Meteorological data preprocessing method for new energy prediction system

Abstract

The invention discloses a meteorological data preprocessing method for a new energy prediction system, and relates to the technical field of new energy power generation. According to the new energy prediction system meteorological data preprocessing method, grid processing is carried out on wide-area meteorological resource data, geographic interpolation calculation is carried out on meteorological data of longitude and latitude positions of photovoltaic systems, an elevation information and air temperature information collaborative Kriging interpolation method is introduced, longitude and latitude are used as main variables, and a new energy prediction system is obtained. Elevation information and air temperature information are added as auxiliary variables, and correlation is described by using a cross covariance and a cross semi-variation function, so that the estimation precision of a main variable is improved, and meteorological data preprocessing is more accurate and is better suitable for a new energy power generation prediction condition.

Description

Technical Field

[0001] The present invention relates to the technical field of renewable energy power generation, and in particular to a meteorological data preprocessing method for a renewable energy forecasting system. Background Art

[0002] Since distributed photovoltaic systems generally have a small installed capacity and no special new energy resource monitoring equipment is installed, directly using the data from the closest point to the photovoltaic system in the wide-area meteorological data for power prediction may result in low accuracy. In addition, wide-area meteorological data often have coarse particles and low longitude and latitude resolution, while the spatial location of each distributed photovoltaic system has a high longitude and latitude resolution.

[0003] Based on this, a meteorological data preprocessing method for a new energy prediction system is proposed. By gridding the wide-area meteorological resource data and performing geographic interpolation calculations on the meteorological data at the longitude and latitude of each photovoltaic system, the problem that the ordinary Kriging method and the inverse distance weighted method can be used to interpolate and obtain meteorological resource data near the location of the new energy power station, but the accuracy of the interpolation results is difficult to guarantee, can be avoided, so as to achieve more accurate meteorological data preprocessing. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the deficiencies in the prior art, the present invention provides a meteorological data preprocessing method for a new energy prediction system, which solves the problem that it is difficult to achieve accurate prediction of each photovoltaic system when the photovoltaic systems are unevenly distributed.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a meteorological data preprocessing method for a new energy prediction system, specifically comprising the following steps:

[0008] Step 1: Obtain geographic information of the location of distributed renewable energy power station A;

[0009] Step 2: Obtain the historical light intensity record time sequence of distributed renewable energy power station A;

[0010] Step 3: Obtain geographic information and elevation information of N meteorological stations Bi within a preset d kilometers around the distributed new energy source;

[0011] Step 4: Obtain the historical light intensity and predicted temperature information of N surrounding weather stations;

[0012] Step 5: Import the historical data of N surrounding meteorological stations and the light intensity recorded by distributed renewable energy power station A into the light intensity prediction calculation formula, and calculate the light intensity at distributed renewable energy power station A by combining the Kriging difference method and the inverse distance weighted interpolation method;

[0013] Step 6: Optimize the light intensity prediction calculation formula to a minimum value calculation formula. According to the historical light intensity and predicted temperature information of N surrounding meteorological stations, obtain the relationship between the light intensity prediction value of distributed renewable energy power station A and the meteorological data of N surrounding meteorological stations within a preset d kilometers, and generate a prediction formula.

[0014] Step 7: Obtain the predicted light intensity and temperature information of N surrounding weather stations, and substitute them into the prediction formula in step 7 for prediction.

[0015] The present invention is further configured such that the geographic information in step one and step three both include longitude and latitude information.

[0016] The present invention is further configured as follows: in the N weather stations Bi in step three, i is 1-N, and the value of N is at least 3.

[0017] The present invention is further configured as follows: the light intensity prediction calculation formula in step 5 includes:

[0018]

[0019] in, is the predicted light intensity at distributed renewable energy power station A, is the measured light intensity of the surrounding N weather stations, H is the actual elevation information of the weather station, T i is the actual temperature at the weather station observation point, and

[0020] The present invention is further configured as follows: the Kriging difference method calculation formula in step 5 includes:

[0021]

[0022] The kriging interpolation result at S0 is Z(S0), and the number of sample points used in the calculation process is n, z i is the value of the corresponding sample point, λ i is the weight coefficient corresponding to the sample point, and it must also meet the conditions of unbiased estimation;

[0023]

[0024] Where z0 is the true value at S0.

[0025] The present invention is further configured as follows: the inverse distance weighted interpolation method calculation formula includes:

[0026]

[0027] The inverse distance weighted interpolation result at S0 is Z(S0), d i is the distance between the interpolation point and the known point, Z i is the value of the corresponding sample point, p is the power parameter, which is 2.

[0028] The present invention is further configured as follows: the minimum value calculation formula in step 6 includes:

[0029]

[0030] For the 3N sets of historical data, we need to minimize Δ and then solve λ by the Lagrange multiplier method. i , μ i , i The specific value of is used to generate the prediction formula.

[0031] There is a strong correlation between meteorological data, geographic elevation information, and temperature. Meteorological data will change significantly with changes in elevation and temperature. Using geographic elevation information and temperature information as auxiliary variables to interpolate meteorological data will effectively improve the interpolation accuracy of meteorological data.

[0032] (III) Beneficial effects

[0033] The present invention provides a meteorological data preprocessing method for a new energy prediction system. It has the following beneficial effects:

[0034] The present invention performs grid processing on wide-area meteorological resource data, performs geographic interpolation calculation on the meteorological data of the longitude and latitude positions of each photovoltaic system, introduces the collaborative Kriging interpolation method of elevation information and temperature information, takes longitude and latitude as the main variables, adds elevation information and temperature information as auxiliary variables, and uses cross-covariance and cross-semivariogram functions to characterize correlation, so as to improve the estimation accuracy of the main variables, make the meteorological data preprocessing more accurate, and better apply to the prediction of new energy power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the process of the present invention;

[0036] Figure 2 This is a schematic diagram of the result of using the ordinary Kriging method to obtain meteorological resource data near the location of a new energy power station in an embodiment of the present invention;

[0037] Figure 3This is a schematic diagram of the results of using the inverse distance weighted interpolation method to obtain meteorological resource data near the location of a new energy power station in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0039] See also Figure 1-3 The embodiment of the present invention provides the following technical solution: a meteorological data preprocessing method for a new energy prediction system, specifically comprising the following steps:

[0040] Step 1: Obtain geographic information of the location of distributed renewable energy power station A;

[0041] Step 2: Obtain the historical light intensity record time sequence of distributed renewable energy power station A;

[0042] Step 3: Obtain the geographic information and elevation information of N meteorological stations Bi within a preset d kilometers around the distributed renewable energy source, where i is 1-N, and the value of N is at least 3;

[0043] Step 4: Obtain the historical light intensity and predicted temperature information of N surrounding weather stations;

[0044] Step 5: Import the historical data of N surrounding meteorological stations and the light intensity recorded by distributed renewable energy power station A into the light intensity prediction calculation formula, and calculate the light intensity at distributed renewable energy power station A by combining the Kriging difference method and the inverse distance weighted interpolation method;

[0045] The Kriging difference method calculation formula includes:

[0046]

[0047] The kriging interpolation result at S0 is Z(S0), and the number of sample points used in the calculation process is n, z i is the value of the corresponding sample point, λ i is the weight coefficient corresponding to the sample point, and it must also meet the conditions of unbiased estimation;

[0048]

[0049] Where z0 is the true value at S0;

[0050] The calculation formula of inverse distance weighted interpolation method includes:

[0051]

[0052] The inverse distance weighted interpolation result at S0 is Z(S0), d iis the distance between the interpolation point and the known point, Z i is the value of the corresponding sample point, p is the power parameter, which is 2;

[0053] The light intensity prediction calculation formula includes:

[0054]

[0055] in, is the predicted light intensity at distributed renewable energy power station A, is the measured light intensity of the surrounding N weather stations, H is the actual elevation information of the weather station, T i is the actual temperature at the weather station observation point, and

[0056] Step 6: Optimize the light intensity prediction calculation formula to a minimum value calculation formula. The minimum value calculation formula includes:

[0057]

[0058] For the 3N sets of historical data, we need to minimize Δ and then solve λ by the Lagrange multiplier method. i , μ i , i The specific value of generates the prediction formula;

[0059] According to the historical light intensity and predicted temperature information of N surrounding meteorological stations, the relationship between the light intensity prediction value of distributed renewable energy power station A and the meteorological data of N surrounding meteorological stations within a preset d kilometers is obtained to generate a prediction formula;

[0060] Step 7: Obtain the predicted light intensity and temperature information of N surrounding weather stations, and substitute them into the prediction formula in step 7 for prediction.

[0061] In summary, the meteorological resource data near the system location can be obtained by using the ordinary kriging method and the inverse distance weighted method interpolation. The results of using the ordinary kriging method to obtain the meteorological resource data near the location of the new energy power station are shown in the attached figure. Figure 2 As shown in the figure, the inverse distance weighted interpolation method is used to obtain the meteorological resource data near the location of the new energy power station. Figure 3 As shown;

[0062] However, the accuracy of the interpolation results in the above two methods is difficult to guarantee;

[0063] Therefore, the present invention introduces the co-kriging interpolation method, takes the geographic information longitude and latitude as the main variables, adds elevation information and temperature information as auxiliary variables, and uses cross-covariance and cross-semivariogram functions to characterize the correlation to improve the estimation accuracy of the main variables. This method makes the preprocessing of meteorological data more accurate and is suitable for more renewable energy power generation prediction problems.

Claims

1. A meteorological data preprocessing method for a new energy forecasting system, characterized by: The specific steps include: Step 1: Obtain geographic information of the location of distributed renewable energy power station A; Step 2: Obtain the historical light intensity record time sequence of distributed renewable energy power station A; Step 3: Obtain geographic information and elevation information of N meteorological stations Bi within a preset d kilometers around the distributed new energy source; Step 4: Obtain the historical light intensity and predicted temperature information of N surrounding weather stations; Step 5: Import the historical data of N surrounding meteorological stations and the light intensity recorded by distributed renewable energy power station A into the light intensity prediction calculation formula, and calculate the light intensity at distributed renewable energy power station A by combining the Kriging difference method and the inverse distance weighted interpolation method; Step 6: Optimize the light intensity prediction calculation formula to a minimum value calculation formula. According to the historical light intensity and predicted temperature information of N surrounding meteorological stations, obtain the relationship between the light intensity prediction value of distributed renewable energy power station A and the meteorological data of N surrounding meteorological stations within a preset d kilometers, and generate a prediction formula. Step 7: Obtain the predicted light intensity and temperature information of N surrounding weather stations, and substitute them into the prediction formula in step 7 for prediction.

2. The meteorological data preprocessing method of a new energy prediction system according to claim 1 is characterized by: The geographic information in step 1 and step 3 both include latitude and longitude information.

3. The meteorological data preprocessing method of a new energy prediction system according to claim 1 is characterized by: In the step 3, among the N weather stations Bi, i is 1-N, and the value of N is at least 3.

4. The meteorological data preprocessing method of a new energy prediction system according to claim 1, characterized in that: The light intensity prediction calculation formula in step 5 include: in, is the predicted light intensity at distributed renewable energy power station A, is the measured light intensity of the surrounding N weather stations, H is the actual elevation information of the weather station, T i is the actual temperature at the weather station observation point, and 5. The meteorological data preprocessing method of a new energy prediction system according to claim 1 is characterized by: The Kriging difference calculation formula in step 5 includes: The kriging interpolation result at S0 is Z(S0), and the number of sample points used in the calculation process is n, z i is the value of the corresponding sample point, λ i is the weight coefficient corresponding to the sample point, and it must also meet the conditions of unbiased estimation; Where z0 is the true value at S0.

6. The meteorological data preprocessing method of a new energy prediction system according to claim 1 is characterized by: The inverse distance weighted interpolation calculation formula includes: The inverse distance weighted interpolation result at S0 is Z(S0), d i is the distance between the interpolation point and the known point, Z i is the value of the corresponding sample point, p is the power parameter, which is 2.

7. The meteorological data preprocessing method of a new energy prediction system according to claim 1 is characterized by: The minimum value calculation formula in step 6 includes: For the 3N sets of historical data, we need to minimize Δ and then solve λ using the Lagrange multiplier method. i , μ i , i The specific value of is used to generate the prediction formula.

Images