A monthly priority power generation amount prediction method based on meteorological data
Patent Information
- Application Number
- CN202211347617.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-10-31
- Publication Date
- 2026-08-18
- Estimated Expiration
- 2042-10-31
AI Technical Summary
目前,月时间尺度下对于风电场、光伏电站月度发电量预测方法研究较少,重点在于历史数据挖掘,缺乏气象因素对月时间尺度下发电量影响方面的研究
[0036] This invention establishes a power generation prediction model based on historical power and meteorological data. The proposed model provides a basis for predicting monthly priority power generation from clean energy sources, helping to ensure a balance between power supply and demand and guaranteeing the consumption of clean energy. Compared with traditional monthly power generation prediction models for wind farms and photovoltaic power plants, this invention considers the inherent influence mechanism of meteorological factors on the priority power generation of wind farms and photovoltaic power plants on a monthly timescale. The proposed method is more scientifically based and improves prediction accuracy.
Smart Images

Figure CN115714378B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for predicting power generation, specifically a method for predicting monthly priority power generation based on meteorological data. Background Technology
[0002] The accuracy and rationality of monthly priority power generation forecasts for clean energy directly affect the proportion of power generation from each type in the power grid, and will further impact the performance evaluation index of the monthly planned power purchase balance rate of power plants under grid dispatch. Improving the accuracy of monthly power generation forecasts for clean energy sources such as wind and solar power is crucial. Currently, there is limited research on monthly power generation forecasting methods for wind farms and photovoltaic power plants on a monthly timescale, with a focus on historical data mining and a lack of research on the impact of meteorological factors on power generation on a monthly timescale. Summary of the Invention
[0003] To address the problems existing in the background technology, the present invention provides a monthly priority power generation prediction method based on meteorological data. It considers the inherent influence mechanism of meteorological factors on the priority power generation of wind farms and photovoltaic power plants on a monthly time scale, and proposes a practical monthly priority power generation prediction method that can improve the accuracy and reliability of monthly priority power generation prediction.
[0004] The technical solution adopted in this invention is:
[0005] The monthly priority power generation prediction method of the present invention includes the following steps:
[0006] Step 1: Establish a power generation prediction model for the power plant.
[0007] Step 2: Obtain meteorological data for the region where the power plant is located in the month to be predicted and perform quantitative processing to obtain quantitative meteorological data.
[0008] Step 3: Input the quantitative meteorological data of the area where the power plant is located in the month to be predicted into the power generation prediction model. The power generation prediction model outputs the number of power generation utilization hours of the power plant in the predicted month.
[0009] Step 4: Based on the predicted monthly power generation utilization hours and the installed capacity of the power station in the month to be predicted, obtain the predicted power generation of the power station in the month to be predicted, thus realizing the prediction of the monthly power generation of the power station.
[0010] In step 1, the power station includes wind farms and photovoltaic power stations, and the power generation prediction model established for the power station includes a wind farm power generation prediction model and a photovoltaic power station power generation prediction model.
[0011] The wind farm power generation prediction model is as follows:
[0012] h θ (γ)=θ0+θ1γ
[0013] Wherein, γ represents the monthly wind force level of the area where the wind farm is located in the month to be predicted; h θ () represents the number of power generation utilization hours of the wind farm in the predicted month of the month to be predicted; θ0 and θ1 represent the first and second fitting parameters of the wind farm power generation prediction model, respectively.
[0014] The first fitting parameter θ0 and the second fitting parameter θ1 of the wind farm power generation prediction model are as follows:
[0015]
[0016]
[0017] Where N represents the total number of historical months preceding the month to be predicted; h (i) This represents the actual monthly power generation utilization hours of the wind farm in the i-th month prior to the month to be predicted; This represents the actual monthly wind force level value of the area where the wind farm is located in the i-th month prior to the month to be predicted.
[0018] The photovoltaic power plant power generation prediction model is as follows:
[0019] h x (x s ,x c ,x o ,x r )=p0+p1x s +p2x c +p3x o +p4x r
[0020]
[0021] Among them, h x () indicates the number of hours of power generation utilization of the photovoltaic power plant in the forecast month of the month to be predicted; x s x c , xo and x r , representing the number of days with four typical weather conditions (sunny, cloudy, overcast, and rainy) in the region where the photovoltaic power station is located during the quarter of the month to be predicted; p0, p1, p2, p3, and p4 represent the first, second, third, fourth, and fifth fitting parameters of the photovoltaic power station power generation prediction model, respectively; X represents the typical weather day matrix; H x This represents a matrix showing the actual monthly power generation utilization hours.
[0022] The typical weather day matrix X and the actual monthly power generation utilization hours matrix H are mentioned above. x Specifically as follows:
[0023]
[0024]
[0025] Where, x s,1 x s,2 x s,3 ... x s,Nx These represent the number of days with typical sunny weather in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted for the photovoltaic power plant; x c,1 x c,2 x c,3 ... x c,Nx These represent the number of days with typical cloudy weather for the photovoltaic power plant in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted; x o,1 x o,2 x o,3 ... x o,Nx These represent the number of days with typical cloudy weather in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted for the photovoltaic power plant; x r,1 x r,2 x r,3 ... x r,Nx These represent the number of days with typical moderate rain in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted for the photovoltaic power plant; h x,1 h x,2 h x,3 ... h x,Nx These represent the actual monthly power generation utilization hours of the photovoltaic power plant in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted.
[0026] In step 2, the meteorological data for the region where the power station is located in the month to be predicted includes the meteorological data for the respective regions of the wind farm and the photovoltaic power station in the month to be predicted. The meteorological data for the region where the wind farm is located in the month to be predicted is the actual monthly wind force level value for the region where the wind farm is located in the month to be predicted. The meteorological data for the region where the photovoltaic power station is located in the month to be predicted is the number of days with four typical weather conditions (sunny, cloudy, overcast, and rainy) in the region where the photovoltaic power station is located in the month to be predicted.
[0027] Meteorological data for the region where the power plant is located in the forecast month is acquired and quantified to obtain quantitative meteorological data. For the wind farm region in the forecast month, the actual daily wind force level value for each day of the forecast month is obtained, and the actual monthly wind force level value for the wind farm region in the forecast month is summed, which serves as the quantitative meteorological data for the wind farm region in the forecast month. For the photovoltaic power plant region in the forecast month, for each typical weather condition, the number of days with typical weather in the quarter in which the forecast month is located is summed, which serves as the quantitative meteorological data for the photovoltaic power plant region in the forecast month. Notably, if the weather condition on a certain day is "sunny turning cloudy," it is recorded as 0.5 days of sunny and 0.5 days of cloudy weather; the same method is used for other weather conditions.
[0028] In step 4, the predicted power generation of the power station in the month to be predicted is obtained based on the predicted monthly power generation utilization hours and the installed capacity of the power station in the month to be predicted. Specifically, the predicted power generation of the wind farm in the month to be predicted is obtained using the predicted monthly power generation utilization hours and the installed capacity of the wind farm in the month to be predicted, and the predicted power generation of the photovoltaic power station in the month to be predicted is obtained using the predicted monthly power generation utilization hours and the installed capacity of the photovoltaic power station in the month to be predicted.
[0029] The specific details of obtaining the predicted power generation of the wind farm in the month to be predicted using the wind farm's monthly power generation utilization hours and the wind farm's installed capacity are as follows:
[0030] E W =h θ (γ)×C W
[0031] Among them, E W γ represents the predicted power generation of the wind farm in the month to be predicted; h represents the monthly wind force rating of the area where the wind farm is located in the month to be predicted; θ () indicates the number of hours of power generation utilization of the wind farm in the forecast month of the month to be predicted; C W This indicates the installed capacity of the wind farm in the month to be predicted.
