CN114358474A - Typical multi-energy user model building method - Google Patents
Typical multi-energy user model building method Download PDFInfo
- Publication number
- CN114358474A CN114358474A CN202111396523.9A CN202111396523A CN114358474A CN 114358474 A CN114358474 A CN 114358474A CN 202111396523 A CN202111396523 A CN 202111396523A CN 114358474 A CN114358474 A CN 114358474A
- Authority
- CN
- China
- Prior art keywords
- energy
- user
- label
- users
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000005265 energy consumption Methods 0.000 claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 20
- 238000013210 evaluation model Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 17
- 238000011156 evaluation Methods 0.000 claims description 15
- 238000007781 pre-processing Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 10
- 230000005611 electricity Effects 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- 238000011161 development Methods 0.000 claims description 5
- 230000008713 feedback mechanism Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000011160 research Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 230000035945 sensitivity Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000006399 behavior Effects 0.000 abstract description 10
- 238000004458 analytical method Methods 0.000 abstract description 6
- 238000013507 mapping Methods 0.000 abstract description 3
- 238000002372 labelling Methods 0.000 abstract 1
- 238000005065 mining Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Educational Administration (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Public Health (AREA)
- Computational Linguistics (AREA)
- Water Supply & Treatment (AREA)
- Operations Research (AREA)
- Primary Health Care (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Quality & Reliability (AREA)
- Molecular Biology (AREA)
- Game Theory and Decision Science (AREA)
- Probability & Statistics with Applications (AREA)
- Power Engineering (AREA)
Abstract
The invention discloses a typical multi-energy user model building method, which overcomes the problems that the prior art lacks multi-energy characteristic portrait and energy demand prediction of typical multi-energy users, does not consider uncertainty of user behaviors and has limitation in an analysis method, and comprises the following steps: extracting and mining data of historical energy consumption of typical multi-energy users; collecting parameters, labeling user characteristics, constructing a user portrait and establishing a prediction model; according to the invention, the subjective characteristics of the user and the market objective environment factors are comprehensively considered, and a reasonable and effective user evaluation model is established; establishing a user image through the label to realize the mapping from the energy utilization characteristics to the user label; and extracting historical energy consumption parameters of typical multi-energy users by using a fuzzy C-means clustering algorithm, and carrying out fine modeling on various user energy characteristics and core appeal on the multi-user from multiple angles.
Description
Technical Field
The invention relates to the field of comprehensive energy systems, in particular to a typical multi-energy user model building method.
Background
The comprehensive energy system relates to various energy forms of electricity, gas, cold/heat and the like, and the various energy forms have obvious characteristic differences in the links of production, transmission, consumption, storage and the like and also have complex interconversion and coupling association. In addition, the deep integration of energy systems with information communication technology allows for significant changes in the operating modes of the integrated energy system relative to conventional single energy systems. The characteristics bring a series of problems to modeling, algorithms and evaluation indexes of the comprehensive energy system reliability evaluation, the reliability evaluation modeling based on big data, cloud computing and machine learning fully utilizes information flow to realize a reliability evaluation algorithm for accurate simulation and rapid evaluation of the system running state, and a reliability evaluation index system aiming at different energy supply form characteristics is a future research direction for the comprehensive energy system reliability evaluation; how to establish a typical energy utilization equipment digital model, further realize network planning and operation simulation support of a multi-energy system, and accumulate long-term operation data is a key problem of the development of a comprehensive energy system, and a breakthrough of related technologies is urgently needed.
For example, a "comprehensive demand response method based on the node energy price strategy of the comprehensive energy system" disclosed in the chinese patent literature, the publication No. CN 113077173a, includes constructing an electric-gas-heat comprehensive energy system operation framework; providing a calculation method of the energy price of an electric-gas-thermal coupling multi-energy flow network node containing fan output; acquiring node energy prices, building an energy consumption cost model considering load change of user comprehensive demand response, and analyzing response behaviors of various flexible loads under the guidance of an energy pricing mechanism of a comprehensive energy system; according to the technical scheme, the prediction of the multi-energy characteristic portrait and the energy demand of a typical multi-energy user is lacked, a differentiated and precise marketing strategy is not provided, the uncertainty of user behaviors is not considered, the analysis method is limited, and the considered energy consumption data is less.
Disclosure of Invention
The invention aims to overcome the problems that the prior art lacks of multi-energy characteristic portrayal and energy demand prediction of typical multi-energy users, uncertainty of user behaviors is not considered, and an analysis method is limited, and provides a typical multi-energy user model building method.
In order to achieve the purpose, the invention adopts the following technical scheme: a typical multi-energy user model building method comprises the following steps:
s1: collecting actual cases and demonstration projects of the comprehensive energy system, and analyzing and summarizing main application scenes of the comprehensive energy system;
s2: extracting historical energy utilization parameters of typical multi-energy users by using a fuzzy C-means clustering algorithm, preprocessing and screening the energy utilization parameters, and analyzing typical data and indexes capable of expressing energy utilization characteristics;
s3: establishing a multi-user evaluation model and indexes in the comprehensive energy market based on an analytic hierarchy process, an introduced entropy weight method and an approximation understanding sorting method and combining subjective judgment and objective information of an operator;
s4: abstracting concrete information of the multi-functional user into a label according to the evaluation model and the index, concretizing the user image through the label, and then constructing a user portrait of the multi-functional user energy characteristic;
s5: defining the energy demand time and space characteristics of typical multi-energy users, and analyzing the long-time scale growth trend of the users;
s6: and establishing a self-adaptive combined prediction model to predict the power load of different label users according to the energy requirements, and improving the prediction accuracy through a data-driven precision feedback mechanism.
The method collects practical cases and demonstration projects of the comprehensive energy system, analyzes and summarizes main application scenes of the comprehensive energy system, extracts historical energy consumption parameters of typical multi-energy users, cleans data, improves value density, analyzes and extracts high-order indexes of typical energy consumption characteristics; according to key indexes embodying the energy characteristics of the multi-energy users, typical users are further classified, and typical energy utilization labels are formed in a concrete mode; establishing a user image through the label to realize the mapping from the energy utilization characteristics to the user label; defining the energy demand time and space characteristics of typical multi-energy users, and analyzing the long-time scale growth trend of the users; a self-adaptive combined prediction model is established to carry out differential prediction on energy requirements of different label users, and prediction accuracy is improved through a data-driven precision feedback mechanism.
Preferably, the step S2 specifically includes the following steps:
s21: cleaning, integrating, transforming and stipulating original data by adopting a data preprocessing related technology;
s22: on the basis of a database generated by preprocessing, a fuzzy C-means clustering algorithm is adopted to research the feature extraction of the energy utilization characteristics of multiple users under different industry categories, energy consumption and reliability requirements;
s23: and fuzzy division is carried out by adopting an FCM clustering algorithm, and energy utilization parameters of each data object for each comprehensive energy user are defined.
Solving an objective function of a minimum clustering algorithm to obtain typical user energy data, economic data and a variation range in a multi-user group, analyzing energy consumption rules of different users, and extracting energy consumption characteristics of the multi-user; and further combining the self characteristics of the multi-user such as industry categories, energy consumption, reliability requirements and the like with the clustering result to extract the comprehensive energy utilization characteristics of the multi-user under the requirements of each industry, energy consumption level and reliability.
Preferably, the step S3 specifically includes the following steps:
s31: determining influence factors of user evaluation by using an analytic result of the electricity change process and the electricity selling market environment by using an analytic hierarchy process;
s32: establishing a judgment matrix according to the power characteristics, demand characteristics and industrial development prospect data of users;
s33: and (4) introducing an entropy weight method to correct the weight of the influencing factor.
Preferably, the step S4 specifically includes the following steps:
s41: carrying out characteristic classification and grading according to the difference of the basic attribute, the electricity consumption behavior, the payment behavior and the appeal behavior of the user;
s42: extracting typical features from each type, and giving a threshold value of the label;
s43: and according to the final label, developing the portrait of the power consumer by combining a business requirement scene.
Preferably, the energy consumption parameter refers to the total amount of energy consumed by a user, the energy consumption time, the energy price sensitivity, the energy consumption expectation degree and the industrial scale.
Preferably, in the FCM clustering algorithm in step S23, the degree of membership between [0,1] of all data objects to each cluster is represented by an objective function:
in the formula, Jm(U, P) represents the degree to which a data object belongs to each cluster, μikE 1 denotes the degree to which the kth data object belongs to the ith cluster center,Pia cluster center representing cluster i; m is an element of [0,2 ]]Represents a weighted index; dkiRepresenting the euclidean distance of the ith cluster center from the kth data object.
Preferably, the step S6 specifically includes the following steps:
s61: selecting a load in a past period of time as a training sample, and performing regression processing;
s62: constructing a network structure and establishing a neural network model;
s63: and predicting the power load through the fitting degree of the neural network model.
Preferably, the regression processing in step S61 includes:
Yt=b0+b1Xt1+b2Xt2+…+bnXtn
in the formula, Xt1,Xt2,…XtnRepresenting factors influencing the load change, b0,b1,…bnRepresenting a parametric variable, YtRepresenting the electrical load.
Preferably, the neural network model in step S62 is:
Y(i)=F(Wi,Y(i-1),M(t-1))
in the formula, YiY (i, t) t 1,2, …,24 represents the load vector on day i; y (i, t) represents the load at the t hour on day i; wiRepresenting a weight vector; m(t-1)=(m(t-1),m(t-2),…,m(t-k)) K is a data length, which indicates a factor affecting load variation.
Preferably, the factors affecting the load change include weather conditions, temperature and humidity.
Preferably, the user representation comprises a personal representation and a group representation; the personal portrait is that each client is pasted with an exclusive label according to the actual situation according to the label in the user label library; the group representation is constructed by screening the user system for personal representations that simultaneously satisfy the selected tags, using known tags.
Therefore, the invention has the following beneficial effects:
1. the subjective characteristics of the user and the market objective environment factors are comprehensively considered, and a reasonable and effective user evaluation model is established;
2. establishing a user image through the label to realize the mapping from the energy utilization characteristics to the user label;
3. and extracting historical energy consumption parameters of typical multi-energy users by using a fuzzy C-means clustering algorithm, and carrying out fine modeling on various user energy characteristics and core appeal on the multi-user from multiple angles.
Drawings
FIG. 1 is a typical multi-energy user model building flow.
Detailed Description
The present embodiment is further described with reference to the following drawings and detailed description.
The embodiment provides a typical multi-energy user model building method, and fig. 1 is a typical multi-energy user model building process, which includes the following steps:
s1: collecting actual cases and demonstration projects of the comprehensive energy system, and analyzing and summarizing main application scenes of the comprehensive energy system;
s2: extracting historical energy utilization parameters of typical multi-energy users by using a fuzzy C-means clustering algorithm, preprocessing and screening the energy utilization parameters, and analyzing typical data and indexes capable of expressing energy utilization characteristics;
s3: establishing a multi-user evaluation model and indexes in the comprehensive energy market based on an analytic hierarchy process, an introduced entropy weight method and an approximation understanding sorting method and combining subjective judgment and objective information of an operator;
s4: abstracting concrete information of the multi-functional user into a label according to the evaluation model and the index, concretizing the user image through the label, and then constructing a user portrait of the multi-functional user energy characteristic;
s5: defining the energy demand time and space characteristics of typical multi-energy users, and analyzing the long-time scale growth trend of the users;
s6: and establishing a self-adaptive combined prediction model to predict the power load of different label users according to the energy requirements, and improving the prediction accuracy through a data-driven precision feedback mechanism.
Step S2 specifically includes the following steps:
s21: cleaning, integrating, transforming and stipulating original data by adopting a data preprocessing related technology;
s22: on the basis of a database generated by preprocessing, a fuzzy C-means clustering algorithm is adopted to research the feature extraction of the energy utilization characteristics of multiple users under different industry categories, energy consumption and reliability requirements;
s23: and fuzzy division is carried out by adopting an FCM clustering algorithm, and energy utilization parameters of each data object for each comprehensive energy user are defined.
In step S23, the FCM clustering algorithm represents the degree of each cluster to which all data objects belong by using the membership degree between [0,1], and the objective function is:
in the formula, Jm(U, P) represents the degree to which a data object belongs to each cluster, μikE 1 denotes the degree to which the kth data object belongs to the ith cluster center,Pia cluster center representing cluster i; m is an element of [0,2 ]]Represents a weighted index; dkiRepresenting the euclidean distance of the ith cluster center from the kth data object.
Step S3 specifically includes the following steps:
s31: determining influence factors of user evaluation by using an analytic result of the electricity change process and the electricity selling market environment by using an analytic hierarchy process;
s32: establishing a judgment matrix according to the power characteristics, demand characteristics and industrial development prospect data of users;
s33: and (4) introducing an entropy weight method to correct the weight of the influencing factor.
Step S4 specifically includes the following steps:
s41: carrying out characteristic classification and grading according to the difference of the basic attribute, the electricity consumption behavior, the payment behavior and the appeal behavior of the user;
s42: extracting typical features from each type, and giving a threshold value of the label;
s43: and according to the final label, developing the portrait of the power consumer by combining a business requirement scene.
Step S6 specifically includes the following steps:
s61: selecting a load in a past period of time as a training sample, and performing regression processing;
s62: constructing a network structure and establishing a neural network model;
s63: and predicting the power load through the fitting degree of the neural network model.
The process of the regression processing in step S61 is:
Yt=b0+b1Xt1+b2Xt2+…+bnXtn
in the formula, Xt1,Xt2,…XtnRepresenting factors influencing the load change, b0,b1,…bnRepresenting a parametric variable, YtRepresenting the electrical load.
The neural network model in step S62 is:
Y(i)=F(Wi,Y(i-1),M(t-1))
in the formula, YiY (i, t) t 1,2, …,24 represents the load vector on day i; y (i, t) represents the load at the t hour on day i; wiRepresenting a weight vector; m(t-1)=(m(t-1),m(t-2),…,m(t-k)) K is a data length, which indicates a factor affecting load variation.
Firstly, the original data needs to be cleaned, integrated, transformed and regulated by adopting a data preprocessing related technology. On the basis of a database generated by preprocessing, Fuzzy C-Means clustering algorithm (FCM) is adopted to research the feature extraction of the energy utilization characteristics of multiple users under the factors of different industry categories, energy consumption, reliability requirements and the like. The FCM clustering algorithm adopts fuzzy division, defines each data object as data information of the total energy consumption, the energy consumption time, the energy price sensitivity, the energy consumption expectation degree and the industrial scale of each comprehensive energy user, and expresses the degree of belonging to each cluster of all the data objects by adopting the membership degree between [0,1 ].
Based on the analysis of diversified user energy characteristics and core appeal, a user energy evaluation index system is determined, user subjective characteristics and market objective environment factors are comprehensively considered, and a reasonable and effective user evaluation model is established.
Based on an analytic hierarchy process, an introduced entropy weight method and an approximate understanding ordering method (TOPSIS method), and combining subjective judgment and objective information of an operator, a multi-user evaluation model and indexes in the comprehensive energy market can be established; the evaluation model is mainly divided into 3 steps, firstly, an analytic hierarchy process is applied to determine influence factors of user evaluation (determined by a power change process and a power selling market environment analysis result), and a judgment matrix is established (the data of the judgment matrix is determined to be from the power characteristics, demand characteristics, industry development prospects and the like of users); then, introducing an entropy weight method to correct the weight of the influencing factors; the TOPSIS method is a common method for multi-target decision analysis of a limited scheme in system engineering; based on the normalized original data matrix, finding out the optimal scheme and the worst scheme (respectively represented by the optimal vector and the worst vector) in the limited scheme, then respectively calculating the distance between each evaluation object and the optimal scheme and the worst scheme, and obtaining the relative proximity degree of the evaluation object and the optimal scheme, wherein the relative proximity degree is used as the basis for evaluating the quality of the evaluation object.
The TOPSIS method is an abbreviation of Technique for Order Preference by Similarity to Ideal Solution, namely a technology approaching an Ideal Solution, and is a multi-target decision method; the basic idea of the method is to define an ideal solution and a negative ideal solution of a decision problem, and then find a scheme in a feasible scheme, so that the distance between the scheme and the ideal solution is the closest, and the distance between the scheme and the negative ideal solution is the farthest; the ideal solution is generally the best scheme to be assumed, and the corresponding attributes at least reach the best values in each scheme; a negative ideal solution assumes a worst case scenario with corresponding individual attributes that are at least not better than the worst values in the individual scenarios. The decision rule of the scheme queuing is to compare the practical feasible solution and the ideal solution with the negative ideal solution, and if a feasible solution is closest to the ideal solution and is also farthest from the negative ideal solution, the solution is a satisfactory solution of the scheme set.
The working process of the invention is as follows: collecting actual cases and demonstration projects of the comprehensive energy system, and analyzing and summarizing main application scenes of the comprehensive energy system; extracting historical energy utilization parameters of typical multi-energy users by using a fuzzy C-means clustering algorithm, preprocessing and screening the energy utilization parameters, and analyzing typical data and indexes capable of expressing energy utilization characteristics; establishing a multi-user evaluation model and indexes in the comprehensive energy market based on an analytic hierarchy process, an introduced entropy weight method and an approximation understanding sorting method and combining subjective judgment and objective information of an operator; abstracting concrete information of the multi-functional user into a label according to the evaluation model and the index, concretizing the user image through the label, and then constructing a user portrait of the multi-functional user energy characteristic; defining the energy demand time and space characteristics of typical multi-energy users, and analyzing the long-time scale growth trend of the users; and establishing a self-adaptive combined prediction model to predict the power load of different label users according to the energy requirements, and improving the prediction accuracy through a data-driven precision feedback mechanism.
The present invention is not limited to the above-described embodiments, and the above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention. Any modification, equivalent transformation, improvement, etc. made in accordance with the technical spirit of the present invention fall within the scope of the claimed invention.
Claims (10)
1. A typical multi-energy user model building method is characterized by comprising the following steps:
s1: collecting actual cases and demonstration projects of the comprehensive energy system, and analyzing and summarizing main application scenes of the comprehensive energy system;
s2: extracting historical energy utilization parameters of typical multi-energy users by using a fuzzy C-means clustering algorithm, preprocessing and screening the energy utilization parameters, and analyzing typical data and indexes capable of expressing energy utilization characteristics;
s3: establishing a multi-user evaluation model and indexes in the comprehensive energy market based on an analytic hierarchy process, an introduced entropy weight method and an approximation understanding sorting method and combining subjective judgment and objective information of an operator;
s4: abstracting concrete information of the multi-functional user into a label according to the evaluation model and the index, concretizing the user image through the label, and then constructing a user portrait of the multi-functional user energy characteristic;
s5: defining the energy demand time and space characteristics of typical multi-energy users, and analyzing the long-time scale growth trend of the users;
s6: and establishing a self-adaptive combined prediction model to predict the power load of different label users according to the energy requirements, and improving the prediction accuracy through a data-driven precision feedback mechanism.
2. The method for building the typical multi-energy user model according to claim 1, wherein the step S2 specifically comprises the following steps:
s21: cleaning, integrating, transforming and stipulating original data by adopting a data preprocessing related technology;
s22: on the basis of a database generated by preprocessing, a fuzzy C-means clustering algorithm is adopted to research the feature extraction of the energy utilization characteristics of multiple users under different industry categories, energy consumption and reliability requirements;
s23: and fuzzy division is carried out by adopting an FCM clustering algorithm, and energy utilization parameters of each data object for each comprehensive energy user are defined.
3. The method for building the typical multi-energy user model according to claim 1, wherein the step S3 specifically comprises the following steps:
s31: determining influence factors of user evaluation by using an analytic result of the electricity change process and the electricity selling market environment by using an analytic hierarchy process;
s32: establishing a judgment matrix according to the power characteristics, demand characteristics and industrial development prospect data of users;
s33: and (4) introducing an entropy weight method to correct the weight of the influencing factor.
4. The method for building the typical multi-energy user model according to claim 1, wherein the step S4 specifically comprises the following steps:
s41: carrying out characteristic classification and grading according to the difference of the basic attribute, the electricity consumption behavior, the payment behavior and the appeal behavior of the user;
s42: extracting typical features from each type, and giving a threshold value of the label;
s43: according to the final label, a service demand scene is combined, and a power user portrait is developed;
the energy consumption parameters refer to the total energy consumption of a user, energy consumption time, energy price sensitivity, energy consumption expectation degree and industrial scale.
5. The method according to claim 2, wherein the FCM clustering algorithm in step S23 represents the degree of membership between [0,1] of all data objects in each cluster, and the objective function is:
in the formula, Jm(U, P) represents the degree to which a data object belongs to each cluster, μikE 1 denotes the degree to which the kth data object belongs to the ith cluster center,Pia cluster center representing cluster i; m is an element of [0,2 ]]Represents a weighted index; dkiRepresenting the euclidean distance of the ith cluster center from the kth data object.
6. The method for building the typical multi-energy user model according to claim 1, wherein the step S6 specifically comprises the following steps:
s61: selecting a load in a past period of time as a training sample, and performing regression processing;
s62: constructing a network structure and establishing a neural network model;
s63: and predicting the power load through the fitting degree of the neural network model.
7. The method for building an exemplary multi-energy user model according to claim 5, wherein the regression process in step S61 comprises:
Yt=b0+b1Xt1+b2Xt2+…+bnXtn
in the formula, Xt1,Xt2,…XtnRepresenting factors influencing the load change, b0,b1,…bnRepresenting a parametric variable, YtRepresenting the electrical load.
8. The method for building the canonical multifunctional user model according to claim 5, wherein the neural network model in step S62 is:
Y(i)=F(Wi,Y(i-1),M(t-1))
in the formula, YiY (i, t) | t ═ 1,2, …,24} represents the load vector on day i; y (i, t) represents the load at the t hour on day i; wiRepresenting a weight vector; m(t-1)=(m(t-1),m(t-2),…,m(t-k)) K is a data length, which indicates a factor affecting load variation.
9. A method as claimed in claim 7 or 8, wherein the factors affecting the change in load include weather conditions, temperature and humidity.
10. A method of building a representative multi-capability user model according to claim 1 or 4, wherein said user representation comprises a personal representation and a group representation; the personal portrait is that each client is pasted with an exclusive label according to the actual situation according to the label in the user label library; the group representation is constructed by screening the user system for personal representations that simultaneously satisfy the selected tags, using known tags.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111396523.9A CN114358474A (en) | 2021-11-23 | 2021-11-23 | Typical multi-energy user model building method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111396523.9A CN114358474A (en) | 2021-11-23 | 2021-11-23 | Typical multi-energy user model building method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114358474A true CN114358474A (en) | 2022-04-15 |
Family
ID=81096427
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111396523.9A Pending CN114358474A (en) | 2021-11-23 | 2021-11-23 | Typical multi-energy user model building method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114358474A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115130006A (en) * | 2022-08-04 | 2022-09-30 | 北京富通智康科技有限公司 | User portrayal method based on health management label |
CN115310888A (en) * | 2022-10-13 | 2022-11-08 | 国网天津市电力公司城东供电分公司 | Comprehensive energy user energy consumption behavior correlation analysis method based on multi-element data processing |
-
2021
- 2021-11-23 CN CN202111396523.9A patent/CN114358474A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115130006A (en) * | 2022-08-04 | 2022-09-30 | 北京富通智康科技有限公司 | User portrayal method based on health management label |
CN115310888A (en) * | 2022-10-13 | 2022-11-08 | 国网天津市电力公司城东供电分公司 | Comprehensive energy user energy consumption behavior correlation analysis method based on multi-element data processing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Cai et al. | Predicting the energy consumption of residential buildings for regional electricity supply-side and demand-side management | |
Bassamzadeh et al. | Multiscale stochastic prediction of electricity demand in smart grids using Bayesian networks | |
CN117725537A (en) | Real-time metering data processing platform | |
CN111695793A (en) | Method and system for evaluating energy utilization flexibility of comprehensive energy system | |
CN115099511A (en) | Photovoltaic power probability estimation method and system based on optimized copula | |
CN117575663A (en) | Fitment cost estimation method and system based on deep learning | |
Raghavendra et al. | Artificial humming bird with data science enabled stability prediction model for smart grids | |
CN115759371A (en) | GCN-LSTM-based short-term load prediction method for power system | |
CN111428766B (en) | Power consumption mode classification method for high-dimensional mass measurement data | |
Sina et al. | Short term load forecasting model based on kernel-support vector regression with social spider optimization algorithm | |
CN114358474A (en) | Typical multi-energy user model building method | |
Shi et al. | Handling uncertainty in financial decision making: a clustering estimation of distribution algorithm with simplified simulation | |
Yiping et al. | An improved multi-view collaborative fuzzy C-means clustering algorithm and its application in overseas oil and gas exploration | |
CN115829683A (en) | Power integration commodity recommendation method and system based on inverse reward learning optimization | |
CN116628534A (en) | Method for dividing energy dynamic images for park based on electric power big data | |
Haq et al. | Classification of electricity load profile data and the prediction of load demand variability | |
CN113052395A (en) | Method for predicting financial data by neural network fusing network characteristics | |
Irfan et al. | Week Ahead Electricity Power and Price Forecasting Using Improved DenseNet-121 Method. | |
Wu et al. | RETRACTED ARTICLE: Artificial neural network based high dimensional data visualization technique for interactive data exploration in E-commerce | |
Chen et al. | K-means clustering method based on nearest-neighbor density matrix for customer electricity behavior analysis | |
Kahraman et al. | Fuzzy and grey forecasting techniques and their applications in production systems | |
Hicham et al. | Hybrid intelligent system for sale forecasting using Delphi and adaptive fuzzy back-propagation neural networks | |
CN116629904A (en) | Client hierarchical matching method based on big data | |
Alikhani et al. | Optimal demand response programs selection using CNN‐LSTM algorithm with big data analysis of load curves | |
CN111401638B (en) | Spatial load prediction method based on extreme learning machine and load density index method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |