CN114638425B - Historical data-based large-user monthly electricity consumption prediction method and system - Google Patents
Historical data-based large-user monthly electricity consumption prediction method and system Download PDFInfo
- Publication number
- CN114638425B CN114638425B CN202210295774.6A CN202210295774A CN114638425B CN 114638425 B CN114638425 B CN 114638425B CN 202210295774 A CN202210295774 A CN 202210295774A CN 114638425 B CN114638425 B CN 114638425B
- Authority
- CN
- China
- Prior art keywords
- model
- historical data
- electricity consumption
- data
- training
- 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.)
- Active
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 85
- 238000012360 testing method Methods 0.000 claims abstract description 27
- 238000007781 pre-processing Methods 0.000 claims abstract description 14
- 238000012417 linear regression Methods 0.000 claims description 32
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 6
- 238000004519 manufacturing process Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 description 5
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 2
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
Images
Classifications
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/08—Learning methods
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Computation (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Game Theory and Decision Science (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to a method and a system for predicting monthly electricity consumption of a large user based on historical data, and belongs to the field of power grid regulation and control and power market. The method comprises the following steps: collecting historical data of the power consumption of a large user, and preprocessing the historical data, wherein the historical data comprises daily load historical data and/or monthly load historical data; dividing the preprocessed historical data into a training set and a testing set, and training a model of a preset type by using the training set to obtain a large-user monthly electricity consumption prediction model; and predicting and outputting a month electricity consumption predicted value of the large user for one month in the future according to the month electricity consumption predicted model of the large user and the history data of the large user. The method has high prediction accuracy, can reflect the power utilization change trend of a large user, and provides a reliable basis for the production plan of the electric company.
Description
Technical Field
The invention belongs to the field of power system automation, and particularly relates to a method and a system for predicting the monthly electricity consumption of a large user based on historical data.
Background
Load prediction is an indispensable link in the power market and grid operation, the necessity of which goes without saying. The large users are used as important clients of the power grid, the power consumption occupies a relatively high proportion of the total power consumption of the area, and the load curve has a considerable influence on the load curve of the area power grid. The power consumption prediction is carried out on large users, and the system load value is predicted from one day to one week to one month in advance, so that the method has an important effect on determining the daily operation mode of the power grid, and is also indispensable for determining a unit combination scheme, an enterprise and regional power grid power transmission scheme and a load scheduling scheme. Therefore, the power company in many places recently requires large users to provide the next daily load curve to provide basis for the power company to make production plans. The accurate load and electricity consumption prediction can truly reflect the electricity consumption change trend, and the safe and economic operation of the power grid is ensured.
The current electricity consumption prediction method can be divided into two main types, namely a prediction method based on a traditional method and a prediction method based on an intelligent algorithm. The traditional method mainly comprises a regression analysis method, a time sequence method, a power consumption derivation method, an exponential smoothing method, a Kalman filtering method and the like; the intelligent method mainly comprises an expert system method, an artificial neural network method, a comprehensive model prediction method, a data mining method and the like. However, prediction of the monthly power consumption of a large user is a relatively complex problem, and various factors are generally considered to obtain relatively high prediction accuracy. How to accurately and effectively predict the monthly electricity consumption of large users is a problem to be solved.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art, and provides a method and a system for predicting the monthly electricity consumption of a large user based on historical data, which are used for preprocessing daily load historical data and/or monthly load historical data of the electricity consumption of the large user, training a model of a preset type by utilizing the preprocessed data to obtain a prediction model of the monthly electricity consumption of the large user, predicting the monthly electricity consumption of the large user for one month in the future by the prediction model, and the prediction accuracy is high, can reflect the electricity consumption change trend of the large user, and provides a reliable basis for an electric company to formulate a production plan.
According to one aspect of the present invention, there is provided a method for predicting monthly electricity consumption of a large user based on historical data, the method comprising the steps of:
s1, collecting historical data of the power consumption of a large user, and preprocessing the historical data, wherein the historical data comprises daily load historical data and/or monthly load historical data;
s2, dividing the preprocessed historical data into a training set and a testing set, and training a model of a preset type by using the training set to obtain a large-user monthly electricity consumption prediction model;
and S3, predicting and outputting a month electricity consumption predicted value of the large user for one month in the future according to the month electricity consumption predicted model of the large user and the history data of the large user.
Preferably, the preprocessing the history data includes:
filling the missing data, identifying and correcting the abnormal data, and carrying out normalization processing.
Preferably, when the historical data is daily load historical data, the model of the preset type is a long-short-period neural network LSTM model; when the historical data is month load historical data, the model of the preset type is a multiple linear regression model.
Preferably, when the model of the preset type is a long-short-term neural network LSTM model, training the model of the preset type by using the training set to obtain the model of the large-user monthly electricity consumption prediction includes:
the super-parameters of the LSTM model adopt a gradient descent adjustment mode, and the cost function is a square reconstruction error; and inputting the training set data into the LSTM model for iteration and calculating an error, judging whether the error is smaller than a preset threshold value and whether the iteration number exceeds the maximum limit number, if not, continuing to execute iteration operation, and if not, inputting the testing set data into the LSTM model for verification, so as to complete model training and obtain the large-user month electricity consumption prediction model.
Preferably, when the model of the preset type is a multiple linear regression model, training the model of the preset type by using the training set to obtain the prediction model of the monthly electricity consumption of the large user includes:
inputting the training set data into the multiple linear regression model to perform multiple linear regression fitting, inputting the test set data to calculate errors, judging whether the errors are smaller than a set value, if not, performing multiple linear regression fitting again by adjusting parameters, if so, judging whether the model is applicable, and if so, completing model training to obtain the large-user monthly electricity consumption prediction model.
According to another aspect of the present invention, there is also provided a system for predicting monthly electricity consumption of a large user based on historical data, the system comprising:
the processing module is used for collecting historical data of the power consumption of the large user and preprocessing the historical data, wherein the historical data comprises daily load historical data and/or monthly load historical data;
the training module is used for dividing the preprocessed historical data into a training set and a testing set, and training a model of a preset type by using the training set to obtain a large-user monthly electricity consumption prediction model;
and the prediction module is used for predicting and outputting a month electricity consumption predicted value of the large user for one month in the future according to the month electricity consumption prediction model of the large user and the history data of the large user.
Preferably, the preprocessing the historical data by the processing module includes:
filling the missing data, identifying and correcting the abnormal data, and carrying out normalization processing.
Preferably, when the historical data is daily load historical data, the model of the preset type is a long-short-period neural network LSTM model; when the historical data is month load historical data, the model of the preset type is a multiple linear regression model.
Preferably, when the model of the preset type is a long-short-period neural network LSTM model, the training module trains the model of the preset type by using the training set, and the obtaining the prediction model of the monthly electricity consumption of the large user includes:
the super-parameters of the LSTM model adopt a gradient descent adjustment mode, and the cost function is a square reconstruction error; and inputting the training set data into the LSTM model for iteration and calculating an error, judging whether the error is smaller than a preset threshold value and whether the iteration number exceeds the maximum limit number, if not, continuing to execute iteration operation, and if not, inputting the testing set data into the LSTM model for verification, so as to complete model training and obtain the large-user month electricity consumption prediction model.
Preferably, when the model of the preset type is a multiple linear regression model, the training module trains the model of the preset type by using the training set, and obtaining the prediction model of the monthly electricity consumption of the large user includes:
inputting the training set data into the multiple linear regression model to perform multiple linear regression fitting, inputting the test set data to calculate errors, judging whether the errors are smaller than a set value, if not, performing multiple linear regression fitting again by adjusting parameters, if so, judging whether the model is applicable, and if so, completing model training to obtain the large-user monthly electricity consumption prediction model.
The beneficial effects are that: according to the method, the long-short-period neural network method and the multiple linear regression method with memory characteristics are respectively adopted according to given daily load or month load historical data and analysis of the similarity and the relevance of the daily loads or the loads of the same month respectively, so that the month electricity consumption predicted value of a large user in the future is obtained, the prediction accuracy is high, the electricity consumption change trend of the large user can be reflected, and a reliable basis is provided for the establishment of a production plan of an electric company.
