CN116865236A - Medium-and-long-term power generation capacity prediction method and system based on new energy power generation - Google Patents

Medium-and-long-term power generation capacity prediction method and system based on new energy power generation Download PDF

Info

Publication number
CN116865236A
CN116865236A CN202310553483.7A CN202310553483A CN116865236A CN 116865236 A CN116865236 A CN 116865236A CN 202310553483 A CN202310553483 A CN 202310553483A CN 116865236 A CN116865236 A CN 116865236A
Authority
CN
China
Prior art keywords
target
data
power
power supply
prediction
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
Application number
CN202310553483.7A
Other languages
Chinese (zh)
Inventor
刘金晓
刘畅
代庚辉
李洋
李中惠
王兴兴
路凤桐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Guodian Investment Energy Marketing Co ltd
Original Assignee
Shandong Guodian Investment Energy Marketing Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong Guodian Investment Energy Marketing Co ltd filed Critical Shandong Guodian Investment Energy Marketing Co ltd
Priority to CN202310553483.7A priority Critical patent/CN116865236A/en
Publication of CN116865236A publication Critical patent/CN116865236A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
    • 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
    • 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
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • 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
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Power Engineering (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Economics (AREA)
  • Public Health (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Geometry (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • Computer Hardware Design (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

A method and a system for predicting medium-long term power generation based on new energy power generation belong to the technical field of new energy, and comprise the following steps: acquiring target environmental weather information and a target power supply object; obtaining target historical power supply information; carrying out electricity demand analysis on the target electricity demand area according to the meteorological information to obtain a target electricity demand predicted value; performing data processing on the target environmental meteorological information to construct a wind turbine generator power prediction model, and performing power supply prediction of a new energy power supply object to obtain a target wind power prediction result; and generating a target electricity demand gap according to a target wind power prediction result and setting an electricity demand supplement threshold value. The method solves the technical problems of low prediction precision of new energy generated power, low grid-connected adaptation degree and low new energy generation cost benefit in the prior art, and achieves the technical effects of accurately predicting new energy generation and adopting thermal power generation control measures, thereby reducing thermal power generation cost and ensuring normal power supply in a power supply area.

