CN112990500A - Transformer area line loss analysis method and system based on improved weighted gray correlation analysis - Google Patents
Transformer area line loss analysis method and system based on improved weighted gray correlation analysis Download PDFInfo
- Publication number
- CN112990500A CN112990500A CN202110356039.7A CN202110356039A CN112990500A CN 112990500 A CN112990500 A CN 112990500A CN 202110356039 A CN202110356039 A CN 202110356039A CN 112990500 A CN112990500 A CN 112990500A
- Authority
- CN
- China
- Prior art keywords
- line loss
- analysis
- correlation
- electrical characteristic
- station area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
本发明提供了一种基于改进加权灰色关联分析的台区线损分析方法及系统，获取不同类型台区的静态参数和动态参数；将不同台区静态参数和动态参数作为电气特征指标，将其和对应线损率归一化处理，利用基于改进分辨系数选取的灰色关联分析方法，确定各参数与台区线损的关联度大小，确定各电气特征指标的客观权重值；将获得的灰色关联度和客观权重结合，得到各电气特征指标的加权灰色关联度，确定最优的强关联度参数；构建线损分析计算模型，基于最优的强关联度参数，对台区线损进行计算、预测。本发明可以有效提高线损预测精度，为台区线损影响因素的挖掘和线损预测提供有效途径。
The invention provides a station line loss analysis method and system based on improved weighted grey correlation analysis, which can obtain static parameters and dynamic parameters of different types of stations; and the corresponding line loss rate normalization processing, using the grey correlation analysis method based on the selection of the improved resolution coefficient, to determine the degree of correlation between each parameter and the line loss of the station area, and to determine the objective weight value of each electrical characteristic index; Combined with the objective weight, the weighted gray correlation degree of each electrical characteristic index is obtained, and the optimal strong correlation degree parameter is determined; the line loss analysis and calculation model is constructed, and the line loss of the station area is calculated and predict. The invention can effectively improve the line loss prediction accuracy, and provides an effective way for mining the influence factors of the line loss in the station area and predicting the line loss.
Description
Technical Field
The invention belongs to the technical field of distribution room line loss analysis, and particularly relates to a distribution room line loss analysis method and system based on improved weighted gray correlation analysis.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
In the management of power enterprises, the line loss rate is an important index for measuring the management level, and meanwhile, the line loss rate also represents the comprehensive economic index of the planning design, the operation management level and the production technology of a power system. Meanwhile, the economy of China enters a new normal state, the electricity selling rate is increased slowly, the market competition is intensified, and the power grid is supported by the high-speed increase of the electricity selling rate and the development of companies is difficult to continue. With the deep advancement of electric power reform, the power transmission and distribution prices of each province are continuously checked and fixed, and electric power enterprises put forward higher requirements for further enhancing line loss management, excavating loss reduction potential and improving operational benefits.
In the line loss management work of power supply enterprises, the line loss of the transformer area is managed in a unified management mode at present, so that the difference of the causes of the line loss of different transformer areas is difficult to reflect, and an effective way for further reducing the loss is not easy to find. Therefore, how to accurately analyze the line loss cause of the distribution room and accurately position the loss reduction breakthrough of the distribution rooms of different categories becomes a key point and a difficult point of line loss management of power supply enterprises. The traditional line loss calculation method of the low-voltage transformer area generally adopts a mechanical method, and due to the fact that models have more simplifying assumptions and low calculation accuracy, the line loss lean analysis requirement of a power supply enterprise cannot be met. In addition, the line loss calculation is mainly based on a network topological structure and line parameters, and typical factors influencing the line loss cannot be deeply excavated to find the reason of the line loss abnormality.
In order to effectively reduce the loss, the reason for analyzing the line loss abnormity is a breakthrough, so the relationship between the line loss influence factor and the line loss needs to be deeply excavated. According to the knowledge of the inventor, certain results are obtained in the aspect of extraction and analysis of key factors of the line loss of the transformer area in the prior art, but the influence weight of each factor is not accurately and comprehensively considered in the prior art, so that the influence degree of each factor on the line loss of the transformer area cannot be accurately quantified, the analysis of the key factors of the line loss lacks optimality and fairness, and the establishment of a loss reduction strategy cannot be guided in a targeted and correct manner.
Disclosure of Invention
The invention provides a distribution room line loss analysis method and system based on improved weighted grey correlation analysis to solve the problems.
