CN103324840A

Movatterモバイル変換

Info

Publication number: CN103324840A
Application number: CN2013102239116A
Authority: CN
Inventors: 潘天红; 张乙
Original assignee: Jiangsu University
Current assignee: Jiangsu University
Priority date: 2013-06-06
Filing date: 2013-06-06
Publication date: 2013-09-25

Abstract

本发明公开了一种电力需求侧用电质量综合评估方法，包括如下步骤：（1）完成电力需求侧配电网监测点的数据采集与传输；（2）对电能质量指标建立电能质量等级，并按着“效益型”指标与“成本型”指标对所建立的等级进行归一化处理；（3）在获取主观权重的基础上，利用贝叶斯方法对其进行修正，并以带约束的优化算法得到各指标的权重优化值；（4）借助于属性识别模型，对监测点的各项电能指标进行分级评估；（5）结合权重优化值，计算得到电能质量的综合评价等级，从而完成对电能质量的定量评估。本发明实现了电能各指标权重的合理计算，提高了电能综合评估结果的合理性与可靠性，能为电网的安全性提供可靠信息。

The invention discloses a method for comprehensive evaluation of power quality on the power demand side, which includes the following steps: (1) completing data collection and transmission of monitoring points of the power demand side distribution network; (2) establishing power quality levels for power quality indicators, And according to the "benefit type" index and "cost type" index, the established grades are normalized; (3) On the basis of obtaining the subjective weight, use the Bayesian method to correct it, and use the constraint (4) With the help of the attribute recognition model, the power index of the monitoring point is graded and evaluated; (5) Combined with the weight optimization value, the comprehensive evaluation level of power quality is calculated, so that Complete the quantitative assessment of power quality. The invention realizes the reasonable calculation of the weight of each index of electric energy, improves the rationality and reliability of the comprehensive evaluation result of electric energy, and can provide reliable information for the safety of the power grid.

Description

Translated fromChinese

一种电力需求侧用电质量综合评估方法A method for comprehensive evaluation of power consumption quality on the power demand side

技术领域technical field

本发明涉及一种电力需求侧用电质量综合评估方法，具体涉及一种基于贝叶斯与属性识别模型的电能综合评估方法，属于电气工程领域。The invention relates to a method for comprehensively evaluating power quality at the power demand side, in particular to a method for comprehensively evaluating electric energy based on a Bayesian and attribute recognition model, and belongs to the field of electrical engineering.

背景技术Background technique

电能作为当今社会不可或缺的能源，其质量的好坏关系到电网的安全运行和电力需求侧用户的用电水平。理想的电能质量是完美的正弦波，但由于电网中存在各种复杂因素，从而导致电能质量出现偏差，于是产生了电能质量问题。Electric energy is an indispensable energy source in today's society, and its quality is related to the safe operation of the power grid and the power consumption level of users on the power demand side. The ideal power quality is a perfect sine wave, but due to various complex factors in the power grid, the power quality is deviated, and power quality problems arise.

电能质量评估是根据电能的相关参数，依据一定的评估方法给出电能的质量等级，以供电力部门和用户查看。传统的电能质量评估主要集中在对单个指标进行评估，忽略了电能各指标之间的影响。2008年我国修改了已颁布的电能质量标准，该五项指标有供电电压偏差，电力系统频率偏差，三相电压不平衡度，电压波动和闪变，公用电网谐波。实际上，电能质量是一个综合性的概念，它是由多个指标共同组合而成，因此，单项指标评估已不能满足电能发展的需求，于是出现了电能的综合评估，其主要思想是求出各个指标的权重，从而利用权重把多个指标综合为一个整体，实现用户用电水平的综合评估。The power quality assessment is based on the relevant parameters of the power, and according to a certain evaluation method, the quality level of the power is given for the power department and users to check. The traditional power quality assessment mainly focuses on the evaluation of a single index, ignoring the influence between the various indexes of electric energy. In 2008, my country revised the promulgated power quality standards. The five indicators include power supply voltage deviation, power system frequency deviation, three-phase voltage unbalance, voltage fluctuation and flicker, and public grid harmonics. In fact, power quality is a comprehensive concept, which is composed of multiple indicators. Therefore, the evaluation of a single indicator can no longer meet the needs of power development, so the comprehensive evaluation of power appears. The main idea is to find out The weight of each indicator can be used to integrate multiple indicators into a whole to achieve a comprehensive evaluation of the user's electricity consumption level.

专利“一种电能质量综合评估的方法”[申请号：CN201110445808.7，公开号：CN101750561A]，公开了一种运用灰色关联系数矩阵的TOPSIS法对电能质量综合评估的方法,通过AHP法与熵权法确定主客观权重，利用TOPSIS法的决策矩阵，最终由贴近度获得电能质量等级。该方法存在不同程度的人为主观因素，计算相对较复杂。The patent "a method for comprehensive evaluation of power quality" [application number: CN201110445808.7, publication number: CN101750561A] discloses a method for comprehensive evaluation of power quality using the TOPSIS method of the gray correlation coefficient matrix, through the AHP method and entropy The weight method determines the subjective and objective weights, uses the decision matrix of the TOPSIS method, and finally obtains the power quality level from the degree of closeness. There are different degrees of human subjective factors in this method, and the calculation is relatively complicated.

专利“电能质量综合评估归一化处理方法”[专利申请号：CN201110051283.9，公开号：CN102339355A]，公开了一种电能质量综合评估归一化处理的方法,将实际测量与计算的电能各项单项指标进行归一化处理,得到各项评估系数,然后与权值计算,最后结合电能质量得分情况对照表,判定电能质量的总体情况。该方法也存在不同程度的人为主观因素，评估结果具有一定的不确定性。The patent "normalized processing method for comprehensive evaluation of power quality" [patent application number: CN201110051283.9, publication number: CN102339355A] discloses a normalized processing method for comprehensive evaluation of power quality. The individual indicators are normalized to obtain the evaluation coefficients, and then calculated with the weight, and finally combined with the power quality score comparison table to determine the overall situation of the power quality. This method also has different degrees of human subjective factors, and the evaluation results have certain uncertainties.

本发明针对上述不足，提出一种结合贝叶斯算法与属性识别的电力需求侧的电能评估方法，利用贝叶斯赋权法修正主观权重，进而得到电能质量各等级下的权重值，并利用带约束的优化算法，得到权重的优化值，实现权重的主客观统一，避免了权重确定时受人为因素影响过大，从而使得评估的结果客观、可信。再借助于属性识别模型，实现了最终的电能质量等级评定，以更好满足现代电能质量综合评估的需求。Aiming at the above-mentioned deficiencies, the present invention proposes a power demand-side electric energy evaluation method combining Bayesian algorithm and attribute identification, uses Bayesian weighting method to correct subjective weights, and then obtains weight values at various levels of power quality, and utilizes The optimization algorithm with constraints obtains the optimized value of the weight, realizes the unity of subjectivity and objectivity of the weight, avoids the excessive influence of human factors when determining the weight, and makes the evaluation result objective and credible. With the help of the attribute recognition model, the final power quality rating is realized to better meet the needs of modern power quality comprehensive evaluation.

发明内容Contents of the invention

本发明提供一种基于贝叶斯与属性识别理论的电力需求侧用电质量综合评估方法，能实现电能各指标权重的合理计算，提高电能综合评估结果的合理性与可靠性，为电网的安全性提供可靠信息，并满足电力需求侧用户知电用电的需求。The invention provides a method for comprehensive evaluation of electricity quality on the demand side based on Bayesian and attribute recognition theory, which can realize reasonable calculation of the weight of each index of electric energy, improve the rationality and reliability of the comprehensive evaluation results of electric energy, and contribute to the safety of the power grid. Provide reliable information, and meet the needs of users on the power demand side to know and use electricity.

根据本发明的目的，提出一种电力需求侧电能质量评估系统，其包含一电力需求侧配电网，安装在若干监测点的电能参数监测仪，一用于传送电能参数监测仪监测数据的GPRS网络，一存储电能评估数据的数据库服务器，一用于评估计算的监测评估中心。According to the purpose of the present invention, a kind of electricity demand side power quality evaluation system is proposed, and it comprises a power demand side distribution network, the electric energy parameter monitoring instrument that is installed in several monitoring points, one is used to transmit the GPRS of electric energy parameter monitoring instrument monitoring data A network, a database server for storing electric energy evaluation data, and a monitoring and evaluation center for evaluation and calculation.

其中，电力需求侧的配电网为任意的电力用户配网点；Among them, the distribution network on the power demand side is any power user distribution network point;

其中，电能参数监测仪主要检测与计算电力需求侧用户的电能质量指标，包括：电压偏差、电压波动、三相不平衡、频率偏差、谐波含量、停电时间等；Among them, the power parameter monitor mainly detects and calculates the power quality indicators of users on the power demand side, including: voltage deviation, voltage fluctuation, three-phase unbalance, frequency deviation, harmonic content, power outage time, etc.;

其中，GPRS网络为用于移动电话用户的一种移动数据业务，结合无线通信与Internet，实现电能质量指标数据的无线传输；Among them, GPRS network is a mobile data service for mobile phone users, combining wireless communication and Internet to realize wireless transmission of power quality index data;

其中，数据库服务器用于接收和存储电能参数监测仪传送的所有电能质量指标数据；Among them, the database server is used to receive and store all the power quality index data transmitted by the power parameter monitor;

其中，监测评估中心用于电能质量指标数据的图表显示与评估结果的计算。Among them, the monitoring and evaluation center is used for graph display of power quality index data and calculation of evaluation results.

本发明的综合评估方法包括如下步骤：Comprehensive assessment method of the present invention comprises the steps:

（1）完成电力需求侧配电网监测点的数据采集与传输；(1) Complete the data collection and transmission of the monitoring points of the distribution network on the power demand side;

（2）对电能质量指标建立电能质量等级，并按着“效益型”指标与“成本型”指标对所建立的等级进行归一化处理，其中“成本型”指标即：属性值愈小愈好的指标，“效益型”指标即：属性值愈大愈好的指标；(2) Establish power quality grades for power quality indicators, and normalize the established grades according to the "benefit-type" indicators and "cost-type" indicators, where the "cost-type" indicators are: the smaller the attribute value, the better Good indicators, "benefit type" indicators are: the larger the attribute value, the better the indicator;

（3）在获取主观权重的基础上，利用贝叶斯方法对其进行修正，并以带约束的优化算法得到各指标的权重优化值；(3) On the basis of obtaining the subjective weight, use the Bayesian method to correct it, and use the optimization algorithm with constraints to obtain the weight optimization value of each index;

（4）借助于属性识别模型，对监测点的各项电能指标进行分级评估；(4) With the help of the attribute recognition model, the various electric energy indicators of the monitoring points are graded and evaluated;

（5）结合权重优化值，计算得到电能质量的综合评价等级，从而完成对电能质量的定量评估。(5) Combined with the weight optimization value, the comprehensive evaluation grade of power quality is calculated to complete the quantitative evaluation of power quality.

本发明与现有技术比较的有益成果是：本发明充分考虑电能质量各指标之间的关系，利用贝叶斯理论，对主观权重加以修正，而不是简单的对主客观权重进行组合，因而所获的指标权重值更符合实习情况，从而使评估结果更加合理。属性识别模型的利用，有利于快速获得评估等级值，使评估结果更加形象客观。The beneficial results of the present invention compared with the prior art are: the present invention fully considers the relationship between the various indicators of power quality, and uses Bayesian theory to correct the subjective weight instead of simply combining the subjective and objective weights. The obtained indicator weight values are more in line with the practice situation, so that the evaluation results are more reasonable. The use of the attribute recognition model is conducive to quickly obtaining the evaluation grade value and making the evaluation result more vivid and objective.

附图说明Description of drawings

图1是电力需求侧用电质量综合评估系统的架构图；Figure 1 is an architecture diagram of a comprehensive evaluation system for electricity quality on the power demand side;

图2是基于贝叶斯与属性识别理论的电能质量综合评估方法的流程示意图。Fig. 2 is a flow diagram of a comprehensive evaluation method for power quality based on Bayesian and attribute recognition theory.

具体实施方式Detailed ways

请参阅第1图，其为本发明用于电力需求侧用电质量综合评估系统的架构图，如图所示，本发明包括一电力需求侧配电网，安装在若干监测点的电能参数监测仪，一用于传送电能参数监测仪监测数据的GPRS网络，一存储电能质量指标数据的数据库服务器，一用于评估计算的监测评估中心。Please refer to Figure 1, which is a structure diagram of the present invention for the comprehensive evaluation system of power quality on the power demand side. As shown in the figure, the present invention includes a power demand side distribution network, and the monitoring of electric energy parameters installed at several monitoring points instrument, a GPRS network for transmitting the monitoring data of the electric energy parameter monitoring instrument, a database server for storing power quality index data, and a monitoring and evaluation center for evaluation and calculation.

其中，电能参数监测仪主要检测电力需求侧配电网各个监测点的电能质量指标数据，包括：电压偏差、电压波动、三相不平衡、频率偏差、谐波含量、停电时间等；Among them, the power parameter monitor mainly detects the power quality index data of each monitoring point of the power demand side distribution network, including: voltage deviation, voltage fluctuation, three-phase imbalance, frequency deviation, harmonic content, power outage time, etc.;

其中，监测评估中心完成本发明所提出的综合评估方法的全部计算过程，详细步骤如图2所示：Wherein, the monitoring and evaluation center completes the entire calculation process of the comprehensive evaluation method proposed by the present invention, and the detailed steps are as shown in Figure 2:

第1步：建立标准指标值矩阵。构建五个等级下的各电能指标值矩阵Y：Step 1: Establish a matrix of standard indicator values. Construct the matrix Y of each electric energy index value under five levels:

$Y Y = = {(({y the y}_{i i,, j j}))}_{c c \times \times m m} = = [\begin{matrix} {y the y}_{1,1 1,1} & {y the y}_{1,2 1,2} & \cdot \cdot \cdot \cdot \cdot \cdot & {y the y}_{11,, m m} \\ {y the y}_{2,1 2,1} & {y the y}_{2,2 2,2} & \cdot &Center Dot; \cdot &Center Dot; \cdot \cdot & {y the y}_{22,, m m} \\ \cdot \cdot & \cdot \cdot & \cdot \cdot & \cdot \cdot \\ \cdot \cdot & \cdot \cdot & \cdot &Center Dot; & \cdot \cdot \\ \cdot &Center Dot; & \cdot &Center Dot; & \cdot \cdot & \cdot &Center Dot; \\ {y the y}_{c c,, 11} & {y the y}_{c c,, 22} & \cdot \cdot \cdot &Center Dot; \cdot &Center Dot; & {y the y}_{c c,, m m} \end{matrix}] - - - - - - ((11))$

其中，m为电能质量指标个数，c为评估等级数（这里c=5），y_i,j为电能参数监测仪检测的电能质量指标数据。再对各指标数据进行标准化处理，归一化为同一量纲，得到标准指标值矩阵