Description

Medium-and-long-term power generation capacity prediction method and system based on new energy power generation
Technical Field
The application relates to the technical field of new energy, in particular to a method and a system for predicting medium-long term power generation based on new energy power generation.
Background
The new energy power generation technology is one of important means for solving the energy crisis of the modern society and reducing environmental pollution. The new energy power generation such as wind power, photovoltaic and the like has the advantages of cleanness, safety, reproducibility and the like, and has been widely applied, however, the new energy belongs to intermittent energy, the generated energy is difficult to predict and schedule, and the benefit of the new energy power generation is low. Therefore, it is necessary to improve the accuracy of predicting the amount of new energy generation, and to perform safe and stable operation of the booster power system.
Disclosure of Invention
The application provides a method and a system for predicting medium-long-term power generation based on new energy power generation, which aim to solve the technical problems in the prior art that the new energy power generation has low prediction accuracy, so that the grid-connected adaptation degree of the new energy power generation is low, and the new energy power generation has a small effect on reducing the power generation cost.
In view of the above problems, the application provides a method and a system for predicting medium-long term power generation based on new energy power generation.
The application discloses a first aspect, which provides a medium-long term power generation capacity prediction method based on new energy power generation, comprising the steps of acquiring target environmental weather information according to target geographic position information of a target power demand area, wherein the target environmental weather information comprises a target environmental temperature parameter set, a target environmental humidity parameter set and a target environmental wind power parameter set; obtaining a target power supply object based on the target power demand area, wherein the target power supply object comprises a new energy power supply object and a fossil power supply object; carrying out data integration on historical power consumption data of a target power supply object to obtain target historical power supply information; carrying out power consumption demand analysis on the target power consumption area according to the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information to obtain a target power consumption predicted value; carrying out data processing on the target environmental meteorological information to construct a wind turbine generator power prediction model, and carrying out power supply prediction of a new energy power supply object to obtain a target wind power prediction result, wherein the target wind power prediction result comprises a middle wind power prediction result and a long-term wind power prediction result; generating a target electricity demand gap based on the target electricity demand predicted value and the mid-term wind power predicted result; and setting a power demand supplement threshold value for the fossil power supply object based on the target power demand gap and the long-term wind power prediction result.
In another aspect of the present disclosure, a system for predicting medium-to-long term power generation based on new energy power generation is provided, the system comprising: the environment weather information module is used for acquiring target environment weather information according to target geographical position information of a target electricity-requiring area, wherein the target environment weather information comprises a target environment temperature parameter set, a target environment humidity parameter set and a target environment wind power parameter set; the target power supply object module is used for obtaining a target power supply object based on a target power demand area, wherein the target power supply object comprises a new energy power supply object and a fossil power supply object; the historical power supply information module is used for carrying out data integration on the historical power consumption data of the target power supply object to obtain target historical power supply information; the power consumption demand analysis module is used for carrying out power consumption demand analysis on the target power consumption area according to the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information to obtain a target power consumption predicted value; the wind power prediction result module is used for carrying out data processing on the target environmental weather information to construct a wind turbine generator power prediction model, and carrying out power supply prediction of a new energy power supply object to obtain a target wind power prediction result, wherein the target wind power prediction result comprises a middle wind power prediction result and a long-term wind power prediction result; the target electricity demand gap module is used for generating a target electricity demand gap based on the target electricity demand predicted value and the mid-term wind power predicted result; and the electricity-required replenishment threshold module is used for setting an electricity-required replenishment threshold for the fossil power supply object based on the target electricity-required gap and the long-term wind power prediction result.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
acquiring target environmental weather information according to geographic position information of a target area, obtaining a target power supply object, performing data integration on historical power supply data, obtaining historical power supply information, and performing power demand analysis according to the environmental information and the historical power supply information to obtain a power demand predicted value for important input data for subsequent power demand prediction and wind power prediction; constructing a wind turbine generator power prediction model, and predicting the generated energy of a new energy power supply object to obtain wind power prediction results, including mid-term and long-term prediction; according to the technical scheme, the power demand gap is determined according to the power demand predicted value and the middle-term wind power predicted result, and the power demand supplement threshold value of fossil fuel power supply is determined according to the power demand gap and the long-term wind power predicted result, so that the technical problems that in the prior art, the prediction accuracy of new energy power generation power is poor, the grid-connected adaptation degree of the new energy power generation is low, the effect of the new energy power generation on reducing the power generation cost is low are solved, the accurate prediction of the new energy power generation by combining environment information and the new energy power generation historical data is achieved, the thermal power generation control is carried out by combining the power consumption predicted result of a power supply area, and the technical effect of guaranteeing the normal power supply of the power supply area is achieved while the thermal power generation cost is reduced.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
Fig. 1 is a schematic diagram of a possible flow chart of a method for predicting medium-long term power generation based on new energy power generation according to an embodiment of the present application;
fig. 2 is a schematic diagram of a possible flow chart for obtaining a target electricity demand predicted value in a method for predicting medium-long-term power generation based on new energy power generation according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a possible flow for obtaining a mid-term wind power prediction result in a mid-term and long-term power generation prediction method based on new energy power generation according to an embodiment of the present application;
fig. 4 is a schematic diagram of a possible structure of a medium-to-long term power generation prediction system based on new energy power generation according to an embodiment of the present application.
Reference numerals illustrate: the system comprises an environment weather information module 11, a target power supply object module 12, a historical power supply information module 13, a power demand analysis module 14, a wind power prediction result module 15, a target power demand gap module 16 and a power demand supplement threshold module 17.
Detailed Description
The technical scheme provided by the application has the following overall thought:
the embodiment of the application provides a method and a system for predicting medium-long term power generation based on new energy power generation. By obtaining power supply objects (including a thermal power plant and a wind power plant) of a power demand region, power demand data of the power demand region is predicted according to environmental weather in combination with historical data. The method comprises the steps of constructing a wavelet neural network according to wind speed and wind power, predicting generating capacity data of a wind power station, subtracting the generating capacity from the required electric capacity to serve as a predicted value of the generating capacity of the thermal power plant, and regulating the generating capacity of the thermal power plant, so that the normal power supply of a power supply area is ensured while the thermal power generation cost is reduced.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides a method for predicting medium-to-long term power generation based on new energy power generation, which includes:
step S100: acquiring target environmental weather information according to target geographical position information of a target electricity-requiring area, wherein the target environmental weather information comprises a target environmental temperature parameter set, a target environmental humidity parameter set and a target environmental wind power parameter set;
Specifically, the target power demand area refers to a geographical area where power supply and demand prediction and power generation scheduling are required, and an area where a certain power system is located can be selected as the target power demand area. The target geographic position information comprises geographic position parameters such as longitude and latitude coordinates, altitude and the like of the target electricity-requiring area, and is used for positioning the target electricity-requiring area and collecting environmental meteorological information in the area. The target environmental weather information refers to weather parameter information such as temperature, humidity, wind power and the like in a target electricity-requiring area. The target environment temperature parameter set is temperature information acquired in a target electricity-requiring area at a plurality of time points; the target environment humidity parameter set is humidity information collected in a target electricity-requiring area at a plurality of time points; the target environment wind power parameter set is wind speed and wind direction information acquired in a target electricity-requiring area at a plurality of time points.
And deploying environment monitoring facilities such as a weather station, a wind tower, satellite remote sensing, a weather observation radar and the like in the target electricity-requiring area to obtain target environment weather information. The acquired meteorological information is stored and forms a meteorological parameter data set which comprises a target environment temperature parameter set, a target environment humidity parameter set and a target environment wind power parameter set, is temperature, humidity and wind power space-time change information within a certain time range of a target electricity-requiring area, and is an important basis for the subsequent prediction analysis.
Step S200: obtaining a target power supply object based on the target power demand region, wherein the target power supply object comprises a new energy power supply object and a fossil power supply object;