According to some embodiments, the invention adopts the following technical scheme:
a distribution room line loss analysis method based on improved weighted gray correlation analysis comprises the following steps:
acquiring static parameters and dynamic parameters of different types of transformer areas;
taking static parameters and dynamic parameters of different transformer areas as electrical characteristic indexes, normalizing the electrical characteristic indexes and corresponding line loss rates, determining the degree of association between each parameter and the line loss of the transformer areas by using a gray association analysis method selected based on improved resolution coefficients, and determining the objective weight value of each electrical characteristic index;
combining the obtained grey correlation degree with the objective weight to obtain the weighted grey correlation degree of each electrical characteristic index, and determining the optimal strong correlation degree parameter;
and constructing a line loss analysis calculation model, and calculating and predicting the line loss of the transformer area based on the optimal strong correlation parameter.
As an alternative embodiment, the static parameters include several of a distribution area capacity, a residential capacity, a non-residential capacity, a residential user proportion, a non-residential user proportion, and a photovoltaic user proportion; the dynamic parameters include several of a maximum daily power curve, a current curve, and a voltage curve for a typical daily month.
As an alternative embodiment, the different types of transformer areas include low-voltage transformer areas mainly used by residential users and low-voltage transformer areas mainly used by non-residential users.
As an alternative embodiment, the specific process of determining the degree of association between each parameter and the line loss of the transformer area based on the gray correlation analysis method selected by improving the resolution coefficient includes: taking the electrical characteristic index influencing the line loss as a comparison sequence, taking the line loss of the transformer area as a reference sequence, and calculating the correlation degree of the comparison sequence and the reference sequence;
in the calculation process, the mean value of all the absolute values of the difference values and the ratio of the mean value to the maximum value of the corresponding difference values are calculated, and the resolution coefficient is adjusted according to the ratio.
As an alternative embodiment, the specific process of determining the objective weight value of each electrical characteristic index includes: and calculating objective weight of the corresponding index based on the variation degree and the conflict of the index quantitative by the variation degree of the index standard variation index and the conflict of the correlation coefficient quantitative index.
As an alternative embodiment, the specific process of constructing the line loss analysis calculation model includes:
constructing a BP neural network model, and selecting strongly-related electrical characteristic indexes and corresponding transformer area line loss rates as input and output of the model respectively, wherein the number of nodes of an input layer depends on the number of the electrical characteristic indexes;
and (4) carrying out normalization on the training data, and training the BP neural network model by using the training data until the set conditions are met.
As a further limitation, the set condition is that the error reaches an expected range or reaches a set learning number.
A distribution room line loss analysis system based on improved weighted gray correlation analysis comprises:
the data receiving module is configured to acquire static parameters and dynamic parameters of different types of transformer areas;
the station area line loss key factor identification module is configured to take static parameters and dynamic parameters of different station areas as electrical characteristic indexes, normalize the electrical characteristic indexes and corresponding line loss rates, determine the degree of association between each parameter and the station area line loss by using a gray association analysis method selected based on an improved resolution coefficient, determine the objective weight value of each electrical characteristic index, combine the obtained gray association degree and the objective weight to obtain the weighted gray association degree of each electrical characteristic index, and determine the optimal strong association degree parameter;
and the transformer area line loss analysis and calculation module is configured to construct a line loss analysis and calculation model, and calculate and predict the transformer area line loss based on the optimal strong correlation parameter.
An electronic device comprising a memory and a processor and computer instructions stored on the memory and executed on the processor, the computer instructions, when executed by the processor, performing the steps of the above method.
A computer readable storage medium storing computer instructions which, when executed by a processor, perform the steps of the above method.
Compared with the prior art, the invention has the beneficial effects that:
(1) the invention introduces the static parameters and the dynamic parameters of the transformer area at the same time, and can more comprehensively reflect the static and dynamic characteristics of the transformer area. Static network loss factors of the transformer area, such as transformer area capacity, resident user proportion and other numerical values are relatively stable, and the essence of line loss change of the transformer area can be reflected; meanwhile, dynamic factors such as fluctuation of a load curve, different capacitor switching modes and different operation modes also have certain influence on the line loss of the power distribution network. Therefore, the static and dynamic parameters of the transformer area are introduced simultaneously, so that key factors of the transformer area can be more accurately identified, and theoretical support is provided for making a loss reduction strategy.
(2) When the correlation between each influence factor and the line loss of the transformer area is analyzed, a CRITIC method is introduced on the basis of the traditional grey correlation analysis, the grey correlation analysis is taken as a central model, and the CRITIC method is taken as an auxiliary model. The grey correlation analysis has obvious advantages in the aspect of processing small samples and bad information, and the CRITIC method is an objective weighting method which considers the degree of variability of indexes and gives consideration to the correlation among the indexes, so that the objective weighting based on the whole index system can be realized. By combining the advantages of the two methods, the relatively optimal association degree can be calculated, the objective optimal sequencing of all factors is realized, and the identified key factors of the distribution room are more in line with the engineering practice.
