-x_i)}^2}---\left(2\right)$

2. by formula (3), obtain three grades of index value z of standardization _ijfor:

$z_{\mathrm{ij}}=\frac{(x_{\mathrm{ij}}-x_i)}{s_i}---\left(3\right)$

2. the sign before reverse index is exchanged, have standardization decision matrix Z=(z _ij) _{n * m};

Because the dimension of different evaluation indexes is different, so carried out standardization and formed standardization decision matrix by arithmetic mean and the standard deviation of each three grades of indexs, the impact that can remove different dimensions, makes different evaluation indexes have comparability.Because evaluation index has just contrary dividing, for reverse index, the sign before it is exchanged.By the desired value after above-mentioned standardization, be centered around 0 fluctuation up and down, be greater than 0 explanation higher than average level, be less than 0 explanation lower than average level.

Step 4, obtain as follows the weight coefficient of three grades of indexs:

By formula (4), obtained the coefficient of variation V of i three grades of indexs _i:

$V_i=\frac{s_i}{x_i}---\left(4\right)$

Have: the weight coefficient ω of i three grades of indexs _ifor

${\mathrm{\ω}}_i=\frac{V_i}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{V}_i}---\left(5\right)$

In evaluation index system, the index that index value difference is larger, the index that is namely more difficult to realization, such index more can reflect the gap of different distributed power source access schemes.Because the coefficient of variation of evaluation index is to try to achieve by the arithmetic mean of evaluation index and standard deviation, can embody the difference degree of index value, therefore the information of directly having utilized indices to comprise by the definite weight coefficient of the coefficient of variation is a kind of objective tax power method.

Step 5, obtain respectively as follows the weight coefficient of each first class index and two-level index:

1. adopt analytical hierarchy process structure consistance judgment matrix

N index is labeled as respectively to a by significance level ₁, a ₂... a _n, and be a according to the sequence of significance level ₁>=a ₂>=... a _n, β index and β+1 index are compared, corresponding scale value is designated as t _β, β=1 wherein, 2 ... N-1, then calculates other element value in judgment matrix according to the transitivity of index significance level, thereby constructs consistance judgment matrix R=(r _sv) _{n * N}, wherein, r _svfor judgment matrix element, s=1,2 ... N, v=1,2 ... N;

Method to traditional Judgement Matricies is improved, and the judgment matrix after improvement has strict consistance, has avoided loaded down with trivial details consistency check, has reduced workload.When Judgement Matricies, adopt scaling law as shown in table 1.

Table 1 importance scale implication table

2. by consistance judgment matrix element r _svbe expressed as Triangular Fuzzy Number form and obtain just reciprocal Triangular Fuzzy Number consistance judgment matrix, consistance judgment matrix element r _svcorresponding to wherein: the value of the possibility maximum of Triangular Fuzzy Number $r_{\mathrm{sv}}^m=r_{\mathrm{sv}};$ The lower limit of Triangular Fuzzy Number $r_{\mathrm{sv}}^l=r_{\mathrm{sv}}^m-0.5;$ The higher limit of Triangular Fuzzy Number $r_{\mathrm{sv}}^u=r_{\mathrm{sv}}^m+0.5;$

Improved just reciprocal Triangular Fuzzy Number consistance judgment matrix has been considered people's judgement ambiguity, has avoided the limitation of traditional heuristic methods, introduces uncertain information, can take into full account numerous factors.

3. build respectively lower limit matrix A ^l, possibility maximum value matrix A ^mwith higher limit matrix A ^ufor:

$A^l={\left(r_{\mathrm{sv}}^l\right)}_{N\×N};A^m={\left(r_{\mathrm{sv}}^m\right)}_{N\×N};A^u={\left(r_{\mathrm{sv}}^u\right)}_{N\×N};$

3. calculate according to the following procedure weight coefficient:

Adopt power method to calculate respectively lower limit matrix A ^l, possibility maximum value matrix A ^mwith higher limit matrix A ^ulower limit normalization proper vector y corresponding to eigenvalue of maximum ^l, possibility maximum value normalization proper vector y ^m, higher limit normalization proper vector y ^u;