specifically, the new energy power supply object refers to a power supply that generates power by using renewable resources such as wind energy, solar energy and the like, such as a wind farm, a photovoltaic power station and the like. And taking the wind power plant as a new energy power supply object as a representative, and constructing a wind turbine generator power prediction model to predict wind power generation capacity. The fossil fuel power supply target is a thermal generator set or a gas turbine set that generates power using fossil fuels such as coal, oil, and natural gas. The target power supply object consists of a new energy power supply object and a fossil fuel power supply object, and forms a main power supply system of the target power demand area together.
The existing electric power infrastructure of the target electricity-requiring area, such as a large-scale wind power plant or a photovoltaic power station, can be selected as a new energy power supply object; larger thermal power plants or gas turbines exist that can be selected as fossil fuel powered subjects. And analyzing the power resource distribution condition of the target power demand area, wherein if abundant wind resources are available, the wind power plant can be selected as a new energy power supply object, and abundant coal resources can be selected as a fossil fuel power supply object. And the power supply source of the target power-requiring area is determined by counting the new energy power supply object and the fossil fuel energy power supply object according to the regional power generation system. Lays a foundation for subsequent collection of historical power generation data and establishment of a prediction model, realizes the confirmation process of a power supply object, and provides support for power supply and demand prediction.
Step S300: performing data integration on the historical power consumption data of the target power supply object to obtain target historical power supply information;
specifically, historical power generation data of the target power supply object is collected, and the data are integrated to form a target historical power supply information data set. The historical electricity consumption data of the target power supply object refers to the generated energy information of the new energy power supply object and the fossil fuel power supply object in a certain time range in the target electricity demand area, reflects the past power generation rule and the past change trend of the power supply object, and is basic data for constructing a power supply prediction model. The data integration refers to the process of collecting, cleaning, converting and storing historical power generation data, firstly, day, week, month and year generating capacity data of a target power supply object are acquired through on-site monitoring equipment or a power generation enterprise database, and then data cleaning operations such as unified format, abnormal value detection, missing value filling and the like are carried out on the data.
And converting the cleaned data into a predictive model input format, and storing the predictive model input format in a database or a data warehouse to form target historical power supply information. The data set contains the time-space change rule of the generated energy of the target power supply object within a certain duration, and is an important data base for power supply and demand prediction and generator set model training.
By the aid of the collection and processing process of the historical power generation data of the target power supply object, high-quality historical power generation data are provided for building an accurate prediction model, data support is provided for subsequent power demand prediction and wind power prediction model building, extraction of power generation change rules of each time-space scale of a power supply is facilitated, and a high-precision prediction model is built.
Step S400: carrying out power consumption demand analysis on the target power consumption area according to the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information to obtain a target power consumption predicted value;
specifically, the target electricity demand predicted value refers to the electricity demand of a target electricity demand region predicted and analyzed to obtain the electricity demand of the region in a future period of time. Based on historical meteorological data of a target electricity-requiring area, such as an ambient temperature parameter set and an ambient humidity parameter set, and historical power supply information of the area, such as a historical electricity consumption data set, and the like, an electricity consumption demand prediction model is built by adopting technologies such as machine learning, and can be a random forest model, a support vector machine model and the like. The prediction model establishes a mapping relation and a rule between the ambient temperature, the humidity and the electricity consumption of a target electricity-requiring area through learning and training a large amount of historical data, then inputs the temperature and the humidity predicted value in a certain time period into the model, and the model can output the corresponding electricity-requiring predicted value.
The obtained target electricity demand prediction value may predict the total electricity demand of the target electricity demand region in a long period of time, such as 1 year or 3 years, and may also predict the electricity demand in a finer time dimension, such as each month or each quarter. By constructing the electricity demand prediction model, the electricity demand in the area within a certain future time range is predicted and estimated, and a foundation is provided for the follow-up dispatching of electric energy.
Step S500: performing data processing on the target environmental meteorological information to construct a wind turbine generator power prediction model, and performing power supply prediction of the new energy power supply object to obtain a target wind power prediction result, wherein the target wind power prediction result comprises a middle wind power prediction result and a long-term wind power prediction result;
specifically, the power generation amount of a new energy power supply object, such as a wind turbine generator, is predicted to obtain wind power prediction results, including prediction results in two time dimensions of medium term and long term. Firstly, processing and screening target environmental weather information, and then selecting characteristic data related to the power generation of the wind turbine generator as input of a power prediction model of the motor set. And (3) constructing a wind turbine generator power prediction model by adopting a machine learning method based on the characteristic data set, such as a wavelet neural network model, a regression tree model, a BP neural network and the like. Through the study of a large number of historical wind turbine generator system generating data and environmental meteorological data, a mapping relation between the historical wind turbine generator system generating data and the environmental meteorological data is established. Then, the environmental weather predicted value in a certain time range in the future is input, and the model can predict the power generation amount of the wind turbine generator on different time scales, such as a mid-term predicted result in the future of 1-3 years and a long-term predicted result in the future of 3-5 years.
By constructing a wind turbine generator power prediction model and training the model based on target environmental weather information, the prediction of the generated energy of a new energy power supply object on different time scales is realized, and the accuracy of the new energy generated power prediction is improved.
Step S600: generating a target electricity demand gap based on the target electricity demand predicted value and the mid-term wind power predicted result;
specifically, based on a power demand predicted value of a target power demand region and a mid-term power generation predicted result of the wind turbine, the power demand gap amount of the target power demand region in a certain time range in the future is calculated. The electricity demand predictor predicts the overall electricity demand of the target electricity demand region over a future period of time, such as 1 year. The mid-term wind power prediction result predicts the power generation amount of the wind turbine generator in the future 1-3 years. Comparing and analyzing the two, and if the predicted power generation amount of the wind turbine in the time range can not meet the power consumption requirement of the power consumption region, generating a target power consumption gap, wherein the gap amount is obtained by subtracting the power generation amount of the wind turbine from the total power consumption requirement of the power consumption region.
The calculation and evaluation of the number of the electric gaps of the region in a certain future age are realized by comparing and analyzing the electric predicted value of the target electric region and the predicted result of the mid-term power generation of the wind turbine unit. If the electricity demand gap is larger, other measures are needed to make up, such as adding other power sources, improving the utilization rate of other power sources, implementing electricity demand side management and the like; if the gap amount is smaller, the fact that the target electricity-requiring area is in the middle time dimension means that the electricity generation amount of new energy sources such as wind power and the like can basically meet the electricity consumption requirement. And the grid-connected adaptation degree of new energy power generation is improved by calculating the target power demand gap, so that the economic benefit of the new energy power generation is improved.
Step S700: and setting a power demand supplement threshold value for the fossil power supply object based on the target power demand gap and the long-term wind power prediction result.
Specifically, the electricity demand gap quantity predicts the difference between the target electricity demand region and the generated energy of the wind turbine in the middle time dimension, and the long-term wind power prediction result predicts the generated energy of the wind turbine in 3-5 years or longer. Considering that the long-term power generation of the wind turbine generator has certain uncertainty, a certain power-demand supplement threshold value is required to be set for a fossil fuel power generation object in order to ensure the power utilization safety and stability of a target power-demand area. For example, a minimum power generation amount to be achieved by the thermal power plant, a minimum plant utilization hour of the thermal power plant, or the like may be set. On the premise of fully utilizing new energy power generation, the power supply gap is made up by fossil fuel power supply, so that the thermal power generation cost is reduced, and meanwhile, the stable operation of the power system is ensured.
If the actual power generation amount of the wind turbine generator is lower than the long-term prediction result, the power demand of the power demand area cannot be met, and the thermal power generation unit needs to start the unit to operate according to the power demand supplement threshold set in the earlier stage so as to make up the power demand gap, so that the wind power benefit can be exerted to the greatest extent, and the safety and stability of the operation of the power system are ensured. The method realizes the optimal coordination and complementary power generation of renewable energy and conventional energy, and ensures the normal power supply of a power supply area while reducing the thermal power generation cost.
Further, as shown in fig. 2, the embodiment of the present application further includes:
step S410: the electricity demand prediction model is pre-constructed, wherein the electricity demand analysis model comprises a data input layer, a data screening layer, a similarity calculation layer and a data output layer;
step S420: constructing a power consumption demand database based on the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information;
step S430: embedding the electricity demand database into the data screening layer of the electricity demand analysis model;
step S440: extracting and obtaining mid-term environment prediction data of the target power demand region based on a time sequence arrangement relation of the target environment temperature parameter set and the target environment humidity parameter set, wherein the mid-term environment prediction data comprises mid-term temperature prediction data and mid-term humidity prediction data;
