CN105741143A - Load characteristic and cluster analysis based electric power commodity pricing model establishment method - Google Patents

Abstract

The invention discloses a load characteristic and cluster analysis based electric power commodity pricing model establishment method. Compared with the prior art, the method overcomes the defect that load characteristics of users cannot be fully reflected in an electric power commodity pricing method. The method comprises the following steps of acquiring and preprocessing original data, obtaining basic load data, and preprocessing the basic load data; clustering the preprocessed load data to obtain a basic load curve; and extracting load characteristic indexes in the basic load curve and constructing an electric power commodity pricing model. The invention provides a massive user load data analysis based modeling method; and through the modeling method, electric power commodities are accurately priced.

Description

A Pricing Model Establishment Method for Electricity Commodities Based on Load Characteristics and Cluster Analysis

technical field

The invention relates to the technical field of data analysis, in particular to a method for establishing an electric commodity pricing model based on load characteristics and cluster analysis thereof.

Background technique

At present, the analysis and processing of power load data is very common. Through the application of statistical methods and the powerful data processing functions of computers, the load data can be analyzed, processed and fitted to solve the manual calculation process. It is widely used in the restoration of load data. And forecasting, load curve modeling and other fields. However, the relationship between the user's load data and the electricity price has not been fully considered, and the impact of load characteristics on the user's electricity cost has not been fully reflected in the price of electricity commodities. In an open electricity market, the user's electricity price is closely related to its load characteristics. How to use massive power user load data to reasonably formulate the price of electricity commodities, which not only compensates for costs and realizes benefits, but also realizes appropriate guidance for users' electricity consumption. Has become an urgent technical problem to be solved.

Contents of the invention

The purpose of the present invention is to solve the defect that the electricity commodity pricing method in the prior art cannot fully reflect user load characteristics, and provide a method for establishing a power commodity pricing model based on load characteristics and cluster analysis to solve the above problems.

In order to achieve the above object, the technical scheme of the present invention is as follows:

A method for establishing a power commodity pricing model based on load characteristics and cluster analysis thereof, comprising the following steps:

Raw data collection and preprocessing, obtaining base load data, and preprocessing base load data;

Cluster the preprocessed load data to obtain the base load curve;

The load characteristic index in the basic load curve is extracted, and the pricing model of electricity commodity is constructed.

The raw data acquisition and preprocessing include the following steps:

Collect basic load data, and represent individual users as n, n∈N, where N represents a set of several users;

Set 96 time collection points i, i ∈ {1, 2, ..., 96};

Express the load data of user n at the i-th collection point as And y _i ⁿ ≥0.

The clustering of the preprocessed load data includes the following steps:

Store N input vectors in the input matrix to construct a matrix Σ of N×96, where N represents the number of users, and 96 represents 96 collection points in one day;

Initialize the weight ω _ij , its value range is [0,1] and ω _ij is not the same;

Get the weight vector set G, select the vector from the N input vectors in the matrix Σ Provide to the network input layer, calculate the weight vector ω _ij (t) and iteratively process the matrix Σ until the N input vectors are all selected and processed, and obtain the weight vector set G; it specifically includes the following steps:

The selected vector in the matrix Σ and the initial weight ω _ij are sent to the network input layer;

Calculate the distance d _j from all neurons in the output layer of the network to the input vector, and the calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>$

Where t is the current number of updates;

Selection of competition winning neurons Neuron i is the winning neuron;

For all neurons, adjust the weight vector between it and the neurons in the neighborhood, the update formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&eta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

Among them, η(t) represents the learning rate (0<η(t)<1) and monotonically decreases, and h _j,i(x) (t) represents the neighborhood function of the winning neuron;

Iteratively process the updated weights and newly selected vectors, and continue to calculate the distance d _j and weight ω _ij of all neurons in the output layer of the network to the input vectors, until all N input vectors are selected and processed ;

Obtain the weight vector set G{ω _ij |j=(1, 2,...,m)}, where m represents the number of clusters;

Taking a single weight vector ω _ij (t) as the cluster center of all samples with it as the winning neuron, the samples are divided into m categories, expressed as {G1, G2, ..., Gm}; the daily load of the cluster center The curve is the base load curve for this class.

The extraction of the load characteristic index in the base load curve includes the following steps:

Calculate the load rate LF ^G of various user daily load curves, the calculation formula is as follows:

in: Indicates the collection point data of this type of base load curve, LF ^G indicates the load rate of this category, T indicates the statistical period, is the average load, is the maximum load;

Calculate the basic electricity price p _i for each time period;

Build pricing models based on load characteristics.

It also includes repairing the missing value in the base load data; using the k-period central moving average method to repair the missing value; setting is the repair value of the missing data y _i , the calculation formula is as follows:

${\mathrm{the\; y}}_i^{*}=\frac{1}{k-1}({\mathrm{the\; y}}_{i-\frac{k-1}{2}}+...{\mathrm{the\; y}}_{i-1}+{\mathrm{the\; y}}_{i+1}+...+{\mathrm{the\; y}}_{i+\frac{k-1}{2}}),$ k is an odd number;

${\mathrm{the\; y}}_i^{*}=\frac{1}{k-1}(\frac{1}{2}{\mathrm{the\; y}}_{i-\frac{k}{2}}+{\mathrm{the\; y}}_{i-\frac{k}{2}+1}+{...\mathrm{the\; y}}_{i-1}+{\mathrm{the\; y}}_{i+1}+...+{\mathrm{the\; y}}_{i+\frac{k}{2}-1}+\frac{1}{2}{\mathrm{the\; y}}_{i+\frac{k}{2}}),$ k is an even number;

Among them, k represents the number of samples used for patching.

The calculation of the basic electricity price p _i for each time period includes the following steps:

The basic average electricity price level p _ave is determined by the cost and expected income, and its relationship with the time-of-use electricity price is as follows:

Among them: p _i represents the electricity price of each collection period, y _i is the system load data, that is, the sum of all user loads at collection point i, and h is the number of collection points;

Assuming that the power price and load data in each collection period have a linear relationship, then

p _i = ay _i + b, i∈{1, 2, ..., h}, where p _i represents the electricity price of each collection period, y _i represents the system load data, and a and b are the parameters of the linear equation ;

Calculate the ratio δ of the peak-to-valley electricity price, and its calculation formula is as follows:

$\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; ay</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}}{{\mathrm{<span\; class="notranslate">\; ay</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; ;</span>$

Calculate the parameters a and b of the linear equation, and the calculation formula is as follows:

$\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; v</span>}\mathrm{<span\; class="notranslate">\; e</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&delta;y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ;</span>$

$\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&delta;y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ;</span>$

Substitute the parameters a and b into the calculation of p _i again, and take p _i as the basic electricity price for each collection period.

The described construction of a pricing model based on load characteristics includes the following steps:

Calculation of additional electricity price based on load rate Its calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; \&lsqb;</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; min</span>}}}{{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; min</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; v</span>}\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; ,</span>$

Among them: G represents the category of users, LF ^max represents the maximum load rate among all categories, LF ^min represents the minimum load rate among all categories, c ₁ represents the ratio of the electricity price reward to the basic electricity price _{pa ave} when the load rate is LF ^max , c ₂ represents the proportion of the electricity price penalty relative to the basic electricity price p _ave when the load rate is LF ^min ;

To standardize the processing load data, divide the data of each collection point by the average load of the day, and the calculation formula is as follows:

Among them, y _ave is the average value of the data of 96 collection points;

Calculate the simultaneity rate surcharge of a single load during peak hours Its calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&lsqb;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}$

in: Indicates the standardized collection point data of a certain category of users, _zi represents the standardized collection point data of the system, d ₁ and d ₂ are the restriction coefficients, d ₁ = z _max -1, d ₂ = -2(z _max -1), z _max indicates the maximum value in the system standardized collection point data, and μ is the deviation coefficient;

Calculate the simultaneous rate additional electricity price during a single load valley period Its calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; \&lsqb;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}$

Among them, d ₃ and d ₄ are also restriction coefficients, d ₃ =1-z _min , d ₄ =-2(1-z _min ), z _min represents the minimum value in the system standardized collection point data, and μ is the deviation coefficient;

To calculate various types of load electricity prices ^PG , the calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>$

Where i is φ(i)=1 during the peak period, Φ(i)=0 when i is the valley period; θ(i)=0 when i is the peak period, and θ(i)=1 when i is the valley period.

Beneficial effect

Compared with the prior art, the present invention provides a method for establishing an electric commodity pricing model based on load characteristics and its cluster analysis, which provides a modeling method based on the analysis of massive user load data, and realizes electric commodity pricing through its modeling method. accurate pricing. By extracting the load characteristic index of the basic load curve, the construction of a power pricing system that fully reflects the user's load characteristics can be realized. On the basis of considering economic benefits, at the same time guide users to use electricity rationally and scientifically, which can more reasonably reflect the cost of electricity consumption for users , to achieve efficient demand management.

Description of drawings

Fig. 1 is a flow chart of the method of the present invention.

detailed description

In order to have a further understanding and understanding of the structural features of the present invention and the achieved effects, the preferred embodiments and accompanying drawings are used for a detailed description, as follows:

As shown in Figure 1, a kind of electric commodity pricing model establishment method based on load characteristic and its cluster analysis described in the present invention, comprises the following steps:

The first step is raw data collection and preprocessing. The basic load data is obtained from the user's electricity consumption information collection system, and the basic load data is preprocessed to provide data support for the construction of a power commodity pricing model. Data preprocessing is the necessary premise and basis for all the operation process of obtaining information through data analysis and establishing models. On the basis of ensuring that the original data used are authentic, reliable and representative, through appropriate statistical analysis and information extraction, the final A power commodity pricing model that can properly reflect user load characteristics is obtained. It specifically includes the following steps:

(1) Collect basic load data, and represent individual users as n, n∈N, where N represents several sets of users. Considering the wide range and types of users here, in order to increase the versatility, so that it can target multiple users, it is necessary to cluster and divide into several categories, and then price each category separately to form a pricing model. When used, the parameters of different categories are different, and different electricity prices will be generated after input into the pricing model.

(2) Set 96 time collection points i, i ∈ {1, 2, ..., 96}.

(3) Express the load data of user n at the i-th collection point as and

The above data can be obtained from relevant organizations. For example, the load data of power users in a certain province can be obtained from the power consumption information collection system of the metering center of the power grid company in this province. The similarity between the daily load curves of users is very high, so it is representative to use the daily load curves for analysis here, which can reflect the load characteristics of users. It is easier to explain the electricity commodity pricing model constructed from this in the following steps. From a technical point of view, in this step, other data can also be used.

Although the collection rate of the user's electricity consumption information collection system is already very high, the collection of individual data is always inevitable. Considering that the lack of individual data will destroy the data structure and make it difficult to carry out subsequent data analysis, a method for repairing missing values in the base load data is also provided here, using the k-period central moving average method to repair missing values .

Assume is the repair value of the missing data y _i , the calculation formula is as follows:

${\mathrm{the\; y}}_i^{*}=\frac{1}{k-1}({\mathrm{the\; y}}_{i-\frac{k-1}{2}}+...{\mathrm{the\; y}}_{i-1}+{\mathrm{the\; y}}_{i+1}+...+{\mathrm{the\; y}}_{i+\frac{k-1}{2}}),$ k is an odd number;

${\mathrm{the\; y}}_i^{*}=\frac{1}{k-1}(\frac{1}{2}{\mathrm{the\; y}}_{i-\frac{k}{2}}+{\mathrm{the\; y}}_{i-\frac{k}{2}+1}+{...\mathrm{the\; y}}_{i-1}+{\mathrm{the\; y}}_{i+1}+...+{\mathrm{the\; y}}_{i+\frac{k}{2}-1}+\frac{1}{2}{\mathrm{the\; y}}_{i+\frac{k}{2}}),$ k is an even number;

Among them, k represents the number of samples used for repairing, which can select the smallest divisor other than 1 and 2 of the total number of collection points.

After patching, the user's daily load curve for analysis is obtained, which is expressed as a time series:

$<span\; class="notranslate">\; \{</span>\begin{array}{cccc}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; ,</span>& {\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; ,</span>& <span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>& {\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 96</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}\end{array}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; .</span>$

The second step is to cluster the preprocessed load data to obtain the base load curve. The specific steps are as follows:

(1) Store N input vectors in the input matrix to construct a matrix Σ of N×96, where N represents the number of users, and 96 represents 96 collection points in a day. Here, the time series of multiple user daily load curves are fused into a matrix for subsequent cluster analysis.

(2) Initialize the weight ω _ij , whose value range is [0,1] and ω _ij is different.

(3) Get the weight vector set G. Randomly select vectors from the N input vectors in matrix Σ Provide it to the network input layer, calculate the weight vector ω _ij (t) and iteratively process the matrix Σ until all N input vectors are selected and processed, and obtain the weight vector set G. The specific steps are as follows:

A. The selected vector in the matrix Σ and the initial weight ω _ij are sent to the network input layer.

B. Calculate the distance d _j from all neurons in the output layer of the network to the input vector, the calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>$

Where t is the current number of updates;

According to the principle that the winner is king, select the winning neuron i(x)=mind _j , and neuron i is the winning neuron.

C. For all neurons, adjust the weight vector between it and the neurons in the neighborhood. The update formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&eta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

Among them, η(t) represents the learning rate (0<η(t)<1) and is monotonously decreasing, h _j,i(x) (t) represents the neighborhood function of the winning neuron, and the GAUSS domain function can be used.

D. Perform iterative processing on the updated weights and newly selected vectors, continue to calculate the distance d _j from all neurons in the network output layer to the input vector and update the weight ω _ij until all N input vectors are selected It must be finished.

E. Obtain a weight vector set G{ω _ij |j=(1, 2, . . . , m)}, where m represents the number of clusters. From the perspective of the training process, the weight vector of the output neuron gradually approaches the winning neuron. Then the weight vector set G{ω _ij |j=(1, 2,...,m)} is a description of all the samples in the training sample set, and a single weight vector can be regarded as all samples with it as the winning neuron The cluster center of , using the cluster center as the base load curve of this class.

(4) Take a single weight vector ω _ij (t) as the clustering center of all samples with it as the winning neuron, and the samples are divided into m categories, expressed as {G1, G2, ..., Gm}; the clustering center The daily load curve is the base load curve for this class.

The third step is to extract the load characteristic index in the base load curve and construct the pricing model of electricity commodity. It specifically includes the following steps:

(1) Calculate the load rate LF ^G of the daily load curves of various users. The pricing model of this method is based on the user's load characteristics, and considers the influence of different characteristics of electricity consumption behavior on the price of electricity commodities. The content of load characteristics is very rich. In 1989, the former Ministry of Energy proposed 14 indicators to describe load characteristics, including load curve and load rate, in the "Interpretation of Statistical Indicators of Electric Power Industry Production". In early 2001, the former State Power Corporation supplemented and revised the above explanation, adding the peak-to-valley difference rate indicators, which are commonly used and standardized load characteristic indicators. Among them, the load rate is the most important indicator. In the "Power Pricing Policy" published by the World Bank, the pricing of electricity commodities is based on the load rate as the only standard. So first calculate the load rate of the user's daily load curve. The general definition of Load Factor refers to the ratio of the average load to the maximum load within a specified period of time, where the maximum load refers to the maximum load recorded during the statistical period, and the average load refers to the average instantaneous load during the statistical period. value, that is, the time-series average of the load time sequence, and its calculation formula is as follows:

in: Indicates the collection point data of this type of base load curve, LF ^G indicates the load rate of this category, T indicates the statistical period, is the average load, is the maximum load.

(2) Calculate the basic electricity price p _i in each time period, the specific steps are as follows:

A. Determine the basic average electricity price level p _ave based on the cost and expected income, and its relationship with the time-of-use electricity price is as follows:

Among them: p _i represents the electricity price of each collection period, y _i is the system load data, that is, the sum of all user loads at collection point i, and h is the number of collection points.

B. Based on factors such as reasonable electricity purchase price, electricity distribution price, electricity sales cost, and operating profit, the electricity price level should be determined, and p _i should be determined by _pave in combination with system load characteristics. Assuming that the power price and load data in each collection period have a linear relationship, then

p _i = ay _i + b, i∈{1, 2, ..., h}, where p _i represents the electricity price of each collection period, y _i represents the system load data, and a and b are the parameters of the linear equation .

C. Calculate the ratio δ of the peak-to-valley electricity price, the calculation formula is as follows:

$\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; ay</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}}{{\mathrm{<span\; class="notranslate">\; ay</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; .</span>$

D. Calculate the parameters a and b of the linear equation, and the calculation formula is as follows:

$\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; v</span>}\mathrm{<span\; class="notranslate">\; e</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&delta;y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ;</span>$

$\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&delta;y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; .</span>$

E. Re-substituting parameters a and b into the calculation of p _i , taking p _i as the basic electricity price for each collection period.

(3) Construct a pricing model based on load characteristics, that is, obtain the calculation formula of the electricity price p ^G of each group of loads, and then complete the establishment of the pricing model. It specifically includes the following steps:

A. Calculate the additional electricity price based on the load rate Its calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; \&lsqb;</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; min</span>}}}{{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; min</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; v</span>}\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; ,</span>$

Among them: G represents the category of users, LF ^max represents the maximum load rate among all categories, LF ^min represents the minimum load rate among all categories, c ₁ represents the ratio of the electricity price reward to the basic electricity price _{pa ave} when the load rate is LF ^max , c ₂ represents the proportion of the electricity price penalty relative to the basic electricity price p _ave when the load rate is LF ^min .

B. Standardize the load data, and divide the data of each collection point by the average load of the day. Here, in order to eliminate the impact of different dimensions on the data, the data of the collection point is usually divided by the maximum value, so that it is convenient to distinguish between peaks and valleys, that is, when the standardized data is greater than 1, it is the peak period, and if it is less than 1, it is the valley period. Consider the additional price of electricity at the same time . Its calculation formula is as follows:

Among them, y _ave is the average value of the data of 96 collection points.

C. Calculate the simultaneous rate additional electricity price of a single load peak period Its calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&lsqb;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}$

in: Indicates the standardized collection point data of a certain category of users, _zi represents the standardized collection point data of the system, d ₁ and d ₂ are the restriction coefficients, d ₁ = z _max -1, d ₂ = -2(z _max -1), z _max represents the maximum value in the system standardized collection point data, and μ is the deviation coefficient.

D. Calculate the simultaneous rate additional electricity price during a single load low period Its calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; \&lsqb;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}$

Among them, d ₃ and d ₄ are also restriction coefficients, d ₃ =1-z _min , d ₄ =-2(1-z _min ), z _min represents the minimum value in the system standardized collection point data, and μ is the deviation coefficient.

E. Calculate various types of load electricity prices P ^G , the calculation formula is as follows:

${\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>$

Where i is Φ(i)=1 during the peak period, Φ(i)=0 when i is the valley period; θ(i)=0 when i is the peak period, and θ(i)=1 when i is the valley period.

The basic principles, main features and advantages of the present invention have been shown and described above. Those skilled in the art should understand that the present invention is not limited by the above-mentioned embodiments. What are described in the above-mentioned embodiments and the description are only the principles of the present invention. Variations and improvements, which fall within the scope of the claimed invention. The scope of protection required by the present invention is defined by the appended claims and their equivalents.

Claims (7)

1. A method for establishing a power commodity pricing model based on load characteristics and cluster analysis thereof, characterized in that, comprising the following steps: 11) Raw data collection and preprocessing, obtaining base load data, and preprocessing base load data; 12) Clustering the preprocessed load data to obtain the base load curve; 13) Extract the load characteristic index in the base load curve, and construct the pricing model of electricity commodity. 2. a kind of electric commodity pricing model establishment method based on load characteristics and cluster analysis thereof according to claim 1, is characterized in that, described raw data collection and preprocessing comprise the following steps: 21) Collect basic load data, and represent individual users as n, n∈N, where N represents a set of several users; 22) Set 96 time collection points i, i∈{1, 2, ..., 96}; 23) Express the load data of user n at the i-th collection point as and 3. A method for establishing a power commodity pricing model based on load characteristics and cluster analysis thereof according to claim 1, wherein said clustering the preprocessed load data comprises the following steps: 31) Store N input vectors in the input matrix to construct a matrix Σ of N×96, where N represents the number of users, and 96 represents 96 collection points in one day; 32) Initialize the weight ω _ij , whose value range is [0,1] and ω _ij are all different; 33) Obtain the weight vector set G, select the vector from the N input vectors in the matrix Σ Provide to the network input layer, calculate the weight vector ω _ij (t) and iteratively process the matrix Σ until the N input vectors are all selected and processed, and obtain the weight vector set G; it specifically includes the following steps: 331) the selected vector in the matrix Σ and the initial weight ω _ij are sent to the network input layer; 332) Calculate the distance d _j of all neurons in the network output layer to the input vector, and the calculation formula is as follows: ${\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>$ Where t is the current number of updates; Select the competitive winning neuron i(x)=mind _j , neuron i is the winning neuron; 333) For all neurons, adjust the weight vectors between it and the neurons in the neighborhood, and its update formula is as follows: ${\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&eta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$ Among them, η(t) represents the learning rate (0<η(t)<1) and monotonically decreases, _{hj, i(x)} (t) represents the neighborhood function of the winning neuron; 334) Iteratively process the updated weights and newly selected vectors, and continue to calculate the distance d _j and weight ω _ij of all neurons in the output layer of the network to the input vectors, until all N input vectors are selected processed; 335) Obtain a weight vector set G{ω _ij |j=(1, 2, ..., m)}, where m represents the number of clusters; 34) Take a single weight vector ω _ij (t) as the clustering center of all samples with it as the winning neuron, and the samples are divided into m categories, expressed as {G1, G2, ..., Gm}; The daily load curve is the base load curve for this class. 4. A method for establishing a pricing model for electricity commodities based on load characteristics and cluster analysis thereof according to claim 1, wherein said extraction of the load characteristic index in the base load curve comprises the following steps: 41) Calculate the load rate LF ^G of the daily load curves of various users, and the calculation formula is as follows: in: Indicates the collection point data of this type of base load curve, LF ^G indicates the load rate of this category, T indicates the statistical period, is the average load, is the maximum load; 42) Calculate the basic electricity price p _i in each time period; 43) Build a pricing model based on load characteristics. 5. A method for establishing a pricing model for electric power commodities based on load characteristics and cluster analysis thereof according to claim 2, characterized in that: it also includes repairing missing values in the base load data; adopting the k-period central moving average method to carry out Repair of missing values; set is the repair value of the missing data y _i , the calculation formula is as follows: ${\mathrm{the\; y}}_i^{*}=\frac{1}{k-1}({\mathrm{the\; y}}_{i-\frac{k-1}{2}}+...{\mathrm{the\; y}}_{i-1}+{\mathrm{the\; y}}_{i+1}+...+{\mathrm{the\; y}}_{i+\frac{k-1}{2}}),$ k is an odd number; ${\mathrm{the\; y}}_i^{*}=\frac{1}{k-1}(\frac{1}{2}{\mathrm{the\; y}}_{i-\frac{k}{2}}+{\mathrm{the\; y}}_{i-\frac{k}{2}+1}+...{\mathrm{the\; y}}_{i-1}+{\mathrm{the\; y}}_{i+1}+...+{\mathrm{the\; y}}_{i+\frac{k}{2}-1}+\frac{1}{2}{\mathrm{the\; y}}_{i+\frac{k}{2}}),$ k is an even number; Among them, k represents the number of samples used for patching. 6. A kind of electric commodity pricing model establishment method based on load characteristic and cluster analysis thereof according to claim 4, it is characterized in that, the basic electricity price p of described calculation each time _period comprises the following steps: 61) The basic average electricity price level p _ave is determined from the cost and expected income, and its relationship with the time-of-use electricity price is as follows: Among them: p _i represents the electricity price of each collection period, y _i is the system load data, that is, the sum of all user loads at collection point i, and h is the number of collection points; 62) Assuming that the power price and load data in each collection period are in a linear relationship, then p _i = ay _i + b, i∈{1, 2, ..., h}, where p _i represents the electricity price of each collection period, y _i represents the system load data, and a and b are the parameters of the linear equation ; 63) Calculate the ratio δ of the peak-to-valley electricity price, the calculation formula is as follows: $\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; ay</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}}{{\mathrm{<span\; class="notranslate">\; ay</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; ;</span>$ 64) Calculate the parameters a and b of the linear equation, and its calculation formula is as follows: $\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; v</span>}\mathrm{<span\; class="notranslate">\; e</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&delta;y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ;</span>$ $\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&delta;y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ;</span>$ 65) Substitute the parameters a and b into the calculation of p _i again, and use p _i as the basic electricity price for each collection period. 7. A method for establishing a pricing model for electricity commodities based on load characteristics and cluster analysis thereof according to claim 4, wherein said construction of a pricing model based on load characteristics comprises the following steps: 71) Calculate the additional electricity price based on the load rate Its calculation formula is as follows: ${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; \&lsqb;</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; min</span>}}}{{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; LF</span>}}^{\mathrm{<span\; class="notranslate">\; min</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; v</span>}\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; ,</span>$ Among them: G represents the category of users, LF ^max represents the maximum load rate among all categories, LF ^min represents the minimum load rate among all categories, c ₁ represents the ratio of the electricity price reward to the basic electricity price _{pa ave} when the load rate is LF ^max , c ₂ represents the proportion of the electricity price penalty relative to the basic electricity price p _ave when the load rate is LF ^min ; 72) To standardize the processing load data, divide the data of each collection point by the average load of the day, and the calculation formula is as follows: Among them, y _ave is the average value of the data of 96 collection points; 73) Calculate the simultaneous rate additional electricity price of a single load peak period Its calculation formula is as follows: ${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; \&lsqb;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}$ in: Indicates the standardized collection point data of a certain category of users, _zi represents the standardized collection point data of the system, d ₁ and d ₂ are the restriction coefficients, d ₁ = z _max -1, d ₂ = -2(z _max -1), z _max indicates the maximum value in the system standardized collection point data, and μ is the deviation coefficient; 74) Calculate the simultaneous rate additional electricity price during a single load low period Its calculation formula is as follows: ${\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; \&lsqb;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}$ Among them, d ₃ and d ₄ are also restriction coefficients, d ₃ =1-z _min , d ₄ =-2(1-z _min ), z _min represents the minimum value in the system standardized collection point data, and μ is the deviation coefficient; 75) Calculate various types of load electricity prices p ^G , the calculation formula is as follows: ${\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Delta;p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>$ Where i is Φ(i)=1 during the peak period, Φ(i)=0 when i is the valley period; θ(i)=0 when i is the peak period, and θ(i)=1 when i is the valley period.