Features and advantages of the present invention will become apparent by reference to the following drawings and detailed description of embodiments of the invention.
Drawings
FIG. 1 is a flow chart of a large user monthly electricity consumption prediction method based on historical data;
FIG. 2 is a flowchart of a large user monthly electricity consumption prediction method based on daily load history data;
FIG. 3 is a flowchart of a large user monthly electricity consumption prediction method based on monthly load history data;
fig. 4 is a schematic diagram of a large user month electricity consumption prediction system based on historical data.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
FIG. 1 is a flow chart of a large user monthly electricity consumption prediction method based on historical data. As shown in fig. 1, the present invention provides a method for predicting the monthly electricity consumption of a large user based on historical data, which comprises the following steps:
s1, collecting historical data of the power consumption of a large user, and preprocessing the historical data, wherein the historical data comprises daily load historical data and/or monthly load historical data;
s2, dividing the preprocessed historical data into a training set and a testing set, and training a model of a preset type by using the training set to obtain a large-user monthly electricity consumption prediction model;
and S3, predicting and outputting a month electricity consumption predicted value of the large user for one month in the future according to the month electricity consumption predicted model of the large user and the history data of the large user.
According to the method, the daily load historical data and/or the monthly load historical data of the power consumption of the large user are preprocessed, the preprocessed data are used for training a model of a preset type to obtain a prediction model of the monthly power consumption of the large user, the prediction model is used for predicting the monthly power consumption of the large user for one month in the future, the prediction accuracy is high, the power consumption change trend of the large user can be reflected, and reliable basis is provided for making a production plan for an electric company.
Preferably, the preprocessing the history data includes:
filling the missing data, identifying and correcting the abnormal data, and carrying out normalization processing.
Specifically, the historical load data acquisition system may have a data missing or abnormal situation. Without processing the input model, a large error would result. Therefore, the data preprocessing needs to fill in the missing data and identify and correct the abnormal data. For convenience of processing, normalization processing is performed on the original data.
Preferably, when the historical data is daily load historical data, the model of the preset type is a long-short-period neural network LSTM model; when the historical data is month load historical data, the model of the preset type is a multiple linear regression model.
Specifically, according to given daily load or month load historical data and analysis of respective similarity and relevance of the daily load or the same month load, a long-short-period neural network method and a multiple linear regression method with memory characteristics are adopted respectively to obtain a month electricity consumption predicted value of a large user in the future, prediction accuracy is improved, and meanwhile, prediction modes are convenient, rapid and various, and operability is high.
Preferably, when the model of the preset type is a long-short-term neural network LSTM model, training the model of the preset type by using the training set to obtain the model of the large-user monthly electricity consumption prediction includes:
the super-parameters of the LSTM model adopt a gradient descent adjustment mode, and the cost function is a square reconstruction error; and inputting the training set data into the LSTM model for iteration and calculating an error, judging whether the error is smaller than a preset threshold value and whether the iteration number exceeds the maximum limit number, if not, continuing to execute iteration operation, and if not, inputting the testing set data into the LSTM model for verification, so as to complete model training and obtain the large-user month electricity consumption prediction model.
Specifically, referring to fig. 2, input and output variables of a model are determined, and a preprocessed data set is divided into 2 parts of a training set and a test set. The training set data are divided into unlabeled data and labeled data according to whether labels exist or not, the training set data are used for model training, and the test set is used for testing the advantages and disadvantages of the model on the prediction performance of the new data.
The training set data is input into an LSTM network model to obtain a predicted value, the model super-parameters adopt a gradient descent adjustment mode, the cost function is a square reconstruction error, and the LSTM realizes supervised learning.
After LSTM network training is completed, test set data is input, and a prediction result is output after inverse normalization.
And comparing and evaluating the load predicted value with the actual load, and measuring the prediction accuracy of the LSTM network model by adopting MAPE and RMSE as rating indexes of the predicted result. The smaller MAPE and RMSE in the power load prediction, the better the load prediction effect.