step S450: and inputting the mid-term environment prediction data into the electricity demand prediction model for data analysis to obtain the target electricity demand prediction value.
Specifically, an electricity demand prediction model is constructed and used for carrying out prediction analysis on the electricity demand of a target electricity demand region, and the electricity demand prediction model comprises a data input layer, a data screening layer, a similarity calculation layer and a data output layer. And constructing an electricity demand database based on historical environment temperature and humidity parameters and historical electricity consumption information of the target electricity demand region, and recording electricity consumption characteristics and electricity consumption information of the target electricity demand region under different environment conditions. The constructed electricity demand database is embedded into a data screening layer of the constructed electricity demand prediction model, and the data screening layer can select data which is most matched with the current input environmental condition from a large amount of historical data for analysis. Based on the time sequence arrangement relation of the environmental temperature and humidity parameters of the target electricity-requiring area, the predicted value of the environmental temperature and humidity in a certain time range in the future, such as the future 1 year, is extracted and used as the input of a model to form mid-term environmental predicted data. And inputting the obtained mid-term environment prediction data into a power consumption demand prediction model for analysis, and finally predicting and obtaining the power consumption demand of the target power demand region in the future time range, namely a target power demand predicted value.
By constructing the electricity demand database and the prediction model and inputting the environment prediction value within a certain time range in the future, the prediction analysis of the electricity demand of the target electricity demand region within the time range is realized, and the technical effect of accurately predicting the new energy power generation by combining the environment information and the new energy power generation historical data is achieved.
Further, the embodiment of the application further comprises:
step S421: presetting a data acquisition period and a period data processing rule, wherein the period data processing rule comprises a temperature period data processing rule, a humidity period data processing rule and a power supply period data processing rule;
step S422: dividing the data of the target environmental temperature parameter set based on the data acquisition period to obtain K groups of environmental temperature change fluctuation values, wherein K is a positive integer;
step S423: carrying out data processing on the K groups of environmental temperature change fluctuation values by adopting the temperature cycle data processing rule to obtain K sample environmental temperature indexes;
step S424: performing data processing on the target environment humidity parameter set based on the data acquisition period and the humidity period data processing rule to obtain K sample environment humidity indexes;
Step S425: performing data processing on the target historical power supply information based on the data acquisition period and the power supply period data processing rule to obtain K sample power demand indexes;
step S426: constructing mapping relations among the K sample environment temperature indexes, the K sample environment humidity indexes and the K sample electricity demand indexes to obtain K groups of sample electricity demand indexes, sample environment temperature indexes and sample environment humidity indexes;
step S427: and constructing the electricity demand database by taking the K groups of sample electricity demand indexes, the sample environmental temperature indexes and the sample environmental humidity indexes as model construction data and adopting a knowledge graph.
Specifically, the data acquisition period is preset, such as daily or weekly, and the period data processing rules corresponding to different data types, such as the processing rules of temperature data, humidity data, electricity consumption data and the like, provide a time reference for subsequent data extraction and processing. According to the data acquisition period, the historical environmental temperature parameter data set of the target power-requiring area is divided into K groups of data, each group represents the temperature change condition in the same time period, and K is a positive integer.
And processing the obtained K groups of temperature data according to a temperature period data processing rule, extracting main characteristics of temperature change in each group, and forming K sample environment temperature indexes, wherein each index can represent the temperature change type in a certain data acquisition period. And processing the obtained K groups of humidity data according to a humidity period data processing rule, extracting main characteristics of humidity change in each group, and forming K sample environment humidity indexes, wherein each index can represent the humidity change type in a certain data acquisition period. And processing the obtained K groups of power supply data according to a power supply period data processing rule, extracting main characteristics of power supply change in each group, and obtaining K sample power demand indexes, wherein each index can represent the power supply change type in a certain data acquisition period.
And constructing and obtaining mapping relations among K sample environment temperature indexes, K sample environment humidity indexes and K sample electricity-required indexes to form K groups of mapping relation combinations so as to establish corresponding relations among temperature, humidity and power supply. Based on the constructed mapping relation combination, a power consumption demand database is constructed by adopting technologies such as a knowledge graph and the like, and complex corresponding relations between different environmental conditions and power supply of a target power consumption area are recorded.
For example, 1 week is selected as a reference period, and meanwhile, processing rules in the period of temperature data, humidity data and electricity consumption data are preset as mean value calculation rules and fluctuation rate calculation rules; the target power demand region was divided into 1560 groups of 30 years old environmental temperature data set, 30 years old environmental humidity data set, and 1560 groups, each representing temperature change data, humidity change data, and power supply change data within 1 week. Calculating characteristic values such as the average value and the fluctuation rate of each group of data according to a temperature data processing rule, and respectively forming 1560 sample environment temperature indexes, environment humidity indexes and electricity demand indexes, wherein the characteristic values represent the temperature change characteristic, the humidity change characteristic and the power supply change characteristic within 1 week; and establishing a corresponding relation among three indexes, namely, establishing a forward mapping relation among the three indexes if the temperature fluctuation rate and the humidity fluctuation rate are both larger and the power supply quantity is also changed greatly in a week, finally obtaining 1560 group of corresponding mapping relation combinations, constructing a power consumption demand database in the form of a knowledge graph based on 1560 group of mapping relation combinations, recording complex corresponding relation between the power consumption demand database and different environmental conditions, providing data support for the model, and greatly improving the model prediction precision.
By extracting, processing and constructing the mapping relation of the historical data, modeling of a complex corresponding relation between an environmental factor and electricity demand is realized, deep mining and associated modeling between the environmental temperature, the humidity and the electricity supply are realized, and the accuracy of new energy power generation prediction is greatly improved, so that the benefit of new energy power generation is improved.
Further, the embodiment of the application further comprises:
step S451: presetting a data screening threshold, wherein the data screening threshold comprises a temperature data screening threshold and a humidity data screening threshold;
step S452: generating a target temperature screening interval based on the temperature data screening threshold and the mid-term temperature prediction data;
step S453: generating a target humidity screening interval based on the humidity data screening threshold and the mid-term humidity prediction data;
step S454: embedding the target temperature screening interval and the target humidity screening interval into the data screening layer of the electricity demand analysis model;
step S455: screening the electricity demand database in the data screening layer of the electricity demand prediction model to obtain a data screening result, wherein the data screening result is M groups of sample electricity demand indexes, sample environment temperature indexes and sample environment humidity indexes, and M is a positive integer smaller than K;
Step S456: in the similarity calculation layer, calculating data similarity of the medium-term environment prediction data and the data screening result based on Euclidean distance to obtain M environment similarity indexes;
step S457: and carrying out serialization processing on the M environmental similarity indexes, and extracting and obtaining a sample electricity demand index corresponding to the minimum value to serve as the target electricity demand predicted value.
Specifically, a screening threshold value of the temperature data and the humidity data is preset for later screening data from the electricity demand database. And generating a target temperature screening interval which represents a temperature range for data screening based on the temperature data screening threshold value and the obtained medium-term temperature prediction data. And generating a target humidity screening interval according to the humidity data and the intermediate humidity prediction data, similar to the generation of the target temperature screening interval.
And embedding the target temperature screening interval and the target humidity screening interval into a data screening layer of the constructed electricity demand prediction model. And in a data screening layer of the model, screening data from the electricity demand database according to the two screening intervals to obtain a data screening result, wherein the data screening result comprises M groups of mapping relation combinations, and M is smaller than the total number K of the database. And in a similarity calculation layer of the model, calculating the similarity between the medium-term environmental prediction data and the obtained data screening result based on the Euclidean distance to obtain M environmental similarity indexes.
The Euclidean distance calculation formula is as follows:
wherein distance (X, Y) is an environmental similarity index, X 1 X is the predicted data of the medium-term temperature 2 For mid-term humidity prediction data, Y 1 For the purpose of sampleThe index of the environmental temperature, Y 2 Is the environmental humidity index of the sample.
And sequencing the obtained M environmental similarity indexes, and selecting data with minimum similarity, wherein the corresponding electricity demand index is used as a final target electricity demand predicted value.
By setting the data screening conditions and the threshold value, the data which is most matched with the current input environmental conditions is filtered from a large-scale historical database to perform model calculation, the calculation efficiency can be improved by using the filtering mechanism, the influence of noise data on the result is eliminated, and the prediction precision of new energy power generation is improved.
Further, the embodiment of the application further comprises:
step S510: performing data integration on the historical power generation data of the new energy power supply object to obtain a historical wind power data set;
step S520: extracting and obtaining a historical wind speed data set and a historical wind direction data set based on the target environmental weather information;
step S530: presetting a data acquisition node, and carrying out data collaborative acquisition on the historical wind power data set, the historical wind speed data set and the historical wind direction data set based on the data acquisition node to obtain a sample data set, wherein the sample data set comprises N groups of sample wind power data, sample wind speed data and sample wind direction data;
Step S540: the wind turbine generator power prediction model is built based on a wavelet neural network model, input data of the wind turbine generator power prediction model are wind speed data and wind direction data, and an output result is a wind power prediction value;
step S550: initializing the wind turbine power prediction model, performing network training and network testing by adopting the sample data set, and stopping training of the wind turbine power prediction model when the error between the wind power predicted value output by the wind turbine power prediction model and the sample wind power data is smaller than a preset threshold value.
Specifically, historical power generation data of the wind turbine generator are integrated to obtain a historical wind power data set, and generating capacity information of the wind turbine generator in different periods is recorded. Based on the historical environmental meteorological information of the target electricity-requiring area, a historical wind speed data set and a historical wind direction data set are extracted, and wind speed and wind direction change information in different periods is recorded respectively. The method comprises the steps of presetting data acquisition nodes, for example, automatically acquiring at small intervals, and then synchronously acquiring data of three historical data sets based on the nodes to obtain a sample data set, wherein the sample data set comprises N groups of wind power data, wind speed data and wind direction data, and each combination represents three data characteristics of the same time node.
And constructing a wind turbine generator power prediction model based on the wavelet neural network model, wherein the input of the model is wind speed and wind direction data, and the output result is a wind power prediction value. Firstly, carrying out wavelet analysis on historical power generation data, wind speed data and wind direction data of a wind turbine generator, extracting feature vectors after wavelet transformation, removing data noise, detecting trend and periodic components in the data, providing more abundant input information for a neural network, and then setting the node number of an input layer, a hidden layer and an output layer of the neural network. The number of nodes at the input layer is the characteristic vector dimension of wind speed and wind direction data, the number of nodes at the output layer is 1, the wind power predicted value is represented, and the number of nodes at the hidden layer is 5-20.
Randomly initializing connection weight and threshold of the neural network model, selecting a Sigmoid function as an activation function, setting model learning rate by adopting an exponential decay method, and performing network training by adopting an error back propagation algorithm and a gradient descent method. And inputting the obtained sample data set into a model for training, and stopping model training when the error between the wind power predicted value output by the model and the sample wind power data is smaller than a preset threshold value.
Through collection and integration of historical data, a wind power prediction model is constructed, model training is carried out by adopting sample data, mining and modeling of a power change rule of a wind turbine generator are achieved, the wind power can be predicted by the trained model through inputting data such as wind speed and wind direction, intelligent prediction of new energy power generation is achieved, and prediction accuracy of power generation is improved.
Further, as shown in fig. 3, the embodiment of the present application further includes:
step S541: based on the time sequence arrangement relation of the data in the target environmental weather information, extracting and obtaining middle-stage weather forecast data of the target electricity demand region from the target environmental weather information, wherein the middle-stage weather forecast data comprises middle-stage wind speed forecast data and middle-stage wind direction forecast data;
step S542: and inputting the predicted data of the middle wind speed and the predicted data of the middle wind direction into the power prediction model of the wind turbine generator to perform power supply prediction of the new energy power supply object, so as to obtain a predicted result of the middle wind power.
Specifically, based on the historical environmental weather information of the target electricity-requiring area and the time sequence rule thereof, wind speed and wind direction prediction data in a certain time range in the future, such as the future 1 year, are extracted and input as a model, and these data form middle weather prediction data. And (3) extracting and obtaining wind speed and direction prediction information in a future period by adopting technical means such as a time sequence model of environmental meteorological data and the like, so as to realize key input of wind power prediction. And inputting the extracted mid-term wind speed prediction data and mid-term wind direction prediction data into a wind turbine generator power prediction model obtained through training, and predicting the generated energy of the wind turbine generator in the future time range of the target electricity demand region based on the input data, namely, a mid-term wind power prediction result, so as to realize calculation and prediction of wind power. The model integrates the mapping relation between the environment information and the wind turbine generator, and can effectively improve the prediction precision.
Through the prediction of environmental meteorological data and the training and application of a machine learning model, the wind power change rule is excavated and predicted, the stability and accuracy of the prediction of new energy power generation are greatly improved, and support is provided for the follow-up prediction of new energy power generation to reduce the thermal power generation cost and improve the benefit of new energy power generation.
Further, as shown in fig. 3, the embodiment of the present application further includes:
step S543: presetting a time sequence data value interval;
step S544: based on the time sequence arrangement relation of the data in the target environmental weather information, extracting and obtaining a weather prediction data set conforming to the preset time sequence data value interval from the target environmental weather information;
step S545: performing periodic division on the weather forecast data set to obtain H periodic weather forecast data;
step S546: inputting the H period weather forecast data into the wind turbine generator power forecast model to conduct power supply forecast, and obtaining H period wind power forecast results;
step S547: calculating and obtaining the data fluctuation indexes of the wind power prediction results of the H periods according to the time sequence relation;
step S548: optimizing and adjusting the mid-term wind power prediction result based on the data fluctuation index to obtain the long-term wind power prediction result;
And the mid-term wind power prediction result and the long-term wind power prediction result form the target wind power prediction result.
Specifically, a value range of time sequence data is preset, such as 1 hour or 1 day, for the subsequent division of the environmental meteorological data set; and extracting a weather forecast data set conforming to a preset interval from the historical environmental weather information based on the data time sequence rule of the historical environmental weather information of the target electricity-requiring area, wherein the data set records wind speed and wind direction forecast information in a future time. Dividing the obtained weather forecast data sets according to preset time intervals to obtain data sets of H time periods, wherein each data set represents forecast information in one time interval, and H is a positive integer and represents the divided time period number. And inputting the weather forecast data sets of the H time periods into a wind turbine generator power forecast model obtained through training to obtain forecast results of wind power in the H time periods, and forming the forecast results of wind power in the H periods.
According to the time sequence relation among the wind power prediction results of the H periods, calculating the data fluctuation degree of the wind power prediction results of the H periods, obtaining a sequence of data fluctuation indexes in the H periods, and carrying out optimization adjustment on the obtained wind power prediction results of the middle period based on the change rule of the data fluctuation indexes in the H periods, so as to obtain a wind power prediction result of the long period.
The wind power generation method comprises the steps that a medium-term wind power prediction result and a long-term wind power prediction result form a final target wind power prediction result, the power generation quantity change condition of a wind turbine generator in a certain period of time is predicted in a target power demand area, and the prediction analysis of the wind power long-term change trend is realized through time sequence division and model prediction of environmental weather prediction data. By adopting the technologies of data mining, machine learning and the like, the time sequence rule in the environmental meteorological data is extracted, and the prediction calculation of wind power change on multiple time scales is realized by combining a learning model, so that the accuracy of a prediction result is greatly improved, and the medium-long term accurate prediction and analysis of the new energy power generation amount are realized.
In summary, the medium-long term power generation capacity prediction method based on new energy power generation provided by the embodiment of the application has the following technical effects:
acquiring target environmental weather information according to target geographical position information of a target electricity demand area, wherein the target environmental weather information comprises a target environmental temperature parameter set, a target environmental humidity parameter set and a target environmental wind power parameter set, and is important input data for carrying out subsequent electricity demand prediction and wind power prediction. Obtaining a target power supply object based on a target power demand area, wherein the target power supply object comprises a new energy power supply object and a fossil power supply object, and a prediction object is provided for subsequent demand prediction and power generation prediction; carrying out data integration on historical electricity consumption data of a target power supply object to obtain target historical power supply information, and constructing an electricity consumption demand prediction model; carrying out power demand analysis on a target power demand region according to a target environment temperature parameter set, a target environment humidity parameter set and target historical power supply information to obtain a target power demand predicted value, and providing an important reference for power supply and demand analysis and power generation scheduling of a power system; carrying out data processing on the target environmental meteorological information to construct a wind turbine generator power prediction model, and carrying out power supply prediction of a new energy power supply object to obtain a target wind power prediction result, wherein the target wind power prediction result comprises a middle wind power prediction result and a long-term wind power prediction result, and predicts the generated energy of new energy wind power; generating a target electricity demand gap based on the target electricity demand predicted value and the medium-term wind power predicted result, and providing a reference basis for the supplement of thermal power generation; the power supply system has the advantages that the power supply system sets the power supply supplement threshold value for the fossil power supply object based on the target power demand gap and the long-term wind power prediction result, is favorable for optimizing and configuring various power generation resources of the power system, ensures supply and demand balance, realizes the accurate prediction of new energy power generation by combining environment information and new energy power generation historical data, and performs thermal power generation control by combining the power consumption prediction result of the power supply area, so that the normal power supply of the power supply area is ensured while the thermal power generation cost is reduced.