[0032] The method for obtaining the predicted power generation of a photovoltaic power station in the month to be predicted by using the number of power generation utilization hours and the installed capacity of the photovoltaic power station in the month to be predicted is as follows:
[0033] E P =h x (x s ,x c ,x o ,x r )×CP
[0034] Among them, E P This indicates the projected power generation of the photovoltaic power plant in the month to be predicted; h n () indicates the number of hours of power generation utilization of the photovoltaic power plant in the forecast month of the month to be predicted; x s x c x o and x r These represent the number of days with four typical weather conditions—sunny, cloudy, overcast, and rainy—occurring in the region where the photovoltaic power station is located during the quarter of the month to be predicted; C P This indicates the installed capacity of photovoltaic power plants in the month to be predicted.
[0035] The beneficial effects of this invention are:
[0036] This invention establishes a power generation prediction model based on historical power and meteorological data. The proposed model provides a basis for predicting monthly priority power generation from clean energy sources, helping to ensure a balance between power supply and demand and guaranteeing the consumption of clean energy. Compared with traditional monthly power generation prediction models for wind farms and photovoltaic power plants, this invention considers the inherent influence mechanism of meteorological factors on the priority power generation of wind farms and photovoltaic power plants on a monthly timescale. The proposed method is more scientifically based and improves prediction accuracy. Attached Figure Description
[0037] Figure 1 This is a flowchart of the method of the present invention;
[0038] Figure 2 This is a schematic diagram illustrating the prediction of monthly power generation utilization hours for wind farms. Detailed Implementation
[0039] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0040] like Figure 1 As shown, the monthly priority power generation prediction method of the present invention includes the following steps:
[0041] Step 1: Establish a power generation prediction model for the power plant.
[0042] In step 1, the power station includes wind farms and photovoltaic power stations, and the power generation prediction model established for the power station includes a wind farm power generation prediction model and a photovoltaic power station power generation prediction model.
[0043] The wind farm power generation prediction model is as follows:
[0044] h θ (γ)=θ0+θ1γ
[0045] Wherein, γ represents the monthly wind force level of the area where the wind farm is located in the month to be predicted; h θ() represents the number of power generation utilization hours of the wind farm in the predicted month of the month to be predicted; θ0 and θ1 represent the first and second fitting parameters of the wind farm power generation prediction model, respectively.
[0046] The first fitting parameter θ0 and the second fitting parameter θ1 of the wind farm power generation prediction model are as follows:
[0047]
[0048]
[0049] Where N represents the total number of historical months preceding the month to be predicted; h (i) This represents the actual monthly power generation utilization hours of the wind farm in the i-th month prior to the month to be predicted; This represents the actual monthly wind force level value of the area where the wind farm is located in the i-th month prior to the month to be predicted.
[0050] The photovoltaic power plant power generation prediction model is as follows:
[0051] h x (x s ,x c ,x o ,x r )=p0+p1x s +p2x c +p3x o +p4x r
[0052]
[0053] Among them, h x () indicates the number of hours of power generation utilization of the photovoltaic power plant in the forecast month of the month to be predicted; x s x c , xo and x r , representing the number of days with four typical weather conditions (sunny, cloudy, overcast, and rainy) in the region where the photovoltaic power station is located during the quarter of the month to be predicted; p0, p1, p2, p3, and p4 represent the first, second, third, fourth, and fifth fitting parameters of the photovoltaic power station power generation prediction model, respectively; X represents the typical weather day matrix; H x This represents a matrix showing the actual monthly power generation utilization hours.
[0054] Typical weather days matrix X and actual monthly power generation utilization hours matrix H x Specifically as follows:
[0055]
[0056]
[0057] Where, x s,1 x s,2 x s,3 ... x s,Nx These represent the number of days with typical sunny weather in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted for the photovoltaic power plant; x c,1 x c,2 x c,3 ... x c,Nx These represent the number of days with typical cloudy weather for the photovoltaic power plant in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted; x o,1 x o,2 x o,3 ... x o,Nx These represent the number of days with typical cloudy weather in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted for the photovoltaic power plant; x r,1 x r,2 x r,3 ... x r,Nx These represent the number of days with typical moderate rain in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted for the photovoltaic power plant; h x,1 h x,2 h x,3 ... h x,Nx These represent the actual monthly power generation utilization hours of the photovoltaic power plant in the 1st, 2nd, 3rd, ..., Nx months prior to the month to be predicted.