Preferably, when the model of the preset type is a multiple linear regression model, training the model of the preset type by using the training set to obtain the prediction model of the monthly electricity consumption of the large user includes:
inputting the training set data into the multiple linear regression model to perform multiple linear regression fitting, inputting the test set data to calculate errors, judging whether the errors are smaller than a set value, if not, performing multiple linear regression fitting again by adjusting parameters, if so, judging whether the model is applicable, and if so, completing model training to obtain the large-user monthly electricity consumption prediction model.
Specifically, referring to fig. 3, the input/output variables of the model are determined, the month electricity consumption history data of the large user is used as the input variables, and the month electricity consumption predicted value of one month in the future is used as the output variables. The preprocessed data set is divided into 2 parts of a training set and a testing set. The training set data is used for model training, and the test set is used for testing the prediction performance of the model on the new data.
And (3) performing multiple linear fitting on the training set data, performing error judgment and parameter adjustment, and simultaneously judging whether the final model is applicable or not by combining the error judgment of the test set.
If the model is applicable, the model training is completed, and a month electricity consumption prediction result in the future is output.
According to the embodiment of the invention, the month electricity consumption of the large user in the future is predicted by constructing the prediction model, the prediction accuracy is high, the electricity consumption change trend of the large user can be reflected, and a reliable basis is provided for the establishment of a production plan of an electric company.
Example 2
Fig. 4 is a schematic diagram of a large user month electricity consumption prediction system based on historical data. As shown in fig. 4, the present invention further provides a system for predicting the monthly electricity consumption of a large user based on historical data, the system comprising:
the processing module 401 is configured to collect historical data of a large user power consumption, and preprocess the historical data, where the historical data includes daily load historical data and/or monthly load historical data;
the training module 402 is configured to divide the preprocessed historical data into a training set and a testing set, and train a model of a preset type by using the training set to obtain a monthly electricity consumption prediction model of a large user;
and the prediction module 403 is configured to predict and output a predicted value of the monthly electricity consumption of the large user for one month according to the prediction model of the monthly electricity consumption of the large user and the historical data of the monthly electricity consumption of the large user.
Preferably, the preprocessing the historical data by the processing module 401 includes:
filling the missing data, identifying and correcting the abnormal data, and carrying out normalization processing.
Preferably, when the historical data is daily load historical data, the model of the preset type is a long-short-period neural network LSTM model; when the historical data is month load historical data, the model of the preset type is a multiple linear regression model.
Preferably, when the model of the preset type is a long-short-term neural network LSTM model, the training module 402 trains the model of the preset type by using the training set, and the obtaining the prediction model of the large-user monthly electricity consumption includes:
the super-parameters of the LSTM model adopt a gradient descent adjustment mode, and the cost function is a square reconstruction error; and inputting the training set data into the LSTM model for iteration and calculating an error, judging whether the error is smaller than a preset threshold value and whether the iteration number exceeds the maximum limit number, if not, continuing to execute iteration operation, and if not, inputting the testing set data into the LSTM model for verification, so as to complete model training and obtain the large-user month electricity consumption prediction model.
Preferably, when the model of the preset type is a multiple linear regression model, the training module 402 trains the model of the preset type by using the training set, and the obtaining the prediction model of the monthly electricity consumption of the large user includes:
inputting the training set data into the multiple linear regression model to perform multiple linear regression fitting, inputting the test set data to calculate errors, judging whether the errors are smaller than a set value, if not, performing multiple linear regression fitting again by adjusting parameters, if so, judging whether the model is applicable, and if so, completing model training to obtain the large-user monthly electricity consumption prediction model.
The specific implementation process of the method steps executed by each module in embodiment 2 of the present invention is the same as that of each step in embodiment 1, and will not be described herein.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structural changes made by the specification and drawings of the present invention or direct/indirect application in other related technical fields are included in the scope of the present invention.
Claims (4)
1. A method for predicting the monthly electricity consumption of a large user based on historical data, which is characterized by comprising the following steps:
s1, collecting historical data of the power consumption of a large user, and preprocessing the historical data, wherein the historical data comprises daily load historical data and/or monthly load historical data;
s2, dividing the preprocessed historical data into a training set and a testing set, and training a model of a preset type by using the training set to obtain a large-user monthly electricity consumption prediction model;
s3, predicting and outputting a month electricity consumption predicted value of the large user for one month in the future according to the month electricity consumption predicted model of the large user and the history data of the large user; the preprocessing of the history data comprises:
filling the missing data, identifying and correcting the abnormal data, and carrying out normalization processing; when the historical data is daily load historical data, the model of the preset type is a long-short-period neural network LSTM model; when the historical data is month load historical data, the model of the preset type is a multiple linear regression model;
when the model of the preset type is a long-short-term neural network LSTM model, training the model of the preset type by using the training set to obtain a large-user monthly electricity consumption prediction model comprises the following steps:
the super-parameters of the LSTM model adopt a gradient descent adjustment mode, and the cost function is a square reconstruction error; and inputting the training set data into the LSTM model for iteration and calculating an error, judging whether the error is smaller than a preset threshold value and whether the iteration number exceeds the maximum limit number, if not, continuing to execute iteration operation, and if not, inputting the testing set data into the LSTM model for verification, so as to complete model training and obtain the large-user month electricity consumption prediction model.
2. The method of claim 1, wherein training the model of the preset type with the training set when the model of the preset type is a multiple linear regression model to obtain the large user monthly electricity consumption prediction model comprises:
inputting the training set data into the multiple linear regression model to perform multiple linear regression fitting, inputting the test set data to calculate errors, judging whether the errors are smaller than a set value, if not, performing multiple linear regression fitting again by adjusting parameters, if so, judging whether the model is applicable, and if so, completing model training to obtain the large-user monthly electricity consumption prediction model.
3. A large user monthly electricity usage prediction system based on historical data, the system comprising:
the processing module is used for collecting historical data of the power consumption of the large user and preprocessing the historical data, wherein the historical data comprises daily load historical data and/or monthly load historical data;
the training module is used for dividing the preprocessed historical data into a training set and a testing set, and training a model of a preset type by using the training set to obtain a large-user monthly electricity consumption prediction model;
the prediction module is used for predicting and outputting a month electricity consumption predicted value of the large user for one month in the future according to the month electricity consumption prediction model of the large user and the history data of the large user electricity consumption;
the preprocessing of the historical data by the processing module comprises the following steps:
filling the missing data, identifying and correcting the abnormal data, and carrying out normalization processing; when the historical data is daily load historical data, the model of the preset type is a long-short-period neural network LSTM model; when the historical data is month load historical data, the model of the preset type is a multiple linear regression model; when the model of the preset type is a long-short-period neural network LSTM model, the training module trains the model of the preset type by utilizing the training set, and the obtaining of the large-user monthly electricity consumption prediction model comprises the following steps:
the super-parameters of the LSTM model adopt a gradient descent adjustment mode, and the cost function is a square reconstruction error; and inputting the training set data into the LSTM model for iteration and calculating an error, judging whether the error is smaller than a preset threshold value and whether the iteration number exceeds the maximum limit number, if not, continuing to execute iteration operation, and if not, inputting the testing set data into the LSTM model for verification, so as to complete model training and obtain the large-user month electricity consumption prediction model.