Example two
Based on the same inventive concept as the method for predicting the medium-long term power generation based on new energy power generation in the foregoing embodiments, as shown in fig. 4, an embodiment of the present application provides a system for predicting the medium-long term power generation based on new energy power generation, including:
the environmental weather information module 11 is configured to acquire target environmental weather information according to target geographical location information of a target electricity-requiring area, where the target environmental weather information includes a target environmental temperature parameter set, a target environmental humidity parameter set, and a target environmental wind power parameter set;
a target power supply object module 12 for obtaining a target power supply object based on the target power demand region, wherein the target power supply object includes a new energy power supply object and a fossil power supply object;
the historical power supply information module 13 is used for carrying out data integration on the historical power consumption data of the target power supply object to obtain target historical power supply information;
the electricity demand analysis module 14 is configured to perform electricity demand analysis on the target electricity demand area according to the target environmental temperature parameter set, the target environmental humidity parameter set and the target historical power supply information, so as to obtain a target electricity demand predicted value;
The wind power prediction result module 15 is configured to perform data processing on the target environmental weather information to construct a wind turbine generator power prediction model, and perform power supply prediction of the new energy power supply object to obtain a target wind power prediction result, where the target wind power prediction result includes a mid-term wind power prediction result and a long-term wind power prediction result;
a target electricity demand gap module 16 for generating a target electricity demand gap based on the target electricity demand predicted value and the mid-term wind power predicted result;
the electricity demand replenishment threshold module 17 sets an electricity demand replenishment threshold for the fossil power supply object based on the target electricity demand gap and the long-term wind power prediction result.
Further, the embodiment of the application further comprises:
the prediction model construction module is used for pre-constructing an electricity demand prediction model, wherein the electricity demand analysis model comprises a data input layer, a data screening layer, a similarity calculation layer and a data output layer;
the demand database construction module is used for constructing a power consumption demand database based on the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information;
the database embedding module is used for embedding the electricity consumption requirement database into the data screening layer of the electricity consumption requirement analysis model;
The medium-term environment prediction module is used for extracting and obtaining medium-term environment prediction data of the target electricity-required area based on a time sequence arrangement relation of the target environment temperature parameter set and the target environment humidity parameter set, wherein the medium-term environment prediction data comprises medium-term temperature prediction data and medium-term humidity prediction data;
and the target electricity demand predicted value module is used for inputting the medium-term environment predicted data into the electricity demand predicted model for data analysis to obtain the target electricity demand predicted value.
Further, the embodiment of the application further comprises:
the rule presetting module is used for presetting a data acquisition period and a period data processing rule, wherein the period data processing rule comprises a temperature period data processing rule, a humidity period data processing rule and a power supply period data processing rule;
the data dividing module is used for carrying out data division on the target environment temperature parameter set based on the data acquisition period to obtain K groups of environment temperature change fluctuation values, wherein K is a positive integer;
the environmental temperature index module is used for carrying out data processing on the K groups of environmental temperature change fluctuation values by adopting the temperature period data processing rule to obtain K sample environmental temperature indexes;
The environment humidity index module is used for carrying out data processing on the target environment humidity parameter set based on the data acquisition period and the humidity period data processing rule to obtain K sample environment humidity indexes;
the sample electricity demand index module is used for carrying out data processing on the target historical power supply information based on the data acquisition period and the power supply period data processing rule to obtain K sample electricity demand indexes;
the mapping relation construction module is used for constructing the mapping relation among the K sample environment temperature indexes, the K sample environment humidity indexes and the K sample electricity demand indexes to obtain K groups of sample electricity demand indexes, sample environment temperature indexes and sample environment humidity indexes;
and the electricity demand database module is used for constructing the electricity demand database by taking the K groups of sample electricity demand indexes, the sample environment temperature indexes and the sample environment humidity indexes as model construction data and adopting a knowledge graph.
Further, the embodiment of the application further comprises:
the threshold value presetting module is used for presetting a data screening threshold value, wherein the data screening threshold value comprises a temperature data screening threshold value and a humidity data screening threshold value;
a temperature screening interval module for generating a target temperature screening interval based on the temperature data screening threshold and the medium-term temperature prediction data;
The humidity screening interval module is used for generating a target humidity screening interval based on the humidity data screening threshold value and the medium-term humidity prediction data;
the data screening layer module is used for embedding the target temperature screening interval and the target humidity screening interval into the data screening layer of the electricity demand analysis model;
the data screening result module is used for screening and obtaining data screening results from the electricity demand database in the data screening layer of the electricity demand prediction model, wherein the data screening results are M groups of sample electricity demand indexes, sample environment temperature indexes and sample environment humidity indexes, and M is a positive integer smaller than K;
the environment similarity index module is used for calculating data similarity of the medium-term environment prediction data and the data screening result based on Euclidean distance in the similarity calculation layer to obtain M environment similarity indexes;
and the electricity demand predicted value module is used for carrying out serialization processing on the M environment similarity indexes, extracting and obtaining a sample electricity demand index corresponding to the minimum value, and taking the sample electricity demand index as the target electricity demand predicted value.
Further, the embodiment of the application further comprises:
the power data set module is used for carrying out data integration on the historical power generation data of the new energy power supply object to obtain a historical wind power data set;
The wind speed and direction data set module is used for extracting and obtaining a historical wind speed data set and a historical wind direction data set based on the target environment meteorological information;
the sample data set module is used for presetting a data acquisition node, and carrying out data collaborative acquisition on the historical wind power data set, the historical wind speed data set and the historical wind direction data set based on the data acquisition node to obtain a sample data set, wherein the sample data set comprises N groups of sample wind power data, sample wind speed data and sample wind direction data;
the wind power prediction value module is used for constructing a wind turbine generator power prediction model based on a wavelet neural network model, wherein input data of the wind turbine generator power prediction model are wind speed data and wind direction data, and an output result is a wind power prediction value;
and the network training test module is used for initializing the wind turbine power prediction model, carrying out network training and network testing by adopting the sample data set, and stopping training of the wind turbine power prediction model when the error between the wind power predicted value output by the wind turbine power prediction model and the sample wind power data is smaller than a preset threshold value.
Further, the embodiment of the application further comprises:
the predicted data extraction module is used for extracting and obtaining middle-stage weather predicted data of the target electricity-required area from the target environmental weather information based on a time sequence arrangement relation of data in the target environmental weather information, wherein the middle-stage weather predicted data comprises middle-stage wind speed predicted data and middle-stage wind direction predicted data;
and the power supply prediction module is used for inputting the mid-term wind speed prediction data and mid-term wind direction prediction data into the wind turbine generator power prediction model to perform power supply prediction of the new energy power supply object, so as to obtain the mid-term wind power prediction result.
Further, the embodiment of the application further comprises:
the value interval presetting module is used for presetting a time sequence data value interval;
the data set extraction module is used for extracting and obtaining a weather prediction data set which accords with the preset time sequence data value interval from the target environmental weather information based on the time sequence arrangement relation of the data in the target environmental weather information;
the weather forecast data module is used for carrying out periodic division on the weather forecast data set to obtain H periodic weather forecast data;
the wind power prediction result module inputs the H period weather prediction data into the wind turbine generator power prediction model to perform power supply prediction, so as to obtain H period wind power prediction results;
The data fluctuation index module is used for calculating and obtaining the data fluctuation indexes of the wind power prediction results of the H periods according to the time sequence relation;
the wind power prediction result module is used for optimizing and adjusting the mid-term wind power prediction result based on the data fluctuation index to obtain the long-term wind power prediction result;
and the mid-term wind power prediction result and the long-term wind power prediction result form the target wind power prediction result.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any method for implementing an embodiment of the present application, without unnecessary limitations.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.

Claims (8)

1. The method for predicting the medium-and-long-term power generation capacity based on new energy power generation is characterized by comprising the following steps of:
acquiring target environmental weather information according to target geographical position information of a target electricity-requiring area, wherein the target environmental weather information comprises a target environmental temperature parameter set, a target environmental humidity parameter set and a target environmental wind power parameter set;
Obtaining a target power supply object based on the target power demand region, wherein the target power supply object comprises a new energy power supply object and a fossil power supply object;
performing data integration on the historical power consumption data of the target power supply object to obtain target historical power supply information;
carrying out power consumption demand analysis on the target power consumption area according to the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information to obtain a target power consumption predicted value;
performing data processing on the target environmental meteorological information to construct a wind turbine generator power prediction model, and performing power supply prediction of the new energy power supply object to obtain a target wind power prediction result, wherein the target wind power prediction result comprises a middle wind power prediction result and a long-term wind power prediction result;
generating a target electricity demand gap based on the target electricity demand predicted value and the mid-term wind power predicted result;
and setting a power demand supplement threshold value for the fossil power supply object based on the target power demand gap and the long-term wind power prediction result.
2. The method of claim 1, wherein the target electricity demand area is subjected to electricity demand analysis according to the target ambient temperature parameter set, the target ambient humidity parameter set, and the target historical power supply information to obtain a target electricity demand predicted value, the method further comprising:
The electricity demand prediction model is pre-constructed, wherein the electricity demand analysis model comprises a data input layer, a data screening layer, a similarity calculation layer and a data output layer;
constructing a power consumption demand database based on the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information;
embedding the electricity demand database into the data screening layer of the electricity demand analysis model;
extracting and obtaining mid-term environment prediction data of the target power demand region based on a time sequence arrangement relation of the target environment temperature parameter set and the target environment humidity parameter set, wherein the mid-term environment prediction data comprises mid-term temperature prediction data and mid-term humidity prediction data;
and inputting the mid-term environment prediction data into the electricity demand prediction model for data analysis to obtain the target electricity demand prediction value.
3. The method of claim 2, wherein a power demand database is constructed based on the target ambient temperature parameter set, the target ambient humidity parameter set, and the target historical power supply information, the method further comprising:
presetting a data acquisition period and a period data processing rule, wherein the period data processing rule comprises a temperature period data processing rule, a humidity period data processing rule and a power supply period data processing rule;
Dividing the data of the target environmental temperature parameter set based on the data acquisition period to obtain K groups of environmental temperature change fluctuation values, wherein K is a positive integer;
carrying out data processing on the K groups of environmental temperature change fluctuation values by adopting the temperature cycle data processing rule to obtain K sample environmental temperature indexes;
performing data processing on the target environment humidity parameter set based on the data acquisition period and the humidity period data processing rule to obtain K sample environment humidity indexes;
performing data processing on the target historical power supply information based on the data acquisition period and the power supply period data processing rule to obtain K sample power demand indexes;
constructing mapping relations among the K sample environment temperature indexes, the K sample environment humidity indexes and the K sample electricity demand indexes to obtain K groups of sample electricity demand indexes, sample environment temperature indexes and sample environment humidity indexes;
and constructing the electricity demand database by taking the K groups of sample electricity demand indexes, the sample environmental temperature indexes and the sample environmental humidity indexes as model construction data and adopting a knowledge graph.
4. The method of claim 3, wherein the mid-term environmental prediction data is input into the electricity demand prediction model for data analysis to obtain the target electricity demand prediction value, the method further comprising:
Presetting a data screening threshold, wherein the data screening threshold comprises a temperature data screening threshold and a humidity data screening threshold;
generating a target temperature screening interval based on the temperature data screening threshold and the mid-term temperature prediction data;
generating a target humidity screening interval based on the humidity data screening threshold and the mid-term humidity prediction data;
embedding the target temperature screening interval and the target humidity screening interval into the data screening layer of the electricity demand analysis model;
screening the electricity demand database in the data screening layer of the electricity demand prediction model to obtain a data screening result, wherein the data screening result is M groups of sample electricity demand indexes, sample environment temperature indexes and sample environment humidity indexes, and M is a positive integer smaller than K;
in the similarity calculation layer, calculating data similarity of the medium-term environment prediction data and the data screening result based on Euclidean distance to obtain M environment similarity indexes;
and carrying out serialization processing on the M environmental similarity indexes, and extracting and obtaining a sample electricity demand index corresponding to the minimum value to serve as the target electricity demand predicted value.
5. The method of claim 1, wherein the target environmental weather information is subjected to data processing to construct a wind turbine generator power prediction model, power supply prediction of the new energy power supply object is performed, and a target wind power prediction result is obtained, and the method further comprises:
performing data integration on the historical power generation data of the new energy power supply object to obtain a historical wind power data set;
extracting and obtaining a historical wind speed data set and a historical wind direction data set based on the target environmental weather information;
presetting a data acquisition node, and carrying out data collaborative acquisition on the historical wind power data set, the historical wind speed data set and the historical wind direction data set based on the data acquisition node to obtain a sample data set, wherein the sample data set comprises N groups of sample wind power data, sample wind speed data and sample wind direction data;
the wind turbine generator power prediction model is built based on a wavelet neural network model, input data of the wind turbine generator power prediction model are wind speed data and wind direction data, and an output result is a wind power prediction value;
initializing the wind turbine power prediction model, performing network training and network testing by adopting the sample data set, and stopping training of the wind turbine power prediction model when the error between the wind power predicted value output by the wind turbine power prediction model and the sample wind power data is smaller than a preset threshold value.
6. The method of claim 5, wherein the method further comprises:
based on the time sequence arrangement relation of the data in the target environmental weather information, extracting and obtaining middle-stage weather forecast data of the target electricity demand region from the target environmental weather information, wherein the middle-stage weather forecast data comprises middle-stage wind speed forecast data and middle-stage wind direction forecast data;
and inputting the predicted data of the middle wind speed and the predicted data of the middle wind direction into the power prediction model of the wind turbine generator to perform power supply prediction of the new energy power supply object, so as to obtain a predicted result of the middle wind power.
7. The method of claim 6, wherein the method further comprises:
presetting a time sequence data value interval;
based on the time sequence arrangement relation of the data in the target environmental weather information, extracting and obtaining a weather prediction data set conforming to the preset time sequence data value interval from the target environmental weather information;
performing periodic division on the weather forecast data set to obtain H periodic weather forecast data;
inputting the H period weather forecast data into the wind turbine generator power forecast model to conduct power supply forecast, and obtaining H period wind power forecast results;
Calculating and obtaining the data fluctuation indexes of the wind power prediction results of the H periods according to the time sequence relation;
optimizing and adjusting the mid-term wind power prediction result based on the data fluctuation index to obtain the long-term wind power prediction result;
and the mid-term wind power prediction result and the long-term wind power prediction result form the target wind power prediction result.
8. A medium-to-long term power generation capacity prediction system based on new energy power generation, the system comprising:
the environment weather information module is used for acquiring target environment weather information according to target geographical position information of a target electricity-requiring area, wherein the target environment weather information comprises a target environment temperature parameter set, a target environment humidity parameter set and a target environment wind power parameter set;
the target power supply object module is used for obtaining a target power supply object based on the target power demand area, wherein the target power supply object comprises a new energy power supply object and a fossil power supply object;
the historical power supply information module is used for carrying out data integration on the historical power consumption data of the target power supply object to obtain target historical power supply information;
The electricity demand analysis module is used for carrying out electricity demand analysis on the target electricity demand area according to the target environment temperature parameter set, the target environment humidity parameter set and the target historical power supply information to obtain a target electricity demand predicted value;
the wind power prediction result module is used for carrying out data processing on the target environmental weather information to construct a wind turbine generator power prediction model, carrying out power supply prediction of the new energy power supply object and obtaining a target wind power prediction result, wherein the target wind power prediction result comprises a middle-term wind power prediction result and a long-term wind power prediction result;
the target electricity demand gap module generates a target electricity demand gap based on the target electricity demand predicted value and the mid-term wind power predicted result;
and the electricity-required replenishment threshold module sets an electricity-required replenishment threshold for the fossil power supply object based on the target electricity-required gap and the long-term wind power prediction result.
CN202310553483.7A 2023-05-17 2023-05-17 Medium-and-long-term power generation capacity prediction method and system based on new energy power generation Pending CN116865236A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310553483.7A CN116865236A (en) 2023-05-17 2023-05-17 Medium-and-long-term power generation capacity prediction method and system based on new energy power generation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310553483.7A CN116865236A (en) 2023-05-17 2023-05-17 Medium-and-long-term power generation capacity prediction method and system based on new energy power generation

Publications (1)

Publication Number Publication Date
CN116865236A true CN116865236A (en) 2023-10-10

Family

ID=88222262

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310553483.7A Pending CN116865236A (en) 2023-05-17 2023-05-17 Medium-and-long-term power generation capacity prediction method and system based on new energy power generation

Country Status (1)

Country Link
CN (1) CN116865236A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117375106A (en) * 2023-10-11 2024-01-09 揭阳前詹风电有限公司 Offshore wind power construction management method and system based on Internet of things
CN117410981A (en) * 2023-11-01 2024-01-16 中嘉能(广东)能源有限公司 Multi-region electric power energy intelligent scheduling method and system based on big data
CN117713217A (en) * 2023-12-27 2024-03-15 华能济南黄台发电有限公司 Coordination control method and system for thermal power generation system
CN117713039A (en) * 2023-11-01 2024-03-15 宁夏青铜峡市华能雷避窑光伏发电有限公司 Power plant power generation control method based on regional new energy power generation prediction
CN117833361A (en) * 2024-01-02 2024-04-05 中国电力工程顾问集团有限公司 Coal-fired power plant depth peak shaving prediction method and device based on time sequence analysis
CN117996757A (en) * 2024-04-07 2024-05-07 南京中核能源工程有限公司 Distributed wind power based power distribution network scheduling method, device and storage medium

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117375106A (en) * 2023-10-11 2024-01-09 揭阳前詹风电有限公司 Offshore wind power construction management method and system based on Internet of things
CN117375106B (en) * 2023-10-11 2024-04-09 揭阳前詹风电有限公司 Offshore wind power construction management method and system based on Internet of Things
CN117410981A (en) * 2023-11-01 2024-01-16 中嘉能(广东)能源有限公司 Multi-region electric power energy intelligent scheduling method and system based on big data
CN117713039A (en) * 2023-11-01 2024-03-15 宁夏青铜峡市华能雷避窑光伏发电有限公司 Power plant power generation control method based on regional new energy power generation prediction
CN117410981B (en) * 2023-11-01 2024-05-17 中嘉能(广东)能源有限公司 Multi-region electric power energy intelligent scheduling method and system based on big data
CN117713217A (en) * 2023-12-27 2024-03-15 华能济南黄台发电有限公司 Coordination control method and system for thermal power generation system
CN117833361A (en) * 2024-01-02 2024-04-05 中国电力工程顾问集团有限公司 Coal-fired power plant depth peak shaving prediction method and device based on time sequence analysis
CN117996757A (en) * 2024-04-07 2024-05-07 南京中核能源工程有限公司 Distributed wind power based power distribution network scheduling method, device and storage medium

Similar Documents

Publication Publication Date Title
CN116865236A (en) Medium-and-long-term power generation capacity prediction method and system based on new energy power generation
CN107341569B (en) Photovoltaic power prediction method combining photovoltaic power physical model and data driving
CN109934395B (en) Multi-hydropower-region short-term power load prediction method based on time-sharing and regional meteorological data
CN104899665A (en) Wind power short-term prediction method
CN110288136A (en) Wind power multi-step Predictive Model method for building up
Melzi et al. Hourly solar irradiance forecasting based on machine learning models
CN108196317B (en) Meteorological prediction method for micro-grid system
CN108075471B (en) Multi-objective constraint optimization power grid scheduling strategy based on stochastic power output prediction
CN117013527A (en) Distributed photovoltaic power generation power prediction method
CN115481918A (en) Active sensing and predictive analysis system for unit state based on source network load storage
CN114021848A (en) Generating capacity demand prediction method based on LSTM deep learning
CN116454928A (en) Multi-type energy storage cooperative scheduling method considering multiple time scales
CN117277304A (en) Photovoltaic power generation ultra-short-term power prediction method and system considering sunrise and sunset time
CN116914719A (en) Photovoltaic power station power prediction method based on space-time diagram network
CN112116127B (en) Photovoltaic power prediction method based on association of meteorological process and power fluctuation
Raju et al. IOT based solar energy prophecy using RNN architecture
CN114626604A (en) Distributed photovoltaic observation method and system based on reference station perception
KR20220021749A (en) System for generating energy for smart farms and method for building the same
CN113537575B (en) Trend load prediction method containing distributed photovoltaic and electric automobile grid connection
Mantri et al. Solar Power Generation Prediction for Better Energy Efficiency using Machine Learning
Yin et al. Photovoltaic power prediction model based on empirical mode decomposition-long-short memory neural network
CN117996757B (en) Distributed wind power based power distribution network scheduling method, device and storage medium
CN117913866B (en) Energy storage system based on photovoltaic power generation
Huang et al. Optimizing battery energy storage prototypes for improved resilience in commercial buildings: Gaussian mixture modeling and hierarchical analysis of energy storage potential
CN117725822A (en) LSTM-based short-term wind power prediction method and system

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