Wherein, $y^l=\left(\begin{array}{c}{y}_1^l\\ y_2^l\\ .\\ .\\ .\\ y_N^l\end{array}\right),y^m=\left(\begin{array}{c}{y}_1^m\\ y_2^m\\ .\\ .\\ .\\ y_N^m\end{array}\right),y^u=\left(\begin{array}{c}{y}_1^u\\ y_2^u\\ .\\ .\\ .\\ y_N^u\end{array}\right)$

Calculate respectively intermediate variable k, h, g;

$k=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\Σ}}}{r}_{\mathrm{sv}}^u}},h=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\Σ}}}{r}_{\mathrm{sv}}^m}},g=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\Σ}}}{r}_{\mathrm{sv}}^l}}$

The Triangular Fuzzy Number weight of s index $w_s=(w_s^l,w_s^m,w_s^u),$ And have: $w_s^l=k{y}_s^l;w_s^m=h{y}_s^m;$ wherein, be the lower limit of the Triangular Fuzzy Number weight of s index, be the value of possibility maximum of the Triangular Fuzzy Number weight of s index, be the higher limit of the Triangular Fuzzy Number weight of s index, for s the element that index is corresponding in lower limit normalization proper vector, for s the element that index is corresponding in the value normalization proper vector of possibility maximum, for s the element that index is corresponding in higher limit normalization proper vector;

Determine the weight coefficient of each index

$w_s^{*}=\frac{1}{2}[(1-\mathrm{\α})w_s^l+w_s^m+\mathrm{\α}{w}_s^u],\mathrm{\α}\∈\left[\mathrm{0,1}\right]---\left(6\right)$

In formula (6), α is risk factor; The attitudes toward risk that has reflected decision maker.When α >0.5, claim that decision maker is Risk-Averse; When α=0.5, claim that decision maker is risk-neutral; When α <0.5, claim that decision maker is risk aversion.

When a described N index is taken as first class index, N value gets 3, the comprehensive assessment satisfaction now based on top distributed power source access, power distribution network being affected and economy, reliability and three first class index of the quality of power supply of the second layer, the just reciprocal Triangular Fuzzy Number consistance judgment matrix on 3 rank of structure, the weight coefficient of obtaining be first class index weight coefficient, be designated as λ _p, p=1 wherein, 2 ... P, P is first class index number, now P value gets 3;

When a described N index is taken as two-level index, N value gets 2, layer 2-based economic index and be subordinate to the cost of investment of the 3rd layer and the reliability index of two two-level index of benefit, the second layer and be subordinate to the system reliability of the 3rd layer and the power quality index of two two-level index of load reliability, the second layer and be subordinate to busbar voltage and two two-level index of load point voltage of the 3rd layer now, the just reciprocal Triangular Fuzzy Number consistance judgment matrix on 32 rank of structure, the weight coefficient of obtaining be two-level index weight coefficient, be designated as η _q, q=1 wherein, 2 ... Q, Q is two-level index number, now Q value gets 6.

Step 6, by formula (7), determine comprehensive weight coefficient W _i ^*:

${W_i}^{*}=\frac{\underset{p=1}{\overset{P}{\mathrm{\&Sigma;}}}\underset{q=1}{\overset{Q}{\mathrm{\&Sigma;}}}\mathrm{\&zeta;}{\mathrm{\&omega;}}_i{\mathrm{\&eta;}}_q{\mathrm{\&lambda;}}_p}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{p=1}{\overset{P}{\mathrm{\&Sigma;}}}\underset{q=1}{\overset{Q}{\mathrm{\&Sigma;}}}\mathrm{\&zeta;}{\mathrm{\&omega;}}_i{\mathrm{\&eta;}}_q{\mathrm{\&lambda;}}_p}---\left(7\right)$

Wherein, ζ is for considering coefficient, ζ=1 when i three grades of indexs are under the jurisdiction of q two-level index and q two-level index and are under the jurisdiction of p first class index, otherwise ζ=0, W _i ^*it is the comprehensive weight coefficient of i three grades of indexs;

By the definite comprehensive weight coefficient of this step, considered subjective and objective factor and people's judgement ambiguity, the data message quantity not sufficient of having avoided simple use objective analysis method to run into, has also avoided the subjectivity of simple use subjective analysis method random simultaneously.

The calculation expression of step 7, comprehensive assessment satisfaction is suc as formula (8):

$S_j=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{W_i}^{*}{z}_{\mathrm{ij}}---\left(8\right)$

Wherein, S _jit is the comprehensive assessment satisfaction of j distributed power source access scheme;

According to formula (8), calculate respectively the comprehensive assessment satisfaction of m distributed power source access scheme, wherein the scheme of comprehensive assessment Maximum Satisfaction is best distribution formula power supply access scheme.

Claims (1)

1. the comprehensive estimation method that access affects power distribution network based on distributed power source, is characterized in that: carry out as follows:

Step 1, set up the distributed power source access comprehensive assessment index system on power distribution network impact:

According to the membership between the character of each evaluation index and each evaluation index, evaluation index is divided into different levels and group as follows: basic evaluation index is designated as to three grades of indexs, and to incorporate into by attribute be separately six two-level index, described six two-level index are respectively cost of investment, benefit, system reliability, load reliability, busbar voltage and load point voltage; It is three first class index that described six two-level index are incorporated into by attribute separately, and described three first class index are respectively economy, reliability and the quality of power supply, obtains the comprehensive assessment index system with four levels; That described four levels are respectively is top, the second layer, the 3rd layer and the bottom, wherein:

Top is the comprehensive assessment satisfaction of distributed power source access on power distribution network impact; The second layer is first class index; The 3rd layer is two-level index; The bottom is three grades of indexs;

Step 2, by analyzing containing the distribution network of distributed power source, calculate three grades of index value x under different distributions formula power supply access scheme _ij, i=1 wherein, 2 ..., n, n is total number of three grades of indexs; J=1,2 ..., m, m is total number of distributed power source access scheme;

Step 3, set up as follows the standardization decision matrix of evaluation index:

1. according to three grades of index value x under different distributions formula power supply access scheme _ij, by formula (1) and formula (2), draw respectively the arithmetic mean x of each three grades of indexs _iwith standard deviation s _i,

$x_i=\frac{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{x}_{\mathrm{ij}}}{m}---\left(1\right)$

$s_i=\sqrt{\frac{1}{m}\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{(x_{\mathrm{ij}}-x_i)}^2}---\left(2\right)$

2. by formula (3), obtain three grades of index value z of standardization _ijfor:

$z_{\mathrm{ij}}=\frac{(x_{\mathrm{ij}}-x_i)}{s_i}---\left(3\right)$

3. the sign before reverse index is exchanged, have standardization decision matrix Z=(z _ij) _{n * m};

Step 4, obtain as follows the weight coefficient of three grades of indexs:

By formula (4), obtained the coefficient of variation V of i three grades of indexs _i:

$V_i=\frac{s_i}{x_i}---\left(4\right)$

Have: the weight coefficient ω of i three grades of indexs _ifor

${\mathrm{\&omega;}}_i=\frac{V_i}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{V}_i}---\left(5\right)$

Step 5, obtain respectively as follows the weight coefficient of each first class index and two-level index:

1. adopt analytical hierarchy process structure consistance judgment matrix

N index is labeled as respectively to a by significance level ₁, a ₂... a _n, and be a according to the sequence of significance level ₁>=a ₂>=... a _n, β index and β+1 index are compared, corresponding scale value is designated as t _β, β=1 wherein, 2 ... N-1, then calculates other element value in judgment matrix according to the transitivity of index significance level, thereby constructs consistance judgment matrix R=(r _sv) _{n * N}, wherein, r _svfor judgment matrix element, s=1,2 ... N, v=1,2 ... N;

2. by consistance judgment matrix element r _svbe expressed as Triangular Fuzzy Number form and obtain just reciprocal Triangular Fuzzy Number consistance judgment matrix, consistance judgment matrix element r _svcorresponding to wherein: the value of the possibility maximum of Triangular Fuzzy Number $r_{\mathrm{sv}}^m=r_{\mathrm{sv}};$ The lower limit of Triangular Fuzzy Number $r_{\mathrm{sv}}^l=r_{\mathrm{sv}}^m-0.5;$ The higher limit of Triangular Fuzzy Number $r_{\mathrm{sv}}^u=r_{\mathrm{sv}}^m+0.5;$

3. build respectively lower limit matrix A ^l, possibility maximum value matrix A ^mwith higher limit matrix A ^ufor:

$A^l={\left(r_{\mathrm{sv}}^l\right)}_{N\&times;N};A^m={\left(r_{\mathrm{sv}}^m\right)}_{N\&times;N};A^u={\left(r_{\mathrm{sv}}^u\right)}_{N\&times;N};$

4. calculate according to the following procedure weight coefficient:

Calculate respectively lower limit matrix A ^l, possibility maximum value matrix A ^mwith higher limit matrix A ^ulower limit normalization proper vector y corresponding to eigenvalue of maximum ^l, possibility maximum value normalization proper vector y ^m, higher limit normalization proper vector y ^u;

Wherein, $y^l=\left(\begin{array}{c}{y}_1^l\\ y_2^l\\ .\\ .\\ .\\ y_N^l\end{array}\right),y^m=\left(\begin{array}{c}{y}_1^m\\ y_2^m\\ .\\ .\\ .\\ y_N^m\end{array}\right),y^u=\left(\begin{array}{c}{y}_1^u\\ y_2^u\\ .\\ .\\ .\\ y_N^u\end{array}\right)$

Calculate respectively intermediate variable k, h, g;

$k=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\&Sigma;}}}{r}_{\mathrm{sv}}^u}},h=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\&Sigma;}}}{r}_{\mathrm{sv}}^m}},g=\sqrt{\underset{v=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{1}{\underset{s=1}{\overset{N}{\mathrm{\&Sigma;}}}{r}_{\mathrm{sv}}^l}}$

The Triangular Fuzzy Number weight of s index $w_s=(w_s^l,w_s^m,w_s^u),$ And have: $w_s^l=k{y}_s^l;w_s^m=h{y}_s^m;$ wherein, be the lower limit of the Triangular Fuzzy Number weight of s index, be the value of possibility maximum of the Triangular Fuzzy Number weight of s index, be the higher limit of the Triangular Fuzzy Number weight of s index, for s the element that index is corresponding in lower limit normalization proper vector, for s the element that index is corresponding in the value normalization proper vector of possibility maximum, for s the element that index is corresponding in higher limit normalization proper vector;

Determine the weight coefficient of each index

$w_s^{*}=\frac{1}{2}[(1-\mathrm{\&alpha;})w_s^l+w_s^m+\mathrm{\&alpha;}{w}_s^u],\mathrm{\&alpha;}\&Element;\left[\mathrm{0,1}\right]---\left(6\right)$

In formula (6), α is risk factor;

When a described N index is taken as first class index, weight coefficient be first class index weight coefficient, be designated as λ _p, p=1 wherein, 2 ... P, P is first class index number;

When a described N index is taken as two-level index, weight coefficient be two-level index weight coefficient, be designated as η _q, q=1 wherein, 2 ... Q, Q is two-level index number;

Step 6, by formula (7), determine comprehensive weight coefficient W _i ^*:

${W_i}^{*}=\frac{\underset{p=1}{\overset{P}{\mathrm{\&Sigma;}}}\underset{q=1}{\overset{Q}{\mathrm{\&Sigma;}}}\mathrm{\&zeta;}{\mathrm{\&omega;}}_i{\mathrm{\&eta;}}_q{\mathrm{\&lambda;}}_p}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{p=1}{\overset{P}{\mathrm{\&Sigma;}}}\underset{q=1}{\overset{Q}{\mathrm{\&Sigma;}}}\mathrm{\&zeta;}{\mathrm{\&omega;}}_i{\mathrm{\&eta;}}_q{\mathrm{\&lambda;}}_p}---\left(7\right)$

Wherein, ζ is for considering coefficient, ζ=1 when i three grades of indexs are under the jurisdiction of q two-level index and q two-level index and are under the jurisdiction of p first class index, otherwise ζ=0, W _i ^*it is the comprehensive weight coefficient of i three grades of indexs;

The calculation expression of step 7, comprehensive assessment satisfaction is suc as formula (8):

$S_j=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{W_i}^{*}{z}_{\mathrm{ij}}---\left(8\right)$

Wherein, S _jit is the comprehensive assessment satisfaction of j distributed power source access scheme;

According to formula (8), calculate respectively the comprehensive assessment satisfaction of m distributed power source access scheme, wherein the scheme of comprehensive assessment Maximum Satisfaction is best distribution formula power supply access scheme.