其标准化处理方法为：Among them, m is the number of power quality indicators, c is the number of evaluation levels (here c=5), y_{i, j} are the data of power quality indicators detected by the power parameter monitor. Then standardize the index data and normalize them to the same dimension to obtain the standard index value matrix

Its standardized processing method is:

${r r}_{i i,, j j} = = \{\begin{matrix} \frac{min min {{{y the y}_{i i,, j j}}}_{i i = = 11}^{c c}}{{y the y}_{i i,, j j}},, {y the y}_{i i,, j j} &Element; &Element; R R \\ \frac{{y the y}_{i i,, j j}}{max max {{{y the y}_{i i,, j j}}}_{i i = = 11}^{c c}},, {y the y}_{i i,, j j} &Element; &Element; A A \end{matrix} - - - - - - ((22))$

式中，R为“成本型”指标，即：属性值愈小愈好的指标；A为“效益型”指标，即：属性值愈大愈好的指标。涉及的电能质量指标有电压偏差、频率偏差、谐波含量、三项不平衡、电压波动、停电时间这六项指标；In the formula, R is a "cost-type" index, that is, the smaller the attribute value, the better the index; A is the "benefit type" index, that is, the larger the attribute value, the better the index. The power quality indicators involved include six indicators: voltage deviation, frequency deviation, harmonic content, three unbalances, voltage fluctuations, and power outage time;

第2步：贝叶斯方法确定权重。在确定各等级下指标主观权重

的基础上，采用贝叶斯方法，用各等级下的标准指标值对主观权重进行修正。设主观权重为u₁,u₂,...,u_m，即指标I_j的概率为p(I_j)=u_j，且在指标I_j下，等级G_i发生的概率为p(G_i|I_j)=r_i,j，利用贝叶斯方法，在等级G_i下，第j个指标的权重（后验概率）为：Step 2: Bayesian method to determine the weights. Determining the subjective weight of indicators at each level

On the basis of , the Bayesian method is used to correct the subjective weights with the standard index values at each level. Let the subjective weights be u₁ , u₂ ,..., u_m , that is, the probability of index I_j is p(I_j )=u_j , and under index I_j , the probability of grade G_i occurring is p(G_i |I_j )=r_i,j , using the Bayesian method, under the level G_i , the weight (posterior probability) of the jth index is:

$p p (({I I}_{j j} | | {G G}_{i i})) = = \frac{p p (({I I}_{j j})) p p (({G G}_{i i} | | {I I}_{j j}))}{{Σ Σ}_{j j = = 11}^{m m} p p (({I I}_{j j})) p p (({G G}_{i i} | | {I I}_{j j}))} = = \frac{{μ μ}_{j j} {r r}_{i i,, j j}}{{Σ Σ}_{j j = = 11}^{m m} {μ μ}_{j j} {r r}_{i i,, j j}} = = {w w}_{j j}^{((i i))} - - - - - - ((33))$

构建目标优化模型：Build the target optimization model:

$J J = = min min (({Σ Σ}_{i i = = 11}^{c c} {Σ Σ}_{j j = = 11}^{m m} {(({w w}_{j j} - - {w w}_{j j}^{((i i))}))}^{22} {r r}_{i i,, j j}^{22}))$

$s the s . . t t . . \{\begin{matrix} {Σ Σ}_{j j = = 11}^{m m} {w w}_{j j} = = 11 \\ 11 > > {w w}_{j j} > > 00 \end{matrix}$

式中，

为主观权重，w_j为权重修正值，m为电能质量指标个数，c为评估等级数，r_i,j为标准指标值矩阵的元素。In the formula,

is the subjective weight, w_j is the weight correction value, m is the number of power quality indicators, c is the number of evaluation levels, r_{i, j} are the elements of the standard index value matrix.

利用非线性迭代寻优算法得到：电能质量整体权重

w_j∈(0,1)。Using nonlinear iterative optimization algorithm to obtain: the overall weight of power quality

w_j ∈ (0,1).

第3步：建立属性测度矩阵。设某监测点的m项指标数据为

为X的某类属性空间的有序分割，且满足q₁>q₂>…>q_c，由每个指标的分类界限可写出分类标准矩阵为

f = [\begin{matrix} a_{1,1} & a_{1,2} & \cdot \cdot \cdot & a_{1, c} \\ a_{2,1} & a_{2,2} & \cdot \cdot \cdot & a_{2, c} \\ \cdot & \cdot & \cdot & \cdot \\ \cdot & \cdot & \cdot & \cdot \\ \cdot & \cdot & \cdot & \cdot \\ a_{m, 1} & a_{m, 2} & \cdot \cdot \cdot & a_{m, c} \end{matrix}],

设u_i,j可表示为x_i属于属性q_j的定量描述（即：x_i∈q_j）的属性测度，则监测点X的属性测度为U＝{u_i,j},i＝1,2,…,m;j＝1,2,…,c：Step 3: Establish attribute measure matrix. Let the m index data of a monitoring point be

is an ordered partition of a certain attribute space of X, and satisfies q₁ >q₂ >…>q_c , the classification criterion matrix can be written from the classification boundary of each index as

f = [\begin{matrix} a_{1,1} & a_{1,2} & &Center Dot; &Center Dot; &Center Dot; & a_{1, c} \\ a_{2,1} & a_{2,2} & \cdot \cdot \cdot & a_{2, c} \\ \cdot & \cdot & \cdot & \cdot \\ \cdot & \cdot & \cdot & \cdot \\ \cdot & \cdot & \cdot & &Center Dot; \\ a_{m, 1} & a_{m, 2} & \cdot \cdot \cdot & a_{m, c} \end{matrix}],

Suppose u_i,j can be expressed as the attribute measure of x_i belonging to the quantitative description of attribute q_j (ie: x_i ∈ q_j ), then the attribute measure of monitoring point X is U={u_i,j },i=1 ,2,...,m;j=1,2,...,c:

（1）当x_i≤a_i,1时，(1) When x_i ≤ a_i,1 ,

u_i,1=1，u_i,2=…=u_i,c=0 （5）u_i,1 =1, u_i,2 =...=u_i,c =0 (5)

（2）当x_i>a_i,c时，(2) When x_i >a_i,c ,

u_i,c=1，u_i,1=…=u_i,c-1=0 （6）u_i,c =1, u_i,1 =...=u_i,c-1 =0 (6)

（3）当a_i,j<x_i≤a_i,j+1时，(3) When a_i,j <x_i ≤a_i,j+1 ,

${u u}_{i i,, j j} = = \frac{| | {x x}_{i i} - - {a a}_{i i,, j j + + 11} | |}{| | {a a}_{i i,, j j} - - {a a}_{i i,, j j + + 11} | |},, {u u}_{i i,, j j + + 11} = = \frac{| | {x x}_{i i} - - {a a}_{i i,, j j} | |}{| | {a a}_{i i,, j j} - - {a a}_{i i,, j j + + 11} | |},, - - - - - - ((77))$

第4步：结合权重和属性测度求电能质量等级。监测点X的综合指标属于各类属性的测度为V=(v₁,v₂,...,v_c)，Step 4: Combining the weight and attribute measure to find the power quality level. The comprehensive index of monitoring point X belongs to the measure of various attributes as V=(v₁ ,v₂ ,...,v_c ),

${v v}_{j j} = = {Σ Σ}_{i i = = 11}^{m m} {w w}_{i i} \cdot \cdot {u u}_{i i,, j j} - - - - - - ((88))$

由此得最终的电能质量等级为 $k_{0} = \min {k : Σ_{j = 1}^{k} v_{j} &GreaterEqual; λ, 1 \leq k \leq c} .$ Thus, the final power quality level is $k_{0} = \min {k : Σ_{j = 1}^{k} v_{j} &Greater Equal; λ, 1 \leq k \leq c} .$

等级值k₀越小，代表电能偏差越小，电能质量越好。在电力需求侧用电质量综合评估中，通过对电能质量进行分级，并通过计算比较评估结果所处的等级，可以直观的把握电能的质量状况。对于某些结果处于同一等级的情况，可由其综合属性测度超过置信度准则的程度进行判断，从而可以比较出同一等级情况下，哪个电能质量更好。The smaller the level value_k0 , the smaller the power deviation and the better the power quality. In the comprehensive evaluation of power quality on the power demand side, by grading the power quality and comparing the grades of the evaluation results through calculation, the quality of power can be intuitively grasped. For some results at the same level, it can be judged by the degree to which the comprehensive attribute measurement exceeds the confidence criterion, so that it can be compared which power quality is better under the same level.

Claims

Translated fromChinese

1.一种电力需求侧用电质量综合评估方法，其评估的系统包括一电力需求侧配电网、安装在若干监测点的电能参数监测仪、一用于传送电能参数监测仪监测数据的GPRS网络、一存储电能评估数据的数据库服务器和一用于评估计算的监测评估中心，其中，电力需求侧配电网为任意的电力用户配网点；电能参数监测仪用来检测和计算电力需求侧用户的电能质量指标；GPRS网络结合无线通信与Internet，实现电能质量指标数据的无线传输；数据库服务器用于接收和存储电能参数监测仪传送的所有电能质量指标数据；监测评估中心用于电能质量指标数据的图表显示与评估结果的计算；其特征在于，综合评估方法包括如下步骤：1. A method for comprehensive evaluation of power consumption quality on the power demand side, the system of its assessment includes a power demand side distribution network, power parameter monitors installed at several monitoring points, a GPRS for transmitting power parameter monitor monitoring data network, a database server storing electric energy evaluation data, and a monitoring and evaluation center for evaluation and calculation, wherein, the power demand side distribution network is any power user distribution network point; the power parameter monitor is used to detect and calculate power demand side users GPRS network combined with wireless communication and Internet to realize wireless transmission of power quality index data; the database server is used to receive and store all power quality index data transmitted by the power parameter monitor; the monitoring and evaluation center is used for power quality index data The chart display and the calculation of the evaluation result; it is characterized in that the comprehensive evaluation method includes the following steps:

（1）所述系统完成电力需求侧配电网监测点的数据采集与传输；(1) The system completes the data collection and transmission of the monitoring points of the distribution network on the power demand side;

2.根据权利要求1所述的一种电力需求侧用电质量综合评估方法，其特征在于，所述电能质量指标包括电压偏差、电压波动、三相不平衡、频率偏差、谐波含量和停电时间。2. A method for comprehensive evaluation of power quality on the power demand side according to claim 1, wherein the power quality indicators include voltage deviation, voltage fluctuation, three-phase unbalance, frequency deviation, harmonic content and power outage time.

3.根据权利要求1或2所述的一种电力需求侧用电质量综合评估方法，其特征在于，所述步骤（3）的权重修正方法为迭代寻优，即构造优化的目标函数，其最小值对应的解即为权重修正值：3. A method for comprehensive evaluation of power consumption quality on the demand side according to claim 1 or 2, characterized in that the weight correction method in step (3) is iterative optimization, that is, constructing an optimized objective function, which The solution corresponding to the minimum value is the weight correction value:

J J = = min min (({Σ Σ}_{i i = = 11}^{c c} {Σ Σ}_{j j = = 11}^{m m} {(({w w}_{j j} - - {w w}_{j j}^{((i i))}))}^{22} {r r}_{i i,, j j}^{22}))

s the s . . t t . . \{\begin{matrix} {Σ Σ}_{j j = = 11}^{m m} {w w}_{j j} = = 11 \\ 11 > > {w w}_{j j} > > 00 \end{matrix}

式中，