A method and device for predicting regional photovoltaic power output

Abstract

The present invention relates to a method and device for predicting regional photovoltaic power output, including: obtaining various climate factors for a future time period; determining the occurrence probabilities of various weather categories in the future time period based on the various climate factors for the future time period and a pre-established multiple linear regression equation; determining the regional photovoltaic total power output prediction sequences of various weather categories in the future time period based on the occurrence probabilities of various weather categories in the future time period; wherein, the multiple linear regression equation is established based on the occurrence probabilities of various weather categories in a historical time period and the various climate factors for the historical time period; the present invention can achieve cross-season prediction of regional photovoltaic power output, making the prediction result not affected by seasonal factors and improving the accuracy of the prediction result.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy power generation prediction, and particularly relates to a method and device for predicting regional photovoltaic output. Background Art

[0002] Electric power is playing an increasingly important role in the economic society and people's lives. The access of large-scale photovoltaic power generation is one of the important ways to innovate the current power grid form. The analysis of its output characteristics can create conditions for optimizing power dispatching and achieve a more stable and reliable energy interaction method.

[0003] The existing technology predicts the regional photovoltaic output characteristics by applying a simple output time series model for stochastic analysis and prediction, or establishing a prediction equation by applying a linear model such as output time series extrapolation for prediction.

[0004] However, large-scale photovoltaic power generation is not only affected by macroscopic factors such as seasonal factors and geographical conditions, but is also very sensitive to short-term weather phenomena such as clouds and fog during the day and within the day. Therefore, in terms of output characteristics, it shows regularity, random interaction and coupling on a short time scale, and at the same time shows significant non-linear change characteristics on a longer time scale. Therefore, the methods of the existing technology are difficult to solve the influence of climate factors on the output prediction results, reducing the accuracy of the prediction results. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a method and device for predicting regional photovoltaic output, and obtain a quantitative estimate of the regional photovoltaic output characteristics across seasons through a correlation analysis method of climate factors.

[0006] The purpose of the present invention is achieved by adopting the following technical solutions:

[0007] The present invention provides a method for predicting regional photovoltaic output, and the improvement lies in that the method includes:

[0008] Obtain each climate factor in the future time period;

[0009] Based on each climate factor in the future time period and a pre-established multiple linear regression equation, determine the occurrence probability of each weather category in the future time period;

[0010] Based on the occurrence probability of each weather category in the future time period, determine the regional photovoltaic total output prediction sequence of each weather category in the future time period;

[0011] Wherein, the multiple linear regression equation is established based on the occurrence probability of each weather category in the historical time period and each climate factor in the historical time period.

[0012] Preferably, the climate factors include: Arctic Oscillation, North Atlantic Oscillation, and Pacific-North American teleconnection pattern.

[0013] Preferably, the weather categories include: clear sky category, rainy category, and cloudy category.

[0014] Preferably, the multiple linear regression equation is established based on the occurrence probabilities of each weather category in the historical period and each climate factor in the historical period, and includes:

[0015] Obtain each climate factor in the historical period, and determine the occurrence probability of weather category i in the historical period according to the regional photovoltaic output sequences of each preset time length in the historical period;

[0016] Respectively take the occurrence probability of weather category i in the historical period and each climate factor in the historical period as the dependent variable sample set and the independent variable sample set, and use the least squares method to determine the multiple linear regression equation between the occurrence probability of weather category i and each climate factor according to the following formula:

[0017]

[0018] In the formula, Y i is the occurrence probability of weather category i, X j is the jth climate factor, β i,0 is the constant term of the multiple linear regression equation between the occurrence probability of weather category i and the climate factor, ε i is the error of the multiple linear regression equation between the occurrence probability of weather category i and the climate factor, β i,j is the weight of the jth climate factor in the multiple linear regression equation between the occurrence probability of weather category i and the climate factor, i ∈ [1, N], N is the total number of weather categories, j ∈ [1, M], M is the total number of climate factors.

[0019] Furthermore, the determining the occurrence probability of weather category i in the historical period according to the regional photovoltaic output sequences of each preset time length in the historical period includes:

[0020] Obtain the average value and the standard deviation of the second-order difference sequence corresponding to the regional photovoltaic output sequence of the s-th preset time length in the historical period;

[0021] When the average value of the regional photovoltaic output sequence of the s-th preset time length in the historical period is between the average value intervals of the regional photovoltaic output sequences corresponding to weather category i and the standard deviation of the second-order difference sequence is between the standard deviation intervals of the second-order difference sequences of the regional photovoltaic output sequences corresponding to weather category i, the weather category of the s-th preset time length in the historical period is weather category i;

[0022] Count the total number S of the preset time lengths corresponding to weather category i in the historical periodi and determine the occurrence probability \(P\) of weather category \(i\) in the historical period according to the following formula i :

[0023]

[0024] where \(s\in[1,S]\), \(S\) is the total number of preset time lengths in the historical period, \(i\in[1,N]\), and \(N\) is the total number of weather categories.

[0025] Furthermore, when weather category \(i\) is the clear sky category, the average value interval of the regional photovoltaic output sequence corresponding to weather category \(i\) is \([0.7,1]\), and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to weather category \(i\) is \([0,0.06)\);

[0026] when weather category \(i\) is the cloudy category, the average value interval of the regional photovoltaic output sequence corresponding to weather category \(i\) is \((0.4,0.9)\), and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to weather category \(i\) is \([0.06,+\infty)\);

[0027] when weather category \(i\) is the rainy and cloudy category, the average value interval of the regional photovoltaic output sequence corresponding to weather category \(i\) is \([0,0.04]\), and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to weather category \(i\) is \([0,+\infty)\).

[0028] Preferably, determining the occurrence probability of each weather category in the future period based on each climate factor in the future period and a pre-established multiple linear regression equation includes:

[0029] Substitute each climate factor in the future period into the multiple linear regression equation established based on each climate factor in the historical period and weather category \(i\) to obtain the occurrence probability of weather category \(i\) in the future period;

[0030] where \(i\in[1,N]\), and \(N\) is the total number of weather categories.

[0031] Preferably, determining the predicted sequence of the total regional photovoltaic output of each weather category in the future period based on the occurrence probability of each weather category in the future period includes:

[0032] Multiply the occurrence probability of weather category \(i\) in the future period by the total number of preset time lengths in the future period to obtain the total number of preset time lengths corresponding to weather category \(i\) in the future period;

[0033] Obtain the average value sequence of the regional photovoltaic output corresponding to weather category \(i\) in the historical period, and multiply it by the total number of preset time lengths corresponding to weather category \(i\) in the future period to obtain the predicted sequence of the total regional photovoltaic output corresponding to weather category \(i\) in the future period;

[0034] Among them, \(i\in[1,N]\), where \(N\) is the total number of weather categories.

[0035] Based on the same inventive concept, the present invention further provides a regional photovoltaic power output prediction device, which is improved in that the device includes:

[0036] An acquisition unit for acquiring various climate factors in a future time period;

[0037] A determination unit for determining the occurrence probability of each weather category in the future time period based on the various climate factors in the future time period and a pre-established multiple linear regression equation;

[0038] A prediction unit for determining a regional photovoltaic total power output prediction sequence of each weather category in the future time period based on the occurrence probability of each weather category in the future time period;

[0039] Among them, the multiple linear regression equation is established based on the occurrence probability of each weather category in the historical time period and the various climate factors in the historical time period.

[0040] Preferably, the device further includes a model construction unit, specifically for:

[0041] Acquiring various climate factors in the historical time period, and determining the occurrence probability of weather category \(i\) in the historical time period according to the regional photovoltaic power output sequences of each preset time length in the historical time period;

[0042] Respectively taking the occurrence probability of weather category \(i\) in the historical time period and the various climate factors in the historical time period as the dependent variable sample set and the independent variable sample set, and using the least squares method to determine the multiple linear regression equation between the occurrence probability of weather category \(i\) and the various climate factors according to the following formula:

[0043]

[0044] In the formula, \(Y\) i is the occurrence probability of weather category \(i\), \(X\) j is the \(j\)th climate factor, \(\beta\) i,0 is the constant term of the multiple linear regression equation between the occurrence probability of weather category \(i\) and the climate factors, \(\varepsilon\) i is the error of the multiple linear regression equation between the occurrence probability of weather category \(i\) and the climate factors, \(\beta\) i,j is the weight of the \(j\)th climate factor in the multiple linear regression equation between the occurrence probability of weather category \(i\) and the climate factors, \(i\in[1,N]\), \(N\) is the total number of weather categories, \(j\in[1,M]\), \(M\) is the total number of climate factors.

[0045] Compared with the closest prior art, the beneficial effects of the present invention are:

[0046] A method and device for predicting regional photovoltaic power output provided by the present invention include: obtaining various climate factors in a future time period; determining the occurrence probabilities of various weather categories in the future time period based on the various climate factors in the future time period and a pre-established multiple linear regression equation; determining the regional photovoltaic total power output prediction sequences of various weather categories in the future time period based on the occurrence probabilities of various weather categories in the future time period; wherein, the multiple linear regression equation is established based on the occurrence probabilities of various weather categories in a historical time period and the various climate factors in the historical time period. The present invention can achieve cross-season prediction of regional photovoltaic power output, making the prediction result not affected by seasonal factors and improving the accuracy of the prediction result; by establishing a multiple linear regression equation between the occurrence probabilities of different weather categories and various climate factors, the occurrence probabilities of various weather categories in the future time period obtained are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flowchart of the method for predicting regional photovoltaic power output of the present invention;

[0048] Figure 2 is a schematic diagram of the device for predicting regional photovoltaic power output of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The following further details the specific embodiments of the present invention with reference to the accompanying drawings.

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] Embodiment 1

[0052] The present invention provides a method for predicting regional photovoltaic power output, as Figure 1 shown, the method includes:

[0053] Step 1. Obtain various climate factors in a future time period;

[0054] Step 2. Determine the occurrence probabilities of various weather categories in the future time period based on the various climate factors in the future time period and a pre-established multiple linear regression equation;

[0055] Step 3. Determine the regional photovoltaic total power output prediction sequences of various weather categories in the future time period based on the occurrence probabilities of various weather categories in the future time period;

[0056] In Embodiment 1 of the present invention, the above-mentioned multiple linear regression equation is established based on the occurrence probabilities of each weather category in a historical period and each climate factor in the historical period.

[0057] In Embodiment 1 of the present invention, the above-mentioned weather categories include: clear sky category, rainy category, and cloudy category; the above-mentioned climate factors include: Arctic Oscillation, North Atlantic Oscillation, and Pacific-North American teleconnection.

[0058] In Embodiment 1 of the present invention, the above-mentioned step 1 includes:

[0059] Obtain the time series of climate factor j in a future period predicted by a weather forecasting system;

[0060] Use an interpolation algorithm to process the time series of climate factor j in the future period predicted by the weather forecasting system into a value, that is, climate factor j in the future period.

[0061] In Embodiment 1 of the present invention, the above-mentioned multiple linear regression equation is established based on the occurrence probabilities of each weather category in a historical period and each climate factor in the historical period, including:

[0062] Obtain each climate factor in the historical period, and determine the occurrence probability of weather category i in the historical period according to the regional photovoltaic power output sequences of each preset time length in the historical period;

[0063] Respectively take the occurrence probability of weather category i in the historical period and each climate factor in the historical period as the dependent variable sample set and the independent variable sample set, and use the least squares method to determine the multiple linear regression equation between the occurrence probability of weather category i and each climate factor according to the following formula:

[0064]

[0065] In the formula, Y i is the occurrence probability of weather category i, X j is the jth climate factor, β i,0 is the constant term of the multiple linear regression equation between the occurrence probability of weather category i and the climate factor, ε i is the error of the multiple linear regression equation between the occurrence probability of weather category i and the climate factor, β i,j is the weight of the jth climate factor in the multiple linear regression equation between the occurrence probability of weather category i and the climate factor, i ∈ [1, N], N is the total number of weather categories, j ∈ [1, M], M is the total number of climate factors;

[0066] Among them, the method for obtaining the climate factors in the historical period is the same as the method for obtaining the climate factors in the future period;

[0067] The time interval of the regional photovoltaic output sequences with each preset time length in the above historical period is 15 minutes, which is obtained by interpolating the original collected regional photovoltaic output sequences with each preset time length in the historical period.

[0068] Exemplarily, in the embodiment of the present invention, the historical month can be used as the historical period, and the historical day can be used as the preset time length.

[0069] Specifically, the determination of the occurrence probability of weather category i in the historical period according to the regional photovoltaic output sequences with each preset time length in the historical period includes:

[0070] Obtain the average value and the standard deviation of the second-order difference sequence corresponding to the regional photovoltaic output sequence with the s-th preset time length in the historical period;

[0071] When the average value of the regional photovoltaic output sequence with the s-th preset time length in the historical period is between the average value intervals of the regional photovoltaic output sequences corresponding to weather category i and the standard deviation of the second-order difference sequence is between the standard deviation intervals of the second-order difference sequences of the regional photovoltaic output sequences corresponding to weather category i, the weather category of the s-th preset time length in the historical period is weather category i;

[0072] Count the total number S of preset time lengths corresponding to weather category i in the historical period i , and determine the occurrence probability P of weather category i in the historical period according to the following formula i :

[0073]

[0074] where s ∈ [1, S], and S is the total number of preset time lengths in the historical period.

[0075] Further, when weather category i is the clear sky category, the average value interval of the regional photovoltaic output sequence corresponding to weather category i is [0.7, 1], and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to weather category i is [0, 0.06);

[0076] When weather category i is the cloudy category, the average value interval of the regional photovoltaic output sequence corresponding to weather category i is (0.4, 0.9), and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to weather category i is [0.06, +∞);

[0077] When weather category i is the rainy and cloudy category, the average value interval of the regional photovoltaic output sequence corresponding to weather category i is [0, 0.04], and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to weather category i is [0, +∞).

[0078] In Embodiment 1 of the present invention, the above step 2 includes:

[0079] Substitute the climate factors in the future time period into the multiple linear regression equation established based on the climate factors and weather category i in the historical time period to obtain the occurrence probability of weather category i in the future time period;

[0080] where i ∈ [1, N], and N is the total number of weather categories.

[0081] In Embodiment 1 of the present invention, the above step 3 includes:

[0082] Multiply the occurrence probability of weather category i in the future time period by the total number of preset time lengths in the future time period to obtain the total number of preset time lengths corresponding to weather category i in the future time period;

[0083] Obtain the average regional photovoltaic output sequence corresponding to weather category i in the historical time period, and multiply it by the total number of preset time lengths corresponding to weather category i in the future time period to obtain the predicted sequence of the total regional photovoltaic output corresponding to weather category i in the future time period;

[0084] where i ∈ [1, N], and N is the total number of weather categories.

[0085] Embodiment 2

[0086] Based on the same inventive concept, the present invention also provides a device for predicting regional photovoltaic output, as Figure 2 shown, the device includes:

[0087] An acquisition unit, configured to acquire the climate factors in the future time period;

[0088] A determination unit, configured to determine the occurrence probabilities of each weather category in the future time period based on the climate factors in the future time period and a pre-established multiple linear regression equation;

[0089] A prediction unit, configured to determine the predicted sequence of the total regional photovoltaic output of each weather category in the future time period based on the occurrence probabilities of each weather category in the future time period;

[0090] where the multiple linear regression equation is established based on the occurrence probabilities of each weather category in the historical time period and the climate factors in the historical time period.

[0091] In Embodiment 2 of the present invention, the above weather categories include: clear sky category, rainy category, and cloudy category;

[0092] The above climate factors include: Arctic Oscillation, North Atlantic Oscillation, and Pacific-North American teleconnection pattern.

[0093] In Embodiment 2 of the present invention, the above acquisition unit is specifically configured to:

[0094] Obtain the time series of climate factor j for the future period predicted by the weather forecasting system;

[0095] Use an interpolation algorithm to process the time series of climate factor j for the future period predicted by the weather forecasting system into a value, that is, the climate factor j for the future period.

[0096] In Embodiment 2 of the present invention, the above device further includes a model construction unit, specifically used for:

[0097] Obtain the climate factors for the historical period, and determine the occurrence probability of weather category i in the historical period according to the regional photovoltaic power output sequences with each preset time length in the historical period;

[0098] Respectively take the occurrence probability of weather category i in the historical period and the climate factors in the historical period as the dependent variable sample set and the independent variable sample set, and use the least squares method to determine the multiple linear regression equation between the occurrence probability of weather category i and the climate factors according to the following formula:

[0099]

[0100] In the formula, Y i is the occurrence probability of weather category i, X j is the jth climate factor, β i,0 is the constant term of the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, ε i is the error of the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, β i,j is the weight of the jth climate factor in the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, i ∈ [1, N], N is the total number of weather categories, j ∈ [1, M], M is the total number of climate factors;

[0101] Among them, the method for obtaining the climate factors in the historical period is the same as the method for obtaining the climate factors in the future period;

[0102] The time interval of the regional photovoltaic power output sequences with each preset time length in the above historical period is 15 min, and it is obtained by processing the original collected regional photovoltaic power output sequences with each preset time length in the historical period through an interpolation algorithm.

[0103] Specifically, the above determination of the occurrence probability of weather category i in the historical period according to the regional photovoltaic power output sequences with each preset time length in the historical period includes:

[0104] Obtain the average value and the standard deviation of the second-order difference sequence corresponding to the regional photovoltaic power output sequence with the sth preset time length in the historical period;

[0105] When the average value of the regional photovoltaic output sequence with the s-th preset time length in the historical period is between the average value intervals of the regional photovoltaic output sequences corresponding to the weather category i and the standard deviation of the second-order difference sequence is between the standard deviation intervals of the second-order difference sequences of the regional photovoltaic output sequences corresponding to the weather category i, the weather category of the s-th preset time length in the historical period is the weather category i;

[0106] Statistically count the total number S of preset time lengths corresponding to the weather category i in the historical period i , and determine the occurrence probability P of the weather category i in the historical period according to the following formula i :

[0107]

[0108] where s ∈ [1, S], and S is the total number of preset time lengths in the historical period.

[0109] Furthermore, when the weather category i is the clear sky category, the average value interval of the regional photovoltaic output sequence corresponding to the weather category i is [0.7, 1], and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to the weather category i is [0, 0.06);

[0110] When the weather category i is the cloudy category, the average value interval of the regional photovoltaic output sequence corresponding to the weather category i is (0.4, 0.9), and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to the weather category i is [0.06, +∞);

[0111] When the weather category i is the rainy and cloudy category, the average value interval of the regional photovoltaic output sequence corresponding to the weather category i is [0, 0.04], and the standard deviation interval of the second-order difference sequence of the regional photovoltaic output sequence corresponding to the weather category i is [0, +∞).

[0112] In Embodiment 2 of the present invention, the above-mentioned determination unit is specifically configured to:

[0113] Substitute the climate factors in the future period into the multiple linear regression equation established based on the climate factors and the weather category i in the historical period to obtain the occurrence probability of the weather category i in the future period;

[0114] where i ∈ [1, N], and N is the total number of weather categories.

[0115] In Embodiment 2 of the present invention, the above-mentioned prediction unit is specifically configured to:

[0116] Multiply the occurrence probability of the weather category i in the future period by the total number of preset time lengths in the future period to obtain the total number of preset time lengths corresponding to the weather category i in the future period;

[0117] Obtain the average value sequence of the regional photovoltaic output corresponding to weather category i in the historical period, and multiply it by the total number of preset time lengths corresponding to weather category i in the future period to obtain the predicted sequence of the total regional photovoltaic output corresponding to weather category i in the future period;

[0118] where i ∈ [1, N], and N is the total number of weather categories.

[0119] In summary, a method and device for predicting regional photovoltaic output provided by the present invention include: obtaining various climate factors in the future period; determining the occurrence probabilities of various weather categories in the future period based on the various climate factors in the future period and a pre-established multiple linear regression equation; determining the predicted sequence of the total regional photovoltaic output of various weather categories in the future period based on the occurrence probabilities of various weather categories in the future period; wherein, the multiple linear regression equation is established based on the occurrence probabilities of various weather categories in the historical period and the various climate factors in the historical period; the present invention can realize cross-season prediction of regional photovoltaic output, make the prediction result not affected by seasonal factors, and improve the accuracy of the prediction result; by establishing a multiple linear regression equation between the occurrence probabilities of different weather categories and various climate factors, the occurrence probabilities of various weather categories in the future period obtained are more accurate.

[0120] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0121] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0122] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one process Figure 1 one process or a plurality of processes and / or boxes Figure 1 specified in one box or a plurality of boxes.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process Figure 1 one process or a plurality of processes and / or boxes Figure 1 specified in one box or a plurality of boxes.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention. Any modification or equivalent replacement without departing from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A method for predicting regional photovoltaic power output, characterized in that, The method includes: Obtaining each climate factor for a future time period; Determining the occurrence probability of each weather category in the future time period based on each climate factor in the future time period and a pre-established multiple linear regression equation; Determining the regional total PV output prediction sequence of each weather category in the future time period based on the occurrence probability of each weather category in the future time period; Wherein, the multiple linear regression equation is established based on the occurrence probability of each weather category in the historical time period and each climate factor in the historical time period; The climate factors include: Arctic Oscillation, North Atlantic Oscillation, and Pacific-North American teleconnection; The multiple linear regression equation is established based on the occurrence probability of each weather category in the historical time period and each climate factor in the historical time period, including: Obtaining each climate factor in the historical time period, and determining the occurrence probability of weather category i in the historical time period according to the regional PV output sequence of each preset time length in the historical time period; Respectively taking the occurrence probability of weather category i in the historical time period and each climate factor in the historical time period as the dependent variable sample set and the independent variable sample set, and using the least squares method to determine the multiple linear regression equation between the occurrence probability of weather category i and each climate factor according to the following formula: Where Y i is the occurrence probability of weather category i, X j is the j-th climate factor, β i，0 is the constant term of the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, ε i is the error of the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, β i，j is the weight of the j-th climate factor in the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, i ∈ [1, N], N is the total number of weather categories, j ∈ [1, M], M is the total number of climate factors; The determining the occurrence probability of weather category i in the historical time period according to the regional PV output sequence of each preset time length in the historical time period includes: Obtaining the average value and the standard deviation of the second-order difference sequence corresponding to the regional PV output sequence of the s-th preset time length in the historical time period; When the average value of the regional PV output sequence of the s-th preset time length in the historical time period is between the average value intervals of the regional PV output sequences corresponding to weather category i and the standard deviation of the second-order difference sequence is between the standard deviation intervals of the second-order difference sequences of the regional PV output sequences corresponding to weather category i, the weather category of the s-th preset time length in the historical time period is weather category i; Statistically count the total number S of preset time lengths corresponding to weather category i in the historical period i , and determine the occurrence probability P of weather category i in the historical period according to the following formula i : Wherein, s ∈ [1, S], and S is the total number of preset time lengths in the historical time period; The determining the regional total PV output prediction sequence of each weather category in the future time period based on the occurrence probability of each weather category in the future time period includes: Multiplying the occurrence probability of weather category i in the future time period by the total number of preset time lengths in the future time period to obtain the total number of preset time lengths corresponding to weather category i in the future time period; Obtaining the average value sequence of the regional PV output corresponding to weather category i in the historical time period, and multiplying it by the total number of preset time lengths corresponding to weather category i in the future time period to obtain the regional total PV output prediction sequence corresponding to weather category i in the future time period; Wherein, i ∈ [1, N], and N is the total number of weather categories.

2. The method according to claim 1, characterized in that, The weather categories include: clear sky category, rainy category, and cloudy category.

3. The method according to claim 1, characterized in that, When weather category i is the clear sky category, the average value interval of the regional PV output sequence corresponding to weather category i is [0.7, 1], and the standard deviation interval of the second-order difference sequence of the regional PV output sequence corresponding to weather category i is [0, 0.06); When the weather category i is the cloudy category, the average value range of the regional photovoltaic output sequence corresponding to the weather category i is (0.4, 0.9), and the standard deviation range of the second-order difference sequence of the regional photovoltaic output sequence corresponding to the weather category i is [0.06, +∞); When the weather category i is the rainy and cloudy category, the average value range of the regional photovoltaic output sequence corresponding to the weather category i is [0, 0.04], and the standard deviation range of the second-order difference sequence of the regional photovoltaic output sequence corresponding to the weather category i is [0, +∞).

4. The method according to claim 1, characterized in that, Determining the occurrence probabilities of the weather categories in the future period based on the climate factors in the future period and the pre-established multiple linear regression equation includes: Substituting the climate factors in the future period into the multiple linear regression equation established for the climate factors and the weather category i in the historical period to obtain the occurrence probability of the weather category i in the future period; where i ∈ [1, N], and N is the total number of weather categories.

5. A device for predicting regional photovoltaic power output, characterized in that, The device includes: An acquisition unit for acquiring the climate factors in the future period; A determination unit for determining the occurrence probabilities of the weather categories in the future period based on the climate factors in the future period and the pre-established multiple linear regression equation; A prediction unit for determining the predicted sequence of the total regional photovoltaic output of the weather categories in the future period based on the occurrence probabilities of the weather categories in the future period; where the multiple linear regression equation is established based on the occurrence probabilities of the weather categories in the historical period and the climate factors in the historical period; The device further includes a model construction unit, specifically for: Acquiring the climate factors in the historical period and determining the occurrence probability of the weather category i in the historical period according to the regional photovoltaic output sequences of each preset time length in the historical period; Respectively taking the occurrence probability of the weather category i in the historical period and the climate factors in the historical period as the dependent variable sample set and the independent variable sample set, and using the least squares method to determine the multiple linear regression equation between the occurrence probability of the weather category i and the climate factors according to the following formula: where Y i is the occurrence probability of weather category i, X j is the j-th climate factor, β i，0 is the constant term of the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, ε i is the error of the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, β i，j is the weight of the j-th climate factor in the multiple linear regression equation between the occurrence probability of weather category i and the climate factors, i ∈ [1, N], N is the total number of weather categories, j ∈ [1, M], M is the total number of climate factors; where determining the occurrence probability of the weather category i in the historical period according to the regional photovoltaic output sequences of each preset time length in the historical period specifically includes: Acquiring the average value and the standard deviation of the second-order difference sequence corresponding to the regional photovoltaic output sequence of the s-th preset time length in the historical period; When the average value of the regional photovoltaic output sequence of the s-th preset time length in the historical period is between the average value range of the regional photovoltaic output sequence corresponding to the weather category i and the standard deviation of the second-order difference sequence is between the standard deviation range of the second-order difference sequence of the regional photovoltaic output sequence corresponding to the weather category i, the weather category of the s-th preset time length in the historical period is the weather category i; Statistically count the total number S of preset time lengths corresponding to weather category i in the historical period i , and determine the occurrence probability P of weather category i in the historical period according to the following formula i : where s ∈ [1, S], and S is the total number of preset time lengths in the historical period; The prediction unit, specifically for: Multiplying the occurrence probability of the weather category i in the future period by the total number of preset time lengths in the future period to obtain the total number of preset time lengths corresponding to the weather category i in the future period; Obtain the average value sequence of the regional photovoltaic power output corresponding to weather category i in the historical period, and multiply it by the total number of preset time lengths corresponding to weather category i in the future period to obtain the predicted sequence of the total regional photovoltaic power output corresponding to weather category i in the future period; where i ∈ [1, N], and N is the total number of weather categories; The climate factors include: Arctic Oscillation, North Atlantic Oscillation, and Pacific-North American teleconnection.

Images