4. The system of claim 3, wherein when the model of the preset type is a multiple linear regression model, the training module training the model of the preset type using the training set to obtain the large user monthly electricity consumption prediction model comprises:
inputting the training set data into the multiple linear regression model to perform multiple linear regression fitting, inputting the test set data to calculate errors, judging whether the errors are smaller than a set value, if not, performing multiple linear regression fitting again by adjusting parameters, if so, judging whether the model is applicable, and if so, completing model training to obtain the large-user monthly electricity consumption prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210295774.6A CN114638425B (en) | 2022-03-24 | 2022-03-24 | Historical data-based large-user monthly electricity consumption prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210295774.6A CN114638425B (en) | 2022-03-24 | 2022-03-24 | Historical data-based large-user monthly electricity consumption prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114638425A CN114638425A (en) | 2022-06-17 |
| CN114638425B true CN114638425B (en) | 2023-05-23 |
Family
ID=81949489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210295774.6A Active CN114638425B (en) | 2022-03-24 | 2022-03-24 | Historical data-based large-user monthly electricity consumption prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114638425B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115374404B (en) * | 2022-10-20 | 2023-01-24 | 国网(宁波)综合能源服务有限公司 | Method for correcting monthly electric energy proportion deviation of industry based on multi-dimensional data |
| CN115473232B (en) * | 2022-11-02 | 2023-03-24 | 广东电网有限责任公司中山供电局 | Load parameter adjusting method, system, equipment and storage medium |
| CN118070134B (en) * | 2024-04-24 | 2024-06-28 | 国网山东省电力公司汶上县供电公司 | Method and system for predicting electricity consumption of users in park |
| CN118690983A (en) * | 2024-08-28 | 2024-09-24 | 航天科工广信智能技术有限公司 | Multi-mode-based energy consumption prediction and energy saving scheme generation method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561156A (en) * | 2020-12-11 | 2021-03-26 | 国网江苏省电力有限公司南通供电分公司 | Short-term power load prediction method based on user load mode classification |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846528B (en) * | 2018-09-19 | 2021-07-09 | 云南电网有限责任公司信息中心 | Long-term load prediction method for large-scale industrial power consumer |
| CN112529301B (en) * | 2020-12-14 | 2023-11-14 | 南京中兴力维软件有限公司 | Power consumption prediction method, device and storage medium |
-
2022
- 2022-03-24 CN CN202210295774.6A patent/CN114638425B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561156A (en) * | 2020-12-11 | 2021-03-26 | 国网江苏省电力有限公司南通供电分公司 | Short-term power load prediction method based on user load mode classification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114638425A (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114638425B (en) | Historical data-based large-user monthly electricity consumption prediction method and system | |
| Zhou et al. | Electricity price forecasting with confidence-interval estimation through an extended ARIMA approach | |
| CN110222882A (en) | A kind of prediction technique and device of electric system Mid-long Term Load | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| Hargreaves et al. | REFLEX: An adapted production simulation methodology for flexible capacity planning | |
| CN105740975A (en) | Data association relationship-based equipment defect assessment and prediction method | |
| CN113487447A (en) | Power distribution network load prediction and line loss cause analysis method based on big data | |
| CN113608975A (en) | Method and system for online evaluation of running state of electric energy metering system in electric power spot market | |
| CN115688993A (en) | Short-term power load prediction method suitable for power distribution station area | |
| CN104834975A (en) | Power network load factor prediction method based on intelligent algorithm optimization combination | |
| CN117856245A (en) | Grid management-based method, device and storage medium for analyzing line loss of transformer area | |
| Yang et al. | Probabilistic revenue analysis of microgrid considering source-load and forecast uncertainties | |
| CN115511230B (en) | Electric energy substitution potential analysis and prediction method | |
| CN117236380A (en) | Power system fault prediction method, system, electronic equipment and medium | |
| Yundra et al. | Hybrid Model Combined Fuzzy Multi-Objective Decision Making with Feed Forward Neural Network (F-MODM-FFNN) For Very Short-Term Load Forecasting Based on Weather Data. | |
| CN108345996B (en) | System and method for reducing wind power assessment electric quantity | |
| CN113239029A (en) | Completion method for missing daily freezing data of electric energy meter | |
| Anderson et al. | Distributed Rules-Based Deconfliction of ADMS Applications: Part 2: Conceptual Implementation | |
| Wen | Research on Power Grid Big Data Short-Term Feature Mining Method Based on Convolution Neural Network | |
| CN117833243B (en) | Method, system, equipment and medium for predicting short-term demand of electric power | |
| CN114243695B (en) | Power load prediction method based on bidirectional long-short-term memory neural network | |
| Zhong et al. | Rapid Identification Method for the Sudden Change of Operating Errors in Smart Electricity Meters | |
| CN118644359B (en) | Hydrogen energy load prediction method and system applied to industrial park | |
| Barahona et al. | A framework for data-driven decision support for operational planning in active distribution networks | |
| Yang et al. | Multi-time Scale Resources Aggregation Demand Response Potential Assessment Method with Stacked Autoencoder |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |