CN113723832A - Working method for carrying out quota screening aiming at digital energy distribution - Google Patents

Abstract

The invention provides a working method for carrying out credit screening aiming at digital energy distribution, which is used for accurately managing the problem of energy consumption and comprises the following steps: s1, collecting basic data of energy users to form an energy user data set, and calculating energy consumption and an energy optimization cost function; s2, screening energy according to the energy usage and the energy optimization cost function, and screening the energy user quota by setting energy screening parameters through the Bayesian network; and S3, after screening the energy user quota, distributing the energy quota to the energy user through the distribution model, and sending the distributed energy quota to the corresponding energy user.

Description

Working method for carrying out quota screening aiming at digital energy distribution

Technical Field

The invention relates to the field of big data screening, in particular to a working method for screening quota aiming at digital energy distribution.

Background

Along with the development of the economic era, energy consumption becomes the basis of the sustainable development of economy, when the production task and the energy consumption project are completed, the energy consumption main body obtains the corresponding energy amount, which belongs to the indispensable precondition, the benign development of green economy is completed by the carbon neutralization and carbon peak reaching technology, but the matching work of the energy consumption main body cannot be accurately completed by judging and screening the existing digital energy consumption amount, the energy consumption and the total amount of carbon emission cannot be accurately managed, and the technical personnel in the field are urgently needed to solve the corresponding technical problems.

Disclosure of Invention

The invention aims to at least solve the technical problems in the prior art, and particularly creatively provides a working method for screening quota aiming at digital energy distribution.

In order to achieve the above object, the present invention provides a method for screening quota for digital energy distribution, comprising the following steps:

s1, collecting basic data information of the energy user, forming an energy user data set, and calculating the energy consumption and an energy optimization cost function;

s2, screening digital energy according to the energy usage and the digital energy optimization cost function, and screening the digital energy user quota by setting digital energy screening parameters through a Bayesian network;

and S3, after the screening of the digital energy user quota is completed, distributing the energy quota to the energy user through the distribution model, and sending the distributed digital energy quota to the digital energy demand party.

Preferably, the S1 includes the following steps:

s1-1, acquiring data of users using theoretical digital energy and data of users using actual digital energy in the energy users, and respectively forming a data set of the users using theoretical digital energy and a data set of the users using actual digital energy;

s1-2, extracting the usage amount of theoretical digital energy and the theoretical digital energy optimization cost function of the theoretical digital energy user data set;

and S1-3, extracting the actual digital energy usage amount and the actual digital energy optimization cost function of the actual digital energy user data set.

Preferably, the default value T of the theoretical digital energy consumption_{Define energy}：

Wherein, T_i ^{Define energy}Representing the energy consumption of the ith theoretical digital energy user; i represents the number of theoretical digital energy users (Define energy units), and mu represents a data factor for training the theoretical digital energy consumption by carrying out data acquisition on the energy consumption of each theoretical digital energy user;

representing the average value of the energy consumption of each theoretical digital energy user; alpha represents a theoretical digital energy consumption regulation coefficient, 0<α<1, adjusting an initial default value of the theoretical digital energy source;

the cost expectation of the theoretical digital energy to each user is completed by acquiring the energy use cost of the theoretical digital energy user, and a theoretical digital energy optimization cost function G is formed_{Define energy}：

Wherein

For the consumption contribution rate of the theoretical digital energy consumption,

in order to calculate the contribution rate of the digital energy scheduling,

contribution rate, delta, for theoretical digital energy consumption₁A satisfaction factor is used for theoretical digital energy.

Preferably, the default value T for the actual digital energy usage_{Reality energy}The calculation formula of (2) is as follows:

wherein,

representing the energy consumption of the jth actual digital energy user; j represents the number of actual digital energy units (real energy units), and epsilon represents a data factor for training the actual digital energy consumption by carrying out data acquisition on the energy consumption of each actual digital energy user;

representing an average value of the energy usage amount of each actual digital energy user; beta represents the actual digital energy consumption regulation coefficient, 0<β<1, adjusting the initial default value of the actual digital energy source;

the cost expectation of the actual digital energy to each user is completed by acquiring the energy use cost of the actual digital energy user, and an actual digital energy optimization cost function G is formed_{Reality energy}：

Wherein

For the consumption contribution rate of the actual digital energy usage,

scheduling contribution rate for real digital energy，

For the actual digital energy consumption contribution, delta₂And using the satisfaction coefficient for the actual digital energy.

Preferably, the S2 includes the following steps:

s2-1, calculating conditional probability mutual information quantity of a data unit X in a theoretical digital energy user data set and a data unit Y in an actual digital energy user data set through prior probability distribution;

s2-2, constructing a naive Bayesian network of a theoretical digital energy user data set and an actual digital energy user data set, and calculating a probability function of an energy consumption attribute D;

and S2-3, calculating a probability quality function of the energy consumption attribute D, and setting an energy screening parameter H according to the Bayesian network to screen the energy user quota.

Preferably, the S2-1 includes the following steps:

the conditional probability mutual information quantity C (X)_i；Y_jI D), D is an energy consumption attribute, subscripts i and j are respectively positive integers, and the subscripts i and j are serial numbers of the data units;

wherein, E (X)_i| D) is traversal X_iConditional probability distribution obtained from the value of D, E (X)_i) For each data unit X in the theoretical digital energy user data set_iE (D) is the probability distribution of the energy consumption attribute D in the theoretical digital energy user data set, F (Y)_j| D) is traversal Y_jConditional probability distribution obtained from the value of D, F (Y)_j) For each data unit Y in the actual digital energy user data set_jF (D) is the probability distribution of the energy consumption attribute D in the actual digital energy user data set, G (X)_i,Y_jD) represents X_i，Y_jJoint probability distribution with D.

Preferably, the S2-2 includes the following steps:

constructing a naive Bayesian network of a theoretical digital energy user data set by taking an energy consumption attribute D as a father node for each data unit in a data unit X in the theoretical digital energy user data set; constructing a naive Bayesian network of the actual digital energy user data set by taking the energy consumption attribute D as a father node for each data unit in the data unit Y of the actual digital energy user data set; putting data units in a theoretical digital energy user data centralized data unit X and a practical digital energy user data centralized data unit Y into a Bayesian network one by one; if C (X)_i；Y_j| D) satisfies C (X)_i；Y_j|D)＞C(X_k；Y_j| D) (k ≠ i), then the next X will be replaced_iAnd Y_jPutting the energy user quota into a network to obtain a complete Bayesian network for screening and sequencing the energy user quota;

calculating a probability function G (D, X) of an energy consumption property D₁,...,X_i,Y₁,...,Y_j) Satisfy the following requirements

Wherein

Representing each data unit X in a theoretical digital energy user data set_iAnd each data unit Y in the actual digital energy user data set_jThe product of the conditional probabilities of (c).

Preferably, the S2-3 includes the following steps:

according to the condition of theoretical digital energy user data and actual digital energy user data, each data unit X of data units X is concentrated into theoretical digital energy user data_iAssignment and each data unit Y of data units Y in the actual digital energy user data set_jAssigning; data unit according to theoretical digital energy user data and data sheet of actual digital energy user dataAnd the elements alternately substitute the corresponding data units into the Bayesian network to calculate a probability mass function G (D, X) related to all energy consumption attributes D₁,...,X_i,Y₁,...,Y_j) Thus, the energy user quota is screened by setting an energy screening parameter H according to the Bayesian network;

eta is a correction parameter for obtaining theoretical digital energy user data, T_{Total energy}In order to achieve the overall use of energy,

as a function of the distribution of theoretical digital energy usage,

a distribution function of theoretical digital energy acquisition quantity, wherein sigma is an average value of the data of the actual digital energy user, and lambda is an energy data gradient threshold value, and the measured value Q of the data of the actual digital energy user is obtained_iAnd S_jAnd (4) carrying out data accumulation by misjudging the attenuation parameters by the energy user data.

Preferably, the S3 includes the following steps:

s3-1, acquiring energy user amount screening data through an energy screening parameter H, obtaining the use sum of predicted theoretical digital energy user data and actual digital energy user data, and pre-estimating the distribution amount of the theoretical digital energy user data and the actual digital energy user data;

s3-2, in order to intelligently distribute corresponding theoretical digital energy user data and actual digital energy user data, calculating the energy distribution quota through a distribution model, wherein the model is as follows:

by judging theoretical digital energy distribution quota q andwhether the actual digital energy distribution quota v is larger than 0 or not, U is the attenuation weight of the energy user usage, and W is a prediction function of the energy user usage; current theoretical digital energy distribution quota q_iAnd the next stage of allocating credit q_i+1(ii) a Current actual digital energy distribution quota v_jAnd the next stage of allocating the amount v_j+1。

In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:

the method comprises the steps of classifying and sorting an energy user data set, judging prior probability according to attributes of energy users, screening use data of the energy users by constructing a Bayesian network, analyzing the attributes of the energy users according to the information extraction capability of the Bayesian network on big data, and setting an allocation model to effectively allocate the energy user quota.

Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.

Drawings

The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

FIG. 1 is a general flow diagram of the present invention.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.

As shown in fig. 1, the present invention provides a working method for screening quota for digital energy distribution, which includes the following steps:

s1, collecting basic data information of the energy user, forming an energy user data set, and calculating the energy consumption and an energy optimization cost function;

s2, screening digital energy according to the energy usage and the digital energy optimization cost function, and screening the digital energy user quota by setting digital energy screening parameters through a Bayesian network;

and S3, after the screening of the digital energy user quota is completed, distributing the energy quota to the energy user through the distribution model, and sending the distributed digital energy quota to the digital energy demand party.

The S1 includes the following steps:

s1-1, acquiring data of users using theoretical digital energy and data of users using actual digital energy in the energy users, and respectively forming a data set of the users using theoretical digital energy and a data set of the users using actual digital energy;

s1-2, extracting the usage amount of theoretical digital energy and the theoretical digital energy optimization cost function of the theoretical digital energy user data set;

s1-3, extracting the actual digital energy usage amount and the actual digital energy optimization cost function of the actual digital energy user data set;

wherein, the default value T of the theoretical digital energy consumption_{Define energy}：

Wherein, T_i ^{Define energy}Representing the energy consumption of the ith theoretical digital energy user; i represents the number of theoretical digital energy users (Define energy units), and mu represents a data factor for training the theoretical digital energy consumption by carrying out data acquisition on the energy consumption of each theoretical digital energy user;

representing the average value of the energy consumption of each theoretical digital energy user; alpha represents a theoretical digital energy consumption regulation coefficient, 0<α<1, adjusting an initial default value of the theoretical digital energy source;

by obtaining theoretical digital energyThe cost of the user using the energy is completed, the cost expectation of the theoretical digital energy to each user is completed, and a theoretical digital energy optimization cost function G is formed_{Define energy}：

Wherein

For the consumption contribution rate of the theoretical digital energy consumption,

the scheduling contribution rate is used for adjusting the energy consumption regulation value of the average value of the energy consumption of each theoretical digital energy user,

contribution rate, delta, for theoretical digital energy consumption₁Using a satisfaction coefficient for theoretical digital energy; the function interval range of the use cost of the theoretical digital energy can be obtained by calculating the theoretical digital energy optimization cost function, so that preparation is made for data screening in a theoretical digital energy user data set;

default value T for actual digital energy usage_{Reality energy}The calculation formula of (2) is as follows:

wherein,

representing the energy consumption of the jth actual digital energy user; j represents the number of actual digital energy units (real energy units) by which data acquisition is performed on the energy usage of each actual digital energy userThe epsilon represents a data factor for training the actual digital energy consumption;

representing an average value of the energy usage amount of each actual digital energy user; beta represents the actual digital energy consumption regulation coefficient, 0<β<1, adjusting the initial default value of the actual digital energy source;

the cost expectation of the actual digital energy to each user is completed by acquiring the energy use cost of the actual digital energy user, and an actual digital energy optimization cost function G is formed_{Reality energy}：

Wherein

For the consumption contribution rate of the actual digital energy usage,

scheduling contribution rate for actual digital energy, the consumption contribution rate being used for adjusting the calculation of consumption value in actual digital energy usage, the scheduling contribution rate being used for adjusting the energy consumption scheduling value of the average value of energy usage by each actual digital energy consumer,

for the actual digital energy consumption contribution, delta₂Using a satisfaction coefficient for actual digital energy; the function interval range of the use cost of the actual digital energy can be obtained by calculating the actual digital energy optimization cost function, so that preparation is made for data screening in the data set of the actual digital energy user;

the S2 includes the following steps:

s2-1, data unit X in theoretical digital energy user data set and data unit in actual digital energy user data set are distributed according to prior probabilityConditional probability mutual information quantity C (X) of Y_i；Y_jI D), wherein D is an energy consumption attribute, and subscripts i and j are respectively positive integers which are serial numbers of the data units;

wherein, E (X)_i| D) is traversal X_iConditional probability distribution obtained from the value of D, E (X)_i) For each data unit X in the theoretical digital energy user data set_iE (D) is the probability distribution of the energy consumption attribute D in the theoretical digital energy user data set, F (Y)_j| D) is traversal Y_jConditional probability distribution obtained from the value of D, F (Y)_j) For each data unit Y in the actual digital energy user data set_jF (D) is the probability distribution of the energy consumption attribute D in the actual digital energy user data set, G (X)_i,Y_jD) represents X_i，Y_jWith the joint probability distribution of D,

s2-2, constructing a naive Bayesian network of the theoretical digital energy user data set by taking the energy consumption attribute D as a father node for each data unit in the data unit X in the theoretical digital energy user data set; constructing a naive Bayesian network of the actual digital energy user data set by taking the energy consumption attribute D as a father node for each data unit in the data unit Y of the actual digital energy user data set; putting data units in a theoretical digital energy user data centralized data unit X and a practical digital energy user data centralized data unit Y into a Bayesian network one by one; if C (X)_i；Y_j| D) satisfies C (X)_i；Y_j|D)＞C(X_k；Y_j| D) (k ≠ i), then the next X will be replaced_iAnd Y_jPutting the energy user quota into a network to obtain a complete Bayesian network for screening and sequencing the energy user quota;

calculating a probability function G (D, X) of an energy consumption property D₁,...,X_i,Y₁,...,Y_j) Satisfy the following requirements

Wherein

Representing each data unit X in a theoretical digital energy user data set_iAnd each data unit Y in the actual digital energy user data set_jThe product of the conditional probabilities of (a);

s2-3, according to the condition of the theoretical digital energy user data and the actual digital energy user data, concentrating each data unit X of the data units X into the theoretical digital energy user data set_iAssignment and each data unit Y of data units Y in the actual digital energy user data set_jAssigning; according to the data unit of the theoretical digital energy user data and the data unit of the actual digital energy user data, substituting the corresponding data units into the Bayesian network in turn, and calculating the probability mass function G (D, X) of all energy consumption attributes D₁,...,X_i,Y₁,...,Y_j) Thus, the energy user quota is screened by setting an energy screening parameter H according to the Bayesian network;

eta is a correction parameter for obtaining theoretical digital energy user data, T_{Total energy}In order to achieve the overall use of energy,

as a function of the distribution of theoretical digital energy usage,

a distribution function of theoretical digital energy acquisition quantity, wherein sigma is an average value of the data of the actual digital energy user, and lambda is an energy data gradient threshold value, and the measured value Q of the data of the actual digital energy user is obtained_iAnd S_jMisjudging attenuation parameters by energy user data to accumulate the data;

the S3 includes the following steps:

s3-1, acquiring energy user amount screening data through an energy screening parameter H, obtaining the use sum of predicted theoretical digital energy user data and actual digital energy user data, and pre-estimating the distribution amount of the theoretical digital energy user data and the actual digital energy user data;

s3-2, in order to intelligently distribute corresponding theoretical digital energy user data and actual digital energy user data, calculating the energy distribution quota through a distribution model, wherein the model is as follows:

by judging whether the theoretical digital energy distribution quota q and the actual digital energy distribution quota v are larger than 0, U is the attenuation weight of the usage of the energy user, and W is a prediction function of the usage of the energy user; current theoretical digital energy distribution quota q_iAnd the next stage of allocating credit q_i+1(ii) a Current actual digital energy distribution quota v_jAnd the next stage of allocating the amount v_j+1。

While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims (9)

1. A working method for screening quota aiming at digital energy distribution is characterized by comprising the following steps:

s1, collecting basic data information of the energy user, forming an energy user data set, and calculating the energy consumption and an energy optimization cost function;

s2, screening digital energy according to the energy usage and the digital energy optimization cost function, and screening the digital energy user quota by setting digital energy screening parameters through a Bayesian network;

and S3, after the screening of the digital energy user quota is completed, distributing the energy quota to the energy user through the distribution model, and sending the distributed digital energy quota to the digital energy demand party.

2. The working method of credit screening for digital energy distribution as claimed in claim 1,

the S1 includes the following steps:

s1-1, acquiring data of users using theoretical digital energy and data of users using actual digital energy in the energy users, and respectively forming a data set of the users using theoretical digital energy and a data set of the users using actual digital energy;

s1-2, extracting the usage amount of theoretical digital energy and the theoretical digital energy optimization cost function of the theoretical digital energy user data set;

and S1-3, extracting the actual digital energy usage amount and the actual digital energy optimization cost function of the actual digital energy user data set.

3. The working method of credit screening for digital energy distribution as claimed in claim 2, wherein the default value T of the theoretical digital energy usage amount_Defineenergy：

Wherein, T_i ^DefineenergyRepresenting the energy consumption of the ith theoretical digital energy user; i represents the number of theoretical digital energy users (Define energy units), and mu represents a data factor for training the theoretical digital energy consumption by carrying out data acquisition on the energy consumption of each theoretical digital energy user;

represents each training sessionThe average value of the energy consumption of each theoretical digital energy user; alpha represents a theoretical digital energy consumption regulation coefficient, 0<α<1, adjusting an initial default value of the theoretical digital energy source;

the cost expectation of the theoretical digital energy to each user is completed by acquiring the energy use cost of the theoretical digital energy user, and a theoretical digital energy optimization cost function G is formed_Defineenergy：

Wherein

For the consumption contribution rate of the theoretical digital energy consumption,

in order to calculate the contribution rate of the digital energy scheduling,

contribution rate, delta, for theoretical digital energy consumption₁A satisfaction factor is used for theoretical digital energy.

4. The working method of credit screening for digital energy distribution as claimed in claim 2, wherein the default value T for the actual digital energy usage is_{Realityenergy}The calculation formula of (2) is as follows:

wherein,

representing the energy consumption of the jth actual digital energy user; j denotes the actual digital energy unit (Reality ene)rgy unit), and epsilon represents a data factor for training the actual digital energy consumption through data acquisition of the energy consumption of each actual digital energy user;

representing an average value of the energy usage amount of each actual digital energy user; beta represents the actual digital energy consumption regulation coefficient, 0<β<1, adjusting the initial default value of the actual digital energy source;

the cost expectation of the actual digital energy to each user is completed by acquiring the energy use cost of the actual digital energy user, and an actual digital energy optimization cost function G is formed_{Realityenergy}：

Wherein

For the consumption contribution rate of the actual digital energy usage,

for the dispatch contribution rate of the actual digital energy,

for the actual digital energy consumption contribution, delta₂And using the satisfaction coefficient for the actual digital energy.

5. The working method of credit screening for digital energy distribution according to claim 1, wherein said S2 comprises the following steps:

s2-1, calculating conditional probability mutual information quantity of a data unit X in a theoretical digital energy user data set and a data unit Y in an actual digital energy user data set through prior probability distribution;

s2-2, constructing a naive Bayesian network of a theoretical digital energy user data set and an actual digital energy user data set, and calculating a probability function of an energy consumption attribute D;

and S2-3, calculating a probability quality function of the energy consumption attribute D, and setting an energy screening parameter H according to the Bayesian network to screen the energy user quota.

6. The working method of credit screening for digital energy distribution according to claim 5, wherein said S2-1 includes the following steps:

the conditional probability mutual information quantity C (X)_i；Y_jI D), D is an energy consumption attribute, subscripts i and j are respectively positive integers, and the subscripts i and j are serial numbers of the data units;

wherein, E (X)_i| D) is traversal X_iConditional probability distribution obtained from the value of D, E (X)_i) For each data unit X in the theoretical digital energy user data set_iE (D) is the probability distribution of the energy consumption attribute D in the theoretical digital energy user data set, F (Y)_j| D) is traversal Y_jConditional probability distribution obtained from the value of D, F (Y)_j) For each data unit Y in the actual digital energy user data set_jF (D) is the probability distribution of the energy consumption attribute D in the actual digital energy user data set, G (X)_i,Y_jD) represents X_i，Y_jJoint probability distribution with D.

7. The working method of credit screening for digital energy distribution according to claim 5, wherein said S2-2 includes the following steps:

constructing a theoretical digital energy user data set plain by taking an energy consumption attribute D as a father node for each data unit in a data unit X in the theoretical digital energy user data setA Bayesian network; constructing a naive Bayesian network of the actual digital energy user data set by taking the energy consumption attribute D as a father node for each data unit in the data unit Y of the actual digital energy user data set; putting data units in a theoretical digital energy user data centralized data unit X and a practical digital energy user data centralized data unit Y into a Bayesian network one by one; if C (X)_i；Y_j| D) satisfies C (X)_i；Y_j|D)＞C(X_k；Y_j| D) (k ≠ i), then the next X will be replaced_iAnd Y_jPutting the energy user quota into a network to obtain a complete Bayesian network for screening and sequencing the energy user quota;

calculating a probability function G (D, X) of an energy consumption property D₁,...,X_i,Y₁,...,Y_j) Satisfy the following requirements

Wherein

Representing each data unit X in a theoretical digital energy user data set_iAnd each data unit Y in the actual digital energy user data set_jThe product of the conditional probabilities of (c).

8. The working method of credit screening for digital energy distribution according to claim 5, wherein said S2-3 includes the following steps:

according to the condition of theoretical digital energy user data and actual digital energy user data, each data unit X of data units X is concentrated into theoretical digital energy user data_iAssignment and each data unit Y of data units Y in the actual digital energy user data set_jAssigning; alternately substituting corresponding data units into the Bayesian network according to the data units of the theoretical digital energy user data and the data units of the actual digital energy user data, and calculating the data units of all the data unitsProbability mass function G (D, X) of energy consumption attribute D₁,...,X_i,Y₁,...,Y_j) Thus, the energy user quota is screened by setting an energy screening parameter H according to the Bayesian network;

eta is a correction parameter for obtaining theoretical digital energy user data, T_TotalenergyIn order to achieve the overall use of energy,

as a function of the distribution of theoretical digital energy usage,

a distribution function of theoretical digital energy acquisition quantity, wherein sigma is an average value of the data of the actual digital energy user, and lambda is an energy data gradient threshold value, and the measured value Q of the data of the actual digital energy user is obtained_iAnd S_jAnd (4) carrying out data accumulation by misjudging the attenuation parameters by the energy user data.

9. The working method of credit screening for digital energy distribution according to claim 1, wherein said S3 comprises the following steps:

s3-1, acquiring energy user amount screening data through an energy screening parameter H, obtaining the use sum of predicted theoretical digital energy user data and actual digital energy user data, and pre-estimating the distribution amount of the theoretical digital energy user data and the actual digital energy user data;

s3-2, in order to intelligently distribute corresponding theoretical digital energy user data and actual digital energy user data, calculating the energy distribution quota through a distribution model, wherein the model is as follows:

by judging whether the theoretical digital energy distribution quota q and the actual digital energy distribution quota v are larger than 0, U is the attenuation weight of the usage of the energy user, and W is a prediction function of the usage of the energy user; current theoretical digital energy distribution quota q_iAnd the next stage of allocating credit q_i+1(ii) a Current actual digital energy distribution quota v_jAnd the next stage of allocating the amount v_j+1。

Images