Wind and light same-field power station yield prediction model based on mechanism electricity price prediction
By constructing a wind and light same-field power station yield prediction model based on mechanism electricity price prediction, combining meteorological data and mechanism electricity price, the problem of insufficient prediction accuracy in the existing technology is solved, and more accurate yield prediction and operational optimization are achieved.
Patent Information
- Application Number
- CN202510494156.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-20
- Publication Date
- 2025-07-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing electricity price prediction methods fail to fully consider meteorological data and complex mechanisms of electricity price mechanisms, resulting in insufficient accuracy in forecasting yields of wind and sun power plants in the same field.
Build a wind and light same-field power station yield prediction model based on mechanism electricity price prediction, including meteorological data acquisition and processing, mechanism electricity price prediction, power station power generation power prediction and cash flow prediction module, and use meteorological big data and machine learning algorithms combined with power market trading rules to calculate indicators such as net present value and internal rate of return.
It improves the accuracy of the yield prediction of Fengguang power stations in the same field, provides scientific investment decision-making basis, optimizes operation strategies, and reduces operating costs.
Smart Images
Figure CN120278747A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electricity price forecasting, and particularly to a forecasting model for the return rate of a wind-solar co-located power station based on mechanism electricity price forecasting. Background Art
[0002] With the continuous growth of the global demand for clean energy, the wind-solar co-located power station, as an efficient power generation mode for utilizing natural resources, has received extensive attention and development. However, the return rate of the wind-solar co-located power station is affected by various factors, and the fluctuation of the electricity price is one of the key factors. Traditional electricity price forecasting methods often fail to fully consider meteorological data and complex mechanism electricity price mechanisms, resulting in insufficient accuracy in forecasting the return rate of the wind-solar co-located power station.
[0003] In the current electricity market environment, the mechanism electricity price mechanism has gradually become an important factor affecting the revenue of new energy power generation. The mechanism electricity price of existing projects is implemented according to the current policy, not higher than the local coal-fired power benchmark price, and the implementation period is connected with the original policy; the mechanism electricity price of incremental projects is formed through provincial bidding. Projects can voluntarily participate in the bidding, and the selected projects are determined according to the bids from low to high. The final electricity price is based on the highest selected bid, but it shall not exceed the provincial-set bidding ceiling.
[0004] Therefore, how to accurately forecast the mechanism electricity price and integrate it into the forecasting model of the return rate of the wind-solar co-located power station has become an urgent problem to be solved.
[0005] At the same time, meteorological data has a significant impact on the power generation efficiency of the wind-solar co-located power station. Wind power generation depends on meteorological conditions such as wind speed and wind direction, while photovoltaic power generation is closely related to light intensity and sunshine duration. However, existing forecasting models for the return rate of power stations rarely combine meteorological big data with mechanism electricity price forecasting, making it difficult to comprehensively and accurately evaluate the investment returns of power stations.
[0006] Therefore, there is a need for a forecasting model for the return rate of a wind-solar co-located power station that can accurately forecast the return rate of the wind-solar co-located power station by using meteorological big data and mechanism electricity price mechanisms, and provide a scientific basis for the investment decision-making and operation management of the power station. Summary of the Invention
[0007] Aiming at the deficiencies of the existing technology, the present invention provides a forecasting model for the return rate of a wind-solar co-located power station based on mechanism electricity price forecasting, which solves the problems raised in the above background art.
[0008] Technical Solution: To solve the above technical problems, according to one aspect of the present invention, more specifically, a forecasting model for the return rate of a wind-solar co-located power station based on mechanism electricity price forecasting includes a meteorological data collection and processing module, a mechanism electricity price forecasting module, a power station power generation power forecasting module, a cash flow forecasting module, and a return rate calculation module. The operation steps are as follows:
[0009] S1. Through the meteorological data collection and processing module, start the meteorological monitoring equipment installed around the wind-solar co-located power station to collect meteorological data (wind speed, wind direction, light intensity, sunshine duration, temperature, humidity, etc.) in real time, and synchronously obtain historical meteorological data from the meteorological department database. After completing the data collection, immediately preprocess the data through data cleaning and interpolation methods to remove outliers and missing values, and output the preprocessed meteorological data.
[0010] S2. The mechanism electricity price prediction module receives the above preprocessed meteorological data, as well as the mechanism electricity price and mechanism electricity quantity information, typical data of power market transactions, etc. released by the competent departments of each province (municipality, autonomous region) collected by itself, and uses the established mechanism electricity price prediction model based on machine learning algorithms for calculation, and outputs the future mechanism electricity price prediction results according to different rules for existing projects and incremental projects (for existing projects, determine the mechanism electricity price according to the current policies and local coal-fired power benchmark prices, etc., combined with the model prediction results; for incremental projects, consider the provincial bidding rules and the characteristics of the project itself to predict the mechanism electricity price after the project participates in the bidding).
[0011] S3. The power generation power prediction module of the power station calls the preprocessed meteorological data to establish a power generation power prediction model for the wind-solar co-located power station. For the wind power part, based on the wind power generation power curve and wind speed and wind direction data, and for the photovoltaic power generation part, based on meteorological data such as light intensity and sunshine duration and the characteristics of photovoltaic cells, predict the wind power generation power and photovoltaic power generation power respectively, and integrate the two results to obtain the total power generation power prediction value of the wind-solar co-located power station.
[0012] S4. The cash flow prediction module predicts the sales revenue of the power station according to the predicted power generation power and mechanism electricity price of the power station, and at the same time considers the operating costs such as equipment maintenance costs, personnel salaries, loan interests, and the initial investment cost (including equipment purchase, site construction, etc.), and constructs and outputs the results of the cash flow prediction model of the power station.
[0013] S5. The rate of return calculation module is based on the net cash flow data output by the cash flow prediction module, uses the discounted cash flow model (DCF), sets a reasonable discount rate, discounts the net cash flows of future periods to the current moment, and calculates the investment measurement data of the net present value (NPV), internal rate of return (IRR), total investment internal rate of return (TotalInvestmentIRR), and return on invested capital (ROIC) of the power station, and completes the operation process of the entire model.
[0014] Furthermore, the meteorological monitoring equipment collects meteorological data every 15 minutes, and uses a data processing program written in Python language to clean the collected meteorological data and supplement missing values.
[0015] Furthermore, the mechanism electricity price prediction module uses the TensorFlow library in Python to build a neural network model for mechanism electricity price prediction, and adjusts the prediction results according to different rules for existing projects and incremental projects.
[0016] Furthermore, in the power generation power prediction module of the power station, the wind power generation power prediction is realized by writing a program using Matlab software based on the power curve of the wind turbine and the wind speed and wind direction data, and the photovoltaic power generation power prediction is realized by establishing a model using Matlab software based on the characteristic parameters of the photovoltaic cell and the light intensity and sunshine duration data.
[0017] Furthermore, in the cash flow prediction module, the equipment maintenance cost is calculated at 5% of the equipment purchase cost per year, the staff salary is determined according to the power station scale and the local labor market price, and the loan interest is calculated according to the loan amount and the loan interest rate.
[0018] Furthermore, the rate of return calculation module sets the discount rate at 8%, and uses the financial calculation library in Python to calculate the investment measurement data of net present value (NPV), internal rate of return (IRR), total investment internal rate of return (TotalInvestmentIRR), and return on invested capital (ROIC).
[0019] The beneficial effects of the wind-solar hybrid power station rate of return prediction model based on mechanism electricity price prediction of the present invention are as follows:
[0020] (1) The present invention combines meteorological big data with mechanism electricity price prediction, and fully considers the key factors affecting the rate of return of the wind-solar hybrid power station. Accurate meteorological data is obtained through the meteorological data collection and processing module. The mechanism electricity price prediction module uses machine learning algorithms to predict electricity prices. The power generation power prediction module of the power station predicts the power generation power based on meteorological data and equipment characteristics. Then, the cash flow of the power station and the investment measurement data are accurately calculated in the cash flow prediction module and the rate of return calculation module, comprehensively improving the accuracy of the rate of return prediction of the power station and providing a reliable basis for investment decisions.
[0021] (2) The present invention predicts electricity prices according to different rules for existing projects and incremental projects. For existing projects, the electricity price is determined by combining the current policies and the local coal-fired power benchmark price; for incremental projects, the electricity price is predicted considering the provincial bidding rules and the characteristics of the project itself, making the prediction results more in line with the actual situation and better serving the investment income evaluation of different types of projects.
[0022] (3) The yield prediction data of the present invention can help power station operators reasonably plan operation strategies. For example, according to the cash flow prediction results, arrange the equipment maintenance plan in advance. When the equipment maintenance cost is calculated at 5% of the purchase cost, reasonably allocate funds; optimize the human resource allocation based on the relationship between the personnel salary, the scale of the power station and the local labor market price; arrange the repayment plan reasonably according to the loan interest calculation results, so as to effectively reduce the operation cost, improve the operation efficiency and economic benefits of the power station. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The present invention will be further described in detail below with reference to the drawings and specific implementation methods.
[0024] Figure 1 It is a structural schematic diagram of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0025] The present invention will be described in detail below with reference to the drawings and embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.
[0026] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and specific embodiments.
[0027] Refer to Figure 1 , a yield prediction model for a wind-solar co-located power station based on mechanism electricity price prediction, including a meteorological data collection and processing module: start the meteorological monitoring equipment installed around the wind-solar co-located power station, and collect meteorological data such as wind speed, wind direction, light intensity, sunshine duration, temperature, humidity, etc. in real time every 15 minutes, and synchronously obtain historical meteorological data from the meteorological department database. Use a data processing program written in Python to preprocess the collected data through data cleaning and interpolation methods, remove outliers and missing values, and output the preprocessed meteorological data.
[0028] Mechanism electricity price prediction module: receive the preprocessed meteorological data output by the meteorological data collection and processing module, and at the same time collect mechanism electricity price and mechanism electricity quantity information, typical data of power market transactions, etc. released by the competent departments of each province (municipality, autonomous region). Use the TensorFlow library in Python to build a neural network model based on machine learning algorithms for mechanism electricity price prediction operations. Output the future mechanism electricity price prediction results according to different rules for existing projects and incremental projects. For existing projects, determine the mechanism electricity price based on the current policies and local coal-fired power benchmark prices, etc., in combination with the model prediction results; for incremental projects, consider the provincial bidding rules and the characteristics of the project itself to predict the mechanism electricity price after the project participates in the bidding.
[0029] Power generation prediction module of power station: Call the meteorological data collected and processed by the meteorological data collection and processing module to establish a power generation prediction model for the wind-solar co-located power station. For the wind power generation part, use Matlab software to write a program to predict the wind power generation according to the power curve of the wind turbine and the wind speed and wind direction data; for the photovoltaic power generation part, use Matlab software to establish a model to predict the photovoltaic power generation according to the characteristic parameters of the photovoltaic cell and the light intensity and sunshine duration data. Integrate the results of both to obtain the total power generation prediction value of the wind-solar co-located power station.
[0030] Cash flow prediction module: According to the power generation of the power station predicted by the power generation prediction module of the power station and the mechanism electricity price predicted by the mechanism electricity price prediction module, combined with the electricity market trading rules, predict the sales revenue of the power station. At the same time, consider the operating costs such as equipment maintenance costs (calculated at 5% of the equipment purchase cost per year), personnel salaries (determined according to the scale of the power station and the local labor market price), loan interests (calculated according to the loan amount and loan interest rate), etc., as well as the initial investment cost (including equipment purchase, site construction, etc.) to construct and output the results of the cash flow prediction model of the power station.
[0031] Yield calculation module: Based on the net cash flow data output by the cash flow prediction module, use the discounted cash flow model (DCF), set the discount rate to 8%, and use the financial calculation library in Python to discount the net cash flows of future periods to the current moment, and calculate the investment measurement data such as the net present value (NPV), internal rate of return (IRR), total investment internal rate of return (TotalInvestmentIRR), and return on invested capital (ROIC) of the power station. The yield indicators calculated using the discounted cash flow model include the net present value (NPV = ∑(cash inflow in the t-th period - cash outflow in the t-th period) / (1 + discount rate)^t, where t is each period of the project operation period, and the discount rate is determined according to factors such as the market interest rate and project risk), internal rate of return (IRR, the discount rate that makes the net present value of the project zero), total investment internal rate of return (calculated considering all investments of the project, including self-owned funds and loan funds), and return on invested capital (ROIC = earnings before interest and taxes (EBIT) / invested capital (IC), invested capital = shareholders' equity + interest-bearing liabilities).
[0032] The above-described embodiments merely represent several implementation manners of the present invention, and their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.
Claims
1. A prediction model for the return rate of a wind-solar co-located power station based on mechanism electricity price prediction, comprising a meteorological data collection and processing module, a mechanism electricity price prediction module, a power generation power prediction module of the power station, a cash flow prediction module, and a return rate calculation module, characterized in that, The running steps are as follows: S1. Through the meteorological data collection and processing module, start the meteorological monitoring equipment installed around the wind-solar co-located power station to collect meteorological data in real time, and simultaneously obtain historical meteorological data from the meteorological department database. After completing the data collection, immediately preprocess the data through data cleaning and interpolation methods to remove outliers and missing values, and output the preprocessed meteorological data. S2. The mechanism electricity price prediction module receives the above-mentioned preprocessed meteorological data, as well as the mechanism electricity price and mechanism electricity quantity information released by the provincial competent departments and typical data of the power market transactions collected by itself, and uses the established mechanism electricity price prediction model based on machine learning algorithms to perform operations, and outputs the future mechanism electricity price prediction results according to different rules for existing projects and incremental projects. S3. The power generation power prediction module of the power station calls the preprocessed meteorological data to establish a power generation power prediction model for the wind-solar co-located power station. For the wind power generation part, based on the wind power generation power curve and wind speed and wind direction data, and for the photovoltaic power generation part, based on the light intensity, sunshine duration meteorological data and the characteristics of photovoltaic cells, predict the wind power generation power and photovoltaic power generation power respectively, and integrate the two results to obtain the total power generation power prediction value of the wind-solar co-located power station. S4. The cash flow prediction module predicts the sales revenue of the power station according to the predicted power generation power and mechanism electricity price of the power station, and combines the power market trading rules. At the same time, considering the equipment maintenance cost, personnel salary, loan interest operation cost and initial investment cost, construct and output the results of the cash flow prediction model of the power station. S5. The rate of return calculation module is based on the net cash flow data output by the cash flow prediction module, adopts the cash flow discount model, sets a reasonable discount rate, discounts the net cash flow of each future period to the current moment, calculates the investment measurement data of the net present value, internal rate of return, total investment internal rate of return, and return on investment of the power station, and completes the operation process of the entire model.
2. The prediction model for the yield of a wind-solar co-located power station based on mechanism electricity price prediction according to claim 1, wherein The meteorological monitoring equipment collects meteorological data every 15 minutes, and uses a data processing program written in the Python language to clean and supplement the missing values of the collected meteorological data.
3. The prediction model for the yield of a wind-solar co-located power station based on mechanism electricity price prediction according to claim 1, wherein The mechanism electricity price prediction module uses the TensorFlow library in Python to build a neural network model for mechanism electricity price prediction, and adjusts the prediction results according to different rules for existing projects and incremental projects.
4. The predicted model of the return rate of the wind-solar co-located power station based on the mechanism electricity price prediction according to claim 1, wherein In the power generation power prediction module of the power station, the wind power generation power prediction is realized by writing a program in Matlab software according to the wind turbine power curve and wind speed and wind direction data, and the photovoltaic power generation power prediction is realized by establishing a model in Matlab software according to the characteristic parameters of photovoltaic cells and light intensity and sunshine duration data.
5. The predicted model of the yield of the wind-solar co-located power station based on the mechanism electricity price prediction according to claim 1, wherein In the cash flow prediction module, the equipment maintenance cost is calculated at 5% of the equipment purchase cost per year, the personnel salary is determined according to the scale of the power station and the local labor market price, and the loan interest is calculated according to the loan amount and loan interest rate.
6. The prediction model for the yield of a wind-solar co-located power station based on mechanism electricity price prediction according to claim 1, wherein The rate of return calculation module sets the discount rate at 8%, and uses the financial calculation library in Python to calculate the investment measurement data of the net present value, internal rate of return, total investment internal rate of return, and return on investment.