[0058] Step 2: Obtain meteorological data for the region where the power plant is located in the month to be predicted and perform quantitative processing to obtain quantitative meteorological data.
[0059] In step 2, the meteorological data for the region where the power station is located in the month to be predicted includes the meteorological data for the respective regions where the wind farm and the photovoltaic power station are located in the month to be predicted. The meteorological data for the region where the wind farm is located in the month to be predicted is the actual monthly wind force level value for the region where the wind farm is located in the month to be predicted. The meteorological data for the region where the photovoltaic power station is located in the month to be predicted is the number of days with four typical weather conditions (sunny, cloudy, overcast, and rainy) in the region where the photovoltaic power station is located in the month to be predicted.
[0060] Meteorological data for the region where the power plant is located in the forecast month is acquired and quantified to obtain quantitative meteorological data. For the wind farm region in the forecast month, the actual daily wind force level value for each day of the forecast month is obtained, and the actual monthly wind force level value for the wind farm region in the forecast month is summed, which serves as the quantitative meteorological data for the wind farm region in the forecast month. For the photovoltaic power plant region in the forecast month, for each typical weather condition, the number of days with typical weather in the quarter in which the forecast month is located is summed, which serves as the quantitative meteorological data for the photovoltaic power plant region in the forecast month. Notably, if the weather condition on a certain day is "sunny turning cloudy," it is recorded as 0.5 days of sunny and 0.5 days of cloudy weather; the same method is used for other weather conditions.
[0061] Step 3: Input the quantitative meteorological data of the area where the power plant is located in the month to be predicted into the power generation prediction model. The power generation prediction model outputs the number of power generation utilization hours of the power plant in the predicted month.
[0062] Step 4: Based on the predicted monthly power generation utilization hours and the installed capacity of the power station in the month to be predicted, obtain the predicted power generation of the power station in the month to be predicted, thus realizing the prediction of the monthly power generation of the power station.
[0063] In step 4, the predicted power generation of the power station in the month to be predicted is obtained based on the predicted monthly power generation utilization hours and the installed capacity of the power station in the month to be predicted. Specifically, the predicted power generation of the wind farm in the month to be predicted is obtained using the predicted monthly power generation utilization hours and the installed capacity of the wind farm in the month to be predicted, and the predicted power generation of the photovoltaic power station in the month to be predicted is obtained using the predicted monthly power generation utilization hours and the installed capacity of the photovoltaic power station in the month to be predicted.
[0064] The predicted power generation of the wind farm in the month to be predicted is obtained using the wind farm's monthly power generation utilization hours and installed capacity as follows:
[0065] E W =h θ (γ)×C W
[0066] Among them, E W γ represents the predicted power generation of the wind farm in the month to be predicted; h represents the monthly wind force rating of the area where the wind farm is located in the month to be predicted; θ () indicates the number of hours of power generation utilization of the wind farm in the forecast month of the month to be predicted; C W This indicates the installed capacity of the wind farm in the month to be predicted.
[0067] The predicted power generation of the photovoltaic power plant in the month to be predicted is obtained by using the monthly power generation utilization hours and the installed capacity of the photovoltaic power plant in the month to be predicted, as follows:
[0068] E P =h x (x s ,x c ,x o ,x r )×C P
[0069] Among them, E P This indicates the projected power generation of the photovoltaic power plant in the month to be predicted; h n () indicates the number of hours of power generation utilization of the photovoltaic power plant in the forecast month of the month to be predicted; x s x c x o and x r These represent the number of days with four typical weather conditions—sunny, cloudy, overcast, and rainy—occurring in the region where the photovoltaic power station is located during the quarter of the month to be predicted; C P This indicates the installed capacity of photovoltaic power plants in the month to be predicted.
[0070] Specific embodiments of the present invention are as follows:
[0071] The forecast month is June 2022. The wind force level values for each day of the 24 months preceding June are summed to obtain the historical monthly wind force level values. June is the second quarter. The number of typical weather events in each month of the second quarter for the previous 5 years is summed to obtain the number of days with typical weather events. The installed capacity and corresponding monthly power generation of the wind farm for the 24 months preceding June are taken, and the installed capacity and corresponding monthly power generation of the photovoltaic power station for the third quarter for each of the previous 5 years are taken.
[0072] Based on the obtained historical meteorological data and historical power output data (i.e., installed capacity and monthly power generation data), the power generation prediction model for the wind farm is as follows:
[0073] h θ (γ) = -157.22 + 5.03γ
[0074] Based on the obtained historical meteorological data and historical power output data, the power generation prediction model for photovoltaic power plants is as follows:
[0075] h n (n s ,n c ,n o ,n r )=793.60-19.82n s -20.82n c -22.18no -23.37n r
[0076] The monthly wind force level in the area where the wind turbine is located is 70. The number of typical weather events (sunny, cloudy, overcast, and rainy) is 1 day, 6 days, 7.5 days, and 15.5 days, respectively. The final predicted monthly power generation of the wind farm in June is 11,694,500 kWh, with an error of 2.05% compared to the actual value. The predicted monthly power generation of the photovoltaic power station in June is 12,033,400 kWh, with an error of 8.30% compared to the actual value. It can be seen that the prediction accuracy of the method of the present invention is high.
Claims
1. A method for predicting monthly priority power generation based on meteorological data, characterized in that: The method includes the following steps: Step 1: Establish a power generation prediction model for the power plant; Step 2: Obtain meteorological data for the region where the power plant is located in the month to be predicted and perform quantitative processing to obtain quantitative meteorological data; Step 3: Input the quantitative meteorological data of the area where the power plant is located in the month to be predicted into the power generation prediction model. The power generation prediction model outputs the number of power generation utilization hours of the power plant in the predicted month. Step 4: Based on the predicted monthly power generation utilization hours of the power station in the month to be predicted and the installed capacity of the power station, obtain the predicted power generation of the power station in the month to be predicted, thus realizing the prediction of the monthly power generation of the power station. In step 1, the power station includes wind farms and photovoltaic power stations, and the power generation prediction model of the power station includes wind farm power generation prediction model and photovoltaic power station power generation prediction model. The wind farm power generation prediction model is as follows: in, This indicates the monthly wind force level for the area where the wind farm is located in the month to be predicted; This indicates the number of hours of power generation utilization for the wind farm in the forecast month of the month to be predicted; and represent the first and second fitting parameters of the wind farm power generation prediction model, respectively; The first fitting parameters of the wind farm power generation prediction model Second fitting parameters Specifically as follows: in, This represents the total number of historical months preceding the month to be predicted. This indicates the wind farm's performance in the month prior to the forecast. Actual monthly power generation utilization hours for the month; This indicates the region where the wind farm is located is in the month preceding the forecast. The actual monthly wind force rating for the month; The photovoltaic power plant power generation prediction model is as follows: in, This indicates the number of hours of power generation utilization of the photovoltaic power plant in the forecast month of the month to be predicted; , , and These represent the number of days with four typical weather conditions—sunny, cloudy, overcast, and rainy—occurring in the region where the photovoltaic power station is located during the quarter of the month to be predicted. , , , and These represent the first, second, third, fourth, and fifth fitting parameters of the photovoltaic power plant power generation prediction model, respectively. Represents a matrix of typical weather days; This represents a matrix showing the actual monthly power generation utilization hours.
2. The method for predicting monthly priority power generation based on meteorological data according to claim 1, characterized in that: The typical weather day matrix Matrix of actual monthly power generation utilization hours Specifically as follows: in, , , … These represent the 1st, 2nd, 3rd, ... months preceding the month to be predicted for the photovoltaic power plant. The number of days with typical sunny weather in a month; , , … These represent the 1st, 2nd, 3rd, ... months preceding the month to be predicted for the photovoltaic power plant. The number of days in a month that are typically cloudy; , , … These represent the 1st, 2nd, 3rd, ... months preceding the month to be predicted for the photovoltaic power plant. The number of days in a month when the weather is typically overcast; , , … These represent the 1st, 2nd, 3rd, ... months preceding the month to be predicted for the photovoltaic power plant. The number of days in a month when rain is typical; , , … These represent the 1st, 2nd, 3rd, ... months preceding the month to be predicted for the photovoltaic power plant. The actual monthly power generation utilization hours in a month.
3. The method for predicting monthly priority power generation based on meteorological data according to claim 1, characterized in that: In step 2, the meteorological data for the region where the power station is located in the month to be predicted includes the meteorological data for the respective regions where the wind farm and the photovoltaic power station are located in the month to be predicted. The meteorological data for the region where the wind farm is located in the month to be predicted is the actual monthly wind force level value for the region where the wind farm is located in the month to be predicted. The meteorological data for the region where the photovoltaic power station is located in the month to be predicted is the number of days with four typical weather conditions (sunny, cloudy, overcast, and rainy) in the region where the photovoltaic power station is located in the month to be predicted. Meteorological data for the region where the power plant is located in the month to be predicted is obtained and quantified to obtain quantitative meteorological data. For the meteorological data of the region where the wind farm is located in the month to be predicted, the actual daily wind force level value of each day in the region where the wind farm is located in the month to be predicted is obtained. The actual monthly wind force level value of the region where the wind farm is located in the month to be predicted is obtained by summing the actual daily wind force level values. This is the quantitative meteorological data of the region where the wind farm is located in the month to be predicted. For the meteorological data of the region where the photovoltaic power plant is located in the month to be predicted, for each typical weather event, the number of days with typical weather events in the quarter in which the month to be predicted occurs is summed to obtain the number of days with typical weather events in the region where the photovoltaic power plant is located in the month to be predicted. This is the quantitative meteorological data of the region where the photovoltaic power plant is located in the month to be predicted.
4. The method for predicting monthly priority power generation based on meteorological data according to claim 1, characterized in that: In step 4, the predicted power generation of the power station in the month to be predicted is obtained based on the predicted monthly power generation utilization hours and the installed capacity of the power station in the month to be predicted. Specifically, the predicted power generation of the wind farm in the month to be predicted is obtained using the predicted monthly power generation utilization hours and the installed capacity of the wind farm in the month to be predicted, and the predicted power generation of the photovoltaic power station in the month to be predicted is obtained using the predicted monthly power generation utilization hours and the installed capacity of the photovoltaic power station in the month to be predicted.
5. The method for predicting monthly priority power generation based on meteorological data according to claim 4, characterized in that: The specific details of obtaining the predicted power generation of the wind farm in the month to be predicted using the wind farm's monthly power generation utilization hours and the wind farm's installed capacity are as follows: in, This indicates the projected power generation of the wind farm in the month to be predicted; This indicates the monthly wind force level for the area where the wind farm is located in the month to be predicted; This indicates the number of hours of power generation utilization for the wind farm in the forecast month of the month to be predicted; This indicates the installed capacity of the wind farm in the month to be predicted.
6. The method for predicting monthly priority power generation based on meteorological data according to claim 4, characterized in that: The method for obtaining the predicted power generation of a photovoltaic power station in the month to be predicted by using the number of power generation utilization hours and the installed capacity of the photovoltaic power station in the month to be predicted is as follows: in, This indicates the projected power generation of the photovoltaic power plant in the month to be predicted; This indicates the number of hours of power generation utilization of the photovoltaic power plant in the forecast month of the month to be predicted; , , and These represent the number of days with four typical weather conditions—sunny, cloudy, overcast, and rainy—occurring in the region where the photovoltaic power station is located during the quarter of the month to be predicted. This indicates the installed capacity of photovoltaic power plants in the month to be predicted.