(3) The invention carries out analysis and calculation of the line loss of the transformer area based on the identified key factors of the transformer area, and the line loss calculation under the condition of optimizing input variables can greatly reduce the burden of model training, improve the learning speed and the accuracy of model calculation and provide an effective way for lean management of the line loss of the transformer area.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of a method according to the present embodiment;
FIG. 2 is a table area electrical characteristic index system according to the present embodiment;
fig. 3 is a schematic structural diagram of a BP neural network model according to this embodiment;
fig. 4 is a flowchart of an algorithm of the BP neural network according to this embodiment.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
According to the technical scheme, the identification of the key influence factors of the transformer area and the corresponding prediction of the line loss of the transformer area are realized aiming at different types of transformer areas, the identification capability of the key factors of the transformer area is enhanced, the current situation of the management of the line loss of the transformer area lacking differentiation in 'one-time cutting' can be improved, and certain data support and direction guidance can be provided for the large technical modification of the transformer area and the planning of a distribution network.
In one or more embodiments, as shown in fig. 1, a method for analyzing line loss of a distribution room based on improved weighted gray correlation analysis includes the following steps:
step 1, selecting low-voltage distribution areas with qualified collection success rate (the daily collection success rate is more than 98%) in distribution areas within one year, wherein the low-voltage distribution areas comprise low-voltage distribution areas mainly comprising residential users and low-voltage distribution areas mainly comprising non-residential users;
and 2, collecting static parameters and dynamic parameters of each distribution area. The static parameters comprise the transformer area capacity, the residential capacity, the non-residential capacity, the residential user proportion, the non-residential user proportion and the photovoltaic user proportion. The dynamic parameters comprise a maximum daily power curve, a current curve and a voltage curve of a typical daily month;
step 3, normalizing the electrical characteristic indexes and the corresponding line loss rates of the distribution areas;
step 4, determining the magnitude of the correlation degree based on the gray correlation analysis selected by the improved resolution coefficient;
step 5, determining objective weight values of all electrical characteristic indexes based on a CRITIC method;
step 6, combining the gray correlation degrees obtained in the step 4 and the step 5 with objective weights, and further obtaining weighted gray correlation degrees of all electrical characteristic indexes;
step 7, determining the index number of the optimal electrical characteristic index system according to a K-fold cross verification method;
and 8, building a BP neural network transformer area line loss analysis calculation model based on the strong correlation factor.
The method can effectively make up the defects of large sample demand, requirement of the sample to obey a certain typical probability distribution, requirement of the mutual independence of all factors, large calculated amount and the like in the traditional correlation analysis methods such as regression analysis, variance analysis, principal component analysis and the like, and has certain practical value for the actual engineering which is difficult to obtain a large number of high-quality samples; the method improves the selection of the traditional grey correlation analysis resolution coefficient, and the resolution coefficient is dynamically selected according to the value of the electrical characteristic index, so that the resolution of the correlation degree of each factor is effectively improved, and the ranking of each electrical characteristic index is more in line with the objective reality; the method introduces a CRITIC method to calculate the objective weight of each electrical index, overcomes the defect that the traditional grey correlation analysis treats each factor in equal weight, highlights the contribution degree of each influence factor to the line loss of the corresponding distribution room, and can better reflect the volatility and the correlation among data; meanwhile, the key factors extracted by the method can obviously improve the calculation precision of the line loss of the transformer area, and have clear practical value and practical significance for loss reduction and efficiency increase of the transformer area.
The above steps are specifically described below.
In step 2, the static parameters and the dynamic parameters of each distribution area are collected, as shown in fig. 2, the system is an electrical characteristic index system of the distribution area, and specifically includes:
(1) residential capacity x1: the electricity consumption capacity of the residential users;
(2) non-resident capacity x2: electricity consumption capacity of non-residential users;
(3) residential user ratio x3: the ratio of the number of resident households to the total number of subscribers in the distribution area;
(4) proportion x of non-residents4: the ratio of the number of non-residential users to the total number of users in the distribution area;
(5) photovoltaic user ratio x5: the ratio of the number of photovoltaic users to the total number of users in the distribution area;
(6) average daily maximum power x6: typical monthly maximum power average;
the calculation formula is shown as formula (1):
wherein N is the number of days of the typical day month, PmaxDaily maximum power value for the typical day of the month.
(7) Average load factor x7: the ratio of typical daily apparent power to cell capacity;
the calculation formula is shown in formula (2):
in the formula: u shapeijAnd IijA typical 24-point daily voltage curve and current curve, SijIs the transformer capacity of the transformer area.
(8) Coefficient of load curve shape x8: typical daily mean root current to current average ratio.
The calculation formula is as follows:
in the formula IjfIs a typical daily root of Chinese scholarCurrent, IavTypical daily current mean. The expression is as follows:
due to line loss and K2Is proportional, so K is usually used2Representing the load curve shape factor.
In the step 4, correlation analysis is mainly performed by adopting gray correlation analysis, and the relation between the electrical characteristic parameters of the transformer area and the corresponding line loss is in a gray state, so that the transformer area is a typical gray system and is suitable for performing correlation analysis by adopting gray correlation analysis. Meanwhile, the grey correlation analysis has low requirement on the number of samples, and has certain application value to the actual engineering which is difficult to obtain high-quality samples in a large quantity. Grey correlation analysis reflects the degree of correlation between several comparison sequences and a reference sequence by determining their relationship, usually expressed as confidence, which is higher the smaller the difference between the two sequences. Using the electric characteristic index influencing the line loss as a comparison sequence X1Line loss of transformer area as reference sequence X0. Wherein x0(k)，x1(k) Is X0And X1The value at time k. After dimensionless transformation, the sequences X are now compared1And reference sequence X0The degree of association is:
in the formula: xi01(k) At a time X of k1And X0A correlation coefficient between; ρ ∈ (0, ∞) is a resolution coefficient, and is generally 0.5.
The correlation coefficient between the two sequences is the average of the correlation coefficients at all time points:
the closer the correlation coefficient is to 1, the better the correlation is, and the greater the correlation between the electrical characteristic index and the corresponding station line loss rate, that is, the greater the influence degree on the station line loss when the index fluctuates.
In step 4, the resolution factor ρ is
The correlation coefficient between each index and the line loss is not only related to the value of each index, but also depends on the values of other indexes. The value of the resolution coefficient rho determines the correlation coefficient xi corresponding to the index itself by other indexes01(k) The influence of (c). Therefore, the resolution coefficient represents the indirect influence degree of each index on the relevance, and the selection principle comprises the following two points:
1) the value of the resolution coefficient rho is dynamically determined according to the value change of each index so as to reflect the integrity of the association degree and increase the resolution of the association degree of different factors as much as possible;
2) and through reasonable value taking of rho, the robustness of the evaluation system is improved, and the influence of singular values on the evaluation result is overcome.
The grey correlation analysis method for improving resolution coefficient selection comprises the following steps:
step 4.1, define ΔavAs the mean of the absolute values of all differences, i.e.
Step 4.2, remember
Step 4.3, when ω is less than 1/3, ω is less than or equal to ρ less than or equal to 1.5 ω, and ρ is generally taken to be 1.5 ω; when ω is equal to or greater than 1/3, 1.5 ω < ρ ≦ 2 ω, and ρ is generally equal to 2 ω.
In step 5, the objective weight of each electrical characteristic index is determined by the CRITIC method in this embodiment. The CRITIC method is an objective weighting method, and measures the objective weight of the index by evaluating the variation degree and the conflict of the index. The relevance between indexes is considered while the index variability is considered, and objective weighting based on the whole index system can be realized.
Wherein the degree of variation of the index is determined by the standard deviation
To quantify:
in the formula, xijIndicating the jth index value in the ith index, N indicating the number of sample data, SjIt represents the variation degree of the jth index value in some kind of index.
The conflict of indexes is determined by the correlation coefficient rjkTo quantify:
in the formula, xikRepresenting the kth index value in the ith index, n representing the number of indexes in the index system, RjAnd a feature conflict index value representing the jth feature quantity.
Objective weight W of jth index in index systemjComprises the following steps:
in the formula, CjAnd (4) representing the information quantity, namely the action size of the jth index in the CRITIC weight evaluation system. As can be seen from the above equation, the larger the amount of information included in a certain index is, the greater the importance is, and the greater the weight value is.
In step 6, the correlation is calculated by combining the improved grey correlation analysisDegree of association r of1' weight value W obtained by equation (11)jFinally, the weighted relevance of each index is obtained:
γ＝r1'×wj (12)
in step 8, the BP neural network structure diagram is shown in fig. 3, and includes three parts, i.e., an input layer, a hidden layer, and an output layer. The BP neural network is a method based on reverse error transfer, and the learning process of the BP neural network consists of two parts, namely forward propagation of input data and reverse propagation of errors. The forward propagation refers to the input of input samples from an input layer and the propagation to an output layer through the layer-by-layer processing of each hidden layer. If the output result of the output layer does not reach the expected value, the error is transferred to the reverse propagation. The error back propagation is to reversely transmit the output error layer by layer through a hidden layer and adjust the weight and the threshold of each neuron. The process of continuously adjusting the weight and the threshold is the network learning training process until the error reaches an expected range or reaches a set learning number. The algorithm flow chart of the BP neural network is shown in FIG. 4, and the whole process can be divided into three steps:
(1) construction of a BP neural network: selecting strongly correlated electrical characteristic indexes and corresponding transformer area line loss rates as input and output of the model respectively, wherein the number of nodes of an input layer depends on the number of the electrical characteristic indexes, and the number of nodes of an implicit layer can be roughly determined according to an equation (13):
in the formula, m and n are the numbers of neurons in the output layer and the input layer, respectively, and a is a constant between [0 and 10 ].
The learning rate determines the amount of weight variation generated in each round of training. Too high a learning rate may result in system instability, but too low a learning rate may result in longer training time, possibly slow convergence, but may ensure that the net error values jump out of the valleys of the error surface and eventually approach the minimum error values. In general, a smaller learning rate tends to be selected to ensure system stability. The learning rate is selected to be 0.01-0.8.
And (4) selecting the expected error. The expected error value should also be determined by comparison training during the training of the network. The so-called "fit" is determined relative to the number of nodes of the required hidden layer, since a smaller expected error is obtained by adding nodes of the hidden layer, and training time.
(2) BP neural network training: first, a normalization process of training data is performed. The normalization processing of the data is to convert all data points into constants between [0,1], and aims to cancel the difference of the magnitude of each dimension of data and avoid network training errors caused by large difference of the magnitude of input and output data. The data normalization method mainly comprises the following two methods:
(1) maximum-minimum method:
xk＝(xk-xmin)/(xmax-xmin) (14)
in the formula, xminIs the minimum value, x, in the data sequencemaxIs the maximum value in the data sequence.
(2) Mean variance method, the functional form is as follows:
xk＝(xk-xmean)/xvar (15)
in the formula, xmeanIs the mean of the data sequence, xvarIs the variance of the data sequence.
In this embodiment, a first data normalization method is adopted, and an MATLAB self-carrying function mapminmax is adopted as a normalization function. Meanwhile, the accuracy of network prediction is greatly influenced by the selection of the hidden layer and output layer functions (node transfer functions). The hidden layer function and the output layer function selected in this embodiment are tansig and purelin, respectively. The calculation formula is as follows:
purelin(n)＝n (17)
in the learning algorithm setting of the BP neural network, the LM algorithm is often selected. The LM algorithm is high in calculation efficiency, and a second derivative is adopted in a weight updating part, as shown in a formula (18):
Δw＝[JT(w)J(w)+μI]-1JT(w)e(w) (18)
wherein, I is a unit matrix, mu is the user-defined learning rate, and J (w) is a Jacobian matrix.
Of course, in other embodiments, other approaches may be selected.
(3) Prediction of the BP neural network: inputting the test and sample data into a trained BP neural network transformer area line loss analysis calculation model, reasonably predicting transformer area line loss, and analyzing prediction errors. The error calculation formula is as follows:
in the formula:
calculating the line loss value of the transformer area; y isijAnd the real line loss value of the platform area.
The above proposed method for analyzing line loss of a distribution room based on the improved weighted gray correlation analysis is verified by using specific examples.
Taking data of a low-voltage distribution area in a certain area as an example to perform distribution area key factor identification and line loss analysis, and selecting 175 samples in total, wherein 140 samples are mainly resident users, and the rest 35 samples are mainly non-resident users. Each sample comprises eight electrical characteristic indexes, namely, residential capacity, non-residential capacity, residential user proportion, non-residential user proportion, photovoltaic user proportion, average daily maximum power, average load rate, load curve shape coefficient and line loss rate of the sample distribution area. In order to determine the relationship between each electrical index and the line loss rate, the correlation degree is calculated by a gray correlation model with improved resolution coefficient and a weighted gray correlation model with further introduced weighting method. The results of the association calculation of the two types of zones are shown in tables 1 and 2, respectively.
Further take the first type of station area as an example. The electrical indexes are sorted according to the relevance degree, the number of the electrical characteristic indexes is increased from 2 to 8, the increasing direction is the direction of reducing the relevance degree, and the average calculation error under different electrical characteristic index systems is calculated to determine the index number of the optimal electrical characteristic index system. The calculation results are shown in table 3.
Further, taking the first type of sample of the transformer area as an example, 140 samples are divided into a training set and a test set according to a ratio of 126: 14. The results of the test set calculations are shown in table 4, comparing the effect of the input quantities selected under the conventional grey correlation analysis and the method of the present invention on the model error.
TABLE 1 degree of correlation between electrical characteristic indexes of first-class distribution area
TABLE 2 correlation degree of each electrical characteristic index of the second type of distribution area
TABLE 3 average calculated error under different electrical characteristic index systems
TABLE 4 average calculated error comparison
(1) Influence of dynamic selection of resolution coefficients on correlation magnitude in grey correlation analysis
The correlation degree intervals calculated by the conventional gray correlation analysis, the gray correlation analysis for improving the resolution coefficient and the improved weighted gray correlation analysis are respectively 0.028,0.090 and 0.480. Therefore, the resolution of the relevance of each influencing factor is improved by dynamically selecting the resolution coefficient, so that the relevance sequencing of each electrical characteristic index has more integrity. Meanwhile, the dynamic selection of the resolution coefficient can overcome the influence of individual oversize or undersize singular values on the correlation evaluation result, and the robustness of the whole evaluation system is improved.
(2) Effect of CRITIC method introduction on correlation size
Aiming at the defect of equal-weight processing of all factors in the traditional grey correlation analysis, the invention introduces an objective weighting method CRITIC to calculate the weight value correction of all factors. The CRITIC method determines the objective weight of each electrical characteristic index according to the variation degree of the indexes and the correlation among the indexes. By combining the objective weighting method with the traditional grey correlation analysis, the relevance ranking can reflect the contribution degree of each electrical characteristic index to the line loss of the transformer area, and the method is more suitable for the application of actual engineering.
Based on the same inventive concept, the present embodiment provides a distribution room line loss analysis system based on improved weighted gray correlation analysis, including:
and the data acquisition module is used for acquiring static parameters and dynamic parameters of the transformer area mainly comprising residential users and the transformer area mainly comprising non-residential users. The static parameters comprise the transformer area capacity, the residential capacity, the non-residential capacity, the residential user proportion, the non-residential user proportion and the photovoltaic user proportion. The dynamic parameters comprise a maximum daily power curve, a current curve and a voltage curve of a typical daily month;
the station area line loss key factor identification module is used for determining the correlation between the electrical characteristic indexes and the corresponding line loss of the different types of station areas by adopting improved weighted gray correlation analysis;
and the transformer area line loss analysis and calculation module is used for constructing a transformer area line loss BP neural network model based on the identified key factors and the corresponding line loss, so that the transformer area line loss is accurately calculated.
The method can also comprise an optimal electrical characteristic index system establishing module, wherein the optimal electrical characteristic index system establishing module is used for calculating the average line loss prediction errors under different numbers of input quantities, and selecting the input quantities with the optimal model performance to form the optimal electrical characteristic index system corresponding to the distribution area.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110356039.7A CN112990500B (en) | 2021-04-01 | 2021-04-01 | Transformer area line loss analysis method and system based on improved weighted gray correlation analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110356039.7A CN112990500B (en) | 2021-04-01 | 2021-04-01 | Transformer area line loss analysis method and system based on improved weighted gray correlation analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112990500A true CN112990500A (en) | 2021-06-18 |
| CN112990500B CN112990500B (en) | 2022-12-23 |
Family
ID=76339071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110356039.7A Active CN112990500B (en) | 2021-04-01 | 2021-04-01 | Transformer area line loss analysis method and system based on improved weighted gray correlation analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112990500B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113762716A (en) * | 2021-07-30 | 2021-12-07 | 国网山东省电力公司营销服务中心(计量中心) | Method and system for evaluating running state of transformer area based on deep learning and attention |
| CN114113800A (en) * | 2021-12-27 | 2022-03-01 | 广西电网有限责任公司 | Online analysis method of line loss based on historical data |
| CN114118855A (en) * | 2021-12-06 | 2022-03-01 | 国网江苏省电力有限公司苏州供电分公司 | A CNN-based calculation method for the benchmark value of line loss rate in station area |
| CN114881429A (en) * | 2022-04-19 | 2022-08-09 | 国网江苏省电力有限公司南通供电分公司 | Data-driven transformer area line loss quantification method and system |
| CN116054265A (en) * | 2023-03-23 | 2023-05-02 | 国网山西省电力公司营销服务中心 | Measuring method and system for photovoltaic access capacity in station area |
| CN118133160A (en) * | 2024-05-07 | 2024-06-04 | 国网上海市电力公司 | Low-voltage active platform area line loss influence factor analysis method based on comprehensive time-varying gray correlation degree |
| CN118154003A (en) * | 2024-05-11 | 2024-06-07 | 国网山东省电力公司肥城市供电公司 | Method, device and storage medium for detecting electricity larceny of users in transformer area |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103577679A (en) * | 2012-08-10 | 2014-02-12 | 深圳市龙电电气有限公司 | Real-time computing method for theoretical line loss of low-voltage distribution room |
| CN105160416A (en) * | 2015-07-31 | 2015-12-16 | 国家电网公司 | Transformer area reasonable line loss prediction method based on principal component analysis and neural network |
| CN106356844A (en) * | 2016-09-30 | 2017-01-25 | 国网上海市电力公司 | Method for analyzing zone area line loss abnormal reason of power distribution network |
| CN107039970A (en) * | 2017-03-13 | 2017-08-11 | 广东电网有限责任公司信息中心 | Gong Biantai areas line loss per unit abnormal cause detection method and system |
| CN108647843A (en) * | 2018-03-09 | 2018-10-12 | 国网浙江省电力有限公司 | A kind of region trade power consumption influence factor judgment method based on grey correlation analysis |
| CN109389332A (en) * | 2018-12-22 | 2019-02-26 | 中铁十八局集团有限公司 | Evaluation method is configured based on combination weighting-improved grey conjunction analytic method Bridge Erector |
| CN110264112A (en) * | 2019-07-17 | 2019-09-20 | 国网河南省电力公司开封供电公司 | Bidirectional weighted gray correlation-based power distribution network reliability influence factor analysis method |
| CN110348713A (en) * | 2019-06-28 | 2019-10-18 | 广东电网有限责任公司 | A kind of platform area line loss calculation method based on association analysis and data mining |
| CN111487488A (en) * | 2020-03-24 | 2020-08-04 | 国网冀北电力有限公司电力科学研究院 | Intelligent station area outdoor transformer identification method based on grey correlation analysis |
| AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
-
2021
- 2021-04-01 CN CN202110356039.7A patent/CN112990500B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103577679A (en) * | 2012-08-10 | 2014-02-12 | 深圳市龙电电气有限公司 | Real-time computing method for theoretical line loss of low-voltage distribution room |
| CN105160416A (en) * | 2015-07-31 | 2015-12-16 | 国家电网公司 | Transformer area reasonable line loss prediction method based on principal component analysis and neural network |
| CN106356844A (en) * | 2016-09-30 | 2017-01-25 | 国网上海市电力公司 | Method for analyzing zone area line loss abnormal reason of power distribution network |
| CN107039970A (en) * | 2017-03-13 | 2017-08-11 | 广东电网有限责任公司信息中心 | Gong Biantai areas line loss per unit abnormal cause detection method and system |
| CN108647843A (en) * | 2018-03-09 | 2018-10-12 | 国网浙江省电力有限公司 | A kind of region trade power consumption influence factor judgment method based on grey correlation analysis |
| CN109389332A (en) * | 2018-12-22 | 2019-02-26 | 中铁十八局集团有限公司 | Evaluation method is configured based on combination weighting-improved grey conjunction analytic method Bridge Erector |
| CN110348713A (en) * | 2019-06-28 | 2019-10-18 | 广东电网有限责任公司 | A kind of platform area line loss calculation method based on association analysis and data mining |
| CN110264112A (en) * | 2019-07-17 | 2019-09-20 | 国网河南省电力公司开封供电公司 | Bidirectional weighted gray correlation-based power distribution network reliability influence factor analysis method |
| CN111487488A (en) * | 2020-03-24 | 2020-08-04 | 国网冀北电力有限公司电力科学研究院 | Intelligent station area outdoor transformer identification method based on grey correlation analysis |
| AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
Non-Patent Citations (4)
| Title |
|---|
| YIJING REN 等: ""Calculation Method of the Line Loss Rate in Transformer District Based on Neural Network with Optimized Input Variables"", 《2020 IEEE 3RD STUDENT CONFERENCE ON ELECTRICAL MACHINES AND SYSTEMS (SCEMS)》 * |
| 张长行: ""可再生能源并网的电网线损计算与分析"", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
| 赵允 等: ""基于灰色关联分析和IPSO-LSSVM的线损预测模型研究"", 《东北电力技术》 * |
| 赵永红等: "基于关联度变权的台区理论线损水平评估方法研究", 《电力需求侧管理》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113762716A (en) * | 2021-07-30 | 2021-12-07 | 国网山东省电力公司营销服务中心(计量中心) | Method and system for evaluating running state of transformer area based on deep learning and attention |
| CN114118855A (en) * | 2021-12-06 | 2022-03-01 | 国网江苏省电力有限公司苏州供电分公司 | A CNN-based calculation method for the benchmark value of line loss rate in station area |
| CN114113800A (en) * | 2021-12-27 | 2022-03-01 | 广西电网有限责任公司 | Online analysis method of line loss based on historical data |
| CN114113800B (en) * | 2021-12-27 | 2023-10-24 | 广西电网有限责任公司 | Line loss on-line analysis method based on historical data |
| CN114881429A (en) * | 2022-04-19 | 2022-08-09 | 国网江苏省电力有限公司南通供电分公司 | Data-driven transformer area line loss quantification method and system |
| CN114881429B (en) * | 2022-04-19 | 2024-04-02 | 国网江苏省电力有限公司南通供电分公司 | Data-driven line loss quantification method and system in Taiwan area |
| CN116054265A (en) * | 2023-03-23 | 2023-05-02 | 国网山西省电力公司营销服务中心 | Measuring method and system for photovoltaic access capacity in station area |
| CN118133160A (en) * | 2024-05-07 | 2024-06-04 | 国网上海市电力公司 | Low-voltage active platform area line loss influence factor analysis method based on comprehensive time-varying gray correlation degree |
| CN118133160B (en) * | 2024-05-07 | 2024-08-06 | 国网上海市电力公司 | Low-voltage active platform area line loss influence factor analysis method based on comprehensive time-varying gray correlation degree |
| CN118154003A (en) * | 2024-05-11 | 2024-06-07 | 国网山东省电力公司肥城市供电公司 | Method, device and storage medium for detecting electricity larceny of users in transformer area |
| CN118154003B (en) * | 2024-05-11 | 2024-07-26 | 国网山东省电力公司肥城市供电公司 | A method, device and storage medium for detecting electricity theft by users in a transformer area |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112990500B (en) | 2022-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112990500A (en) | Transformer area line loss analysis method and system based on improved weighted gray correlation analysis | |
| CN110659779B (en) | A Loss Prediction Method of Distribution System Based on Long Short-Term Memory Network | |
| CN113554466B (en) | Short-term electricity consumption prediction model construction method, prediction method and device | |
| CN112733417A (en) | Abnormal load data detection and correction method and system based on model optimization | |
| CN106779129A (en) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor | |
| CN109647899B (en) | A method for predicting power consumption of multi-specification rolling stock in the process of hot strip finishing rolling | |
| CN112446534A (en) | Construction period prediction method and device for power transmission and transformation project | |
| CN106600037B (en) | Multi-parameter auxiliary load prediction method based on principal component analysis | |
| CN111931354A (en) | Transformer top layer oil temperature prediction method based on gray-autoregressive differential moving average model | |
| CN108898259A (en) | Adaptive Evolutionary planning Methods of electric load forecasting and system based on multi-factor comprehensive | |
| CN116823008A (en) | Park energy utilization efficiency evaluation method, system, equipment and storage medium | |
| CN106127303A (en) | A kind of short-term load forecasting method towards multi-source data | |
| CN113868954B (en) | Air conditioner load prediction method and system | |
| CN113554213A (en) | Natural gas demand prediction method, system, storage medium and equipment | |
| CN105005878B (en) | A kind of comprehensive estimation method of strong intelligent grid | |
| CN111985845B (en) | Node priority optimization method of heterogeneous Spark cluster | |
| WO2025145610A1 (en) | Urban drainage, regulation and storage facility operation step-by-step decision analysis method based on risk assessment | |
| CN107239850A (en) | A kind of long-medium term power load forecasting method based on system dynamics model | |
| CN119359053B (en) | Overseas high-speed rail urban tunnel construction risk assessment method and system | |
| CN112633762A (en) | Building energy efficiency obtaining method and equipment | |
| CN109376937B (en) | Self-adaptive scheduling end-of-term water level prediction method based on ensemble empirical mode decomposition | |
| CN111311026A (en) | Runoff nonlinear prediction method considering data characteristics, model and correction | |
| CN115374938A (en) | A distribution network voltage prediction method based on XGBoost | |
| CN118523309A (en) | Medium-long-term power load combination prediction method, device and storage medium | |
| CN118644018A (en) | Reservoir flooding forecasting and dispatching method and system based on water surface line prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |