New energy power prediction method and system based on meteorological conditions and cluster division

Abstract

The present invention relates to the field of renewable energy power generation technology, and in particular to a renewable energy power prediction method and system based on meteorological conditions and cluster division. The method and system provided by the present invention are used to collect digital weather forecast data of multiple spatial and temporal scales in a region and digital weather forecast data reported by a station, historical operation data and measured meteorological data. Cluster division principles, benchmark power station screening models, cluster prediction models, electric energy calculation models and confidence intervals are established through big data analysis technology. A benchmark power station is selected and a cluster power prediction curve is calculated. An electric energy model is used to calculate a confidence interval curve. The meteorological conditions are integrated with the power prediction to achieve more accurate correction of the cluster prediction data.

Description

Technical Field

[0001] The present invention relates to the field of renewable energy power generation technology, and in particular to a renewable energy power prediction method and system based on meteorological conditions and cluster division. Background Art

[0002] In traditional technology, power plant power prediction methods are mainly divided into statistical methods and physical methods, and each method has its own advantages and disadvantages. The physical method does not require a large amount of measurement data, but requires an accurate mathematical description of the physical properties of the atmosphere and the characteristics of the power plant. These equations are difficult to solve and require a large amount of calculation. The statistical method does not require solving physical equations and has a fast calculation speed, but requires a large amount of historical data. At the same time, most of the current prediction manufacturers do not provide cluster power predictions. Some manufacturers calculate cluster power predictions by simply accumulating power plant power predictions, which has low accuracy and cannot meet the actual needs of guiding scheduling decisions.

[0003] Therefore, how to integrate meteorological conditions with power forecasts to improve the accuracy of power forecasts and achieve accurate prediction of the total power of renewable energy in a larger range has become an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0004] The purpose of the present invention is to provide a new energy power prediction method and system based on meteorological conditions and cluster division to solve the above problems in the prior art.

[0005] The embodiments of the present invention are implemented by the following technical solutions:

[0006] In a first aspect, the present invention provides a new energy power prediction method based on meteorological conditions and cluster division, comprising:

[0007] Obtain historical operating data and obtain measured meteorological data through the weather forecast system;

[0008] Divide the power stations in the area into clusters and output several clusters;

[0009] A benchmark power plant screening model is established based on historical operation data, and benchmark power plants are screened out from a number of clusters through the benchmark power plant screening model;

[0010] Obtain historical operating data and measured meteorological data of the benchmark power station, establish a cluster prediction model, and output the cluster power prediction curve of the benchmark power station through the cluster prediction model;

[0011] The confidence interval curve is calculated through the electric energy model to correct the cluster prediction data.

[0012] Preferably, the dividing of the power stations in the area into clusters includes:

[0013] Obtain the geographical location information of all power stations connected to the power system and set the coverage area of ​​a single cluster;

[0014] Get the area of ​​the region to which it belongs, and divide the area of ​​the region according to the coverage area of ​​the set single cluster;

[0015] The power stations located within the coverage area of ​​a single cluster are divided into one cluster.

[0016] Preferably, the establishing of the benchmark power station screening model comprises:

[0017] Obtain the real-time power data of all power plants and cluster power data, calculate the correlation coefficient between power plants and cluster power, and sort them;

[0018] Set a base point value of the correlation coefficient, and retain the first power station corresponding to the correlation coefficient greater than the base point value;

[0019] Obtain data from the first power station, establish a data matrix, and calculate the correlation coefficient in the data matrix;

[0020] Calculate the correlation between each power station's power generation series and the cluster's power generation series, and sort the correlations;

[0021] A correlation selection threshold is set, the second power stations with correlations greater than the correlation selection threshold are retained, and the power station with the largest installed capacity among the second power stations is selected as the benchmark power station.

[0022] Preferably, the calculation of the correlation coefficient between the power station and the cluster power includes:

[0023]

[0024] In the formula, r XY is the correlation coefficient, E is the mathematical expectation, cov(X,Y) is the covariance of random variables X and Y, σ X is the standard deviation of the random variable X, σ Y is the standard deviation of the random variable Y, μ X is the mean of X, μ Y is the mean of Y.

[0025] Preferably, the calculation of the correlation coefficient in the data matrix includes:

[0026]

[0027] Where, L oi (k) is the correlation coefficient, Δ min is the minimum value of the absolute difference between the power generation of the cluster and the power plant at each moment, Δ maxis the maximum absolute difference of the values ​​at each moment in the comparison sequence of the power generation of the cluster and the power plant, p is the resolution coefficient, Δ oi (k) is the absolute difference between the power generation sequence of the cluster in year k and the power generation sequence of each power station in the ith year.

[0028] Preferably, the calculating of the correlation between each power station power generation series and the cluster power generation series includes:

[0029]

[0030] In the formula, r oi is the correlation degree, n and k are constants.

[0031] Preferably, the establishing of the cluster prediction model comprises:

[0032] Obtaining the measured annual meteorological data of the benchmark power station, establishing a first time series, obtaining the forecast annual meteorological data, establishing a second time series, and establishing a function model based on the first time series and the second time series;

[0033] Correct the forecasted annual meteorological data through the function model, and establish a probability distribution diagram based on the corrected annual meteorological data;

[0034] The cluster prediction power value is calculated through the probability distribution diagram of the annual meteorological data, and the cluster prediction curve is obtained through the cluster prediction power.

[0035] Preferably, the computing cluster predicting power value includes:

[0036] P=0.5C p ρAv 3 +KI mpp U mpp

[0037] Where P is the cluster prediction power value, C p is the wind energy utilization coefficient, ρ is the air density, A is the total area of ​​the wind turbine swept by the airflow, v is the wind speed in the forecast annual meteorological data, K is the photovoltaic module utilization coefficient, I mpp is the output current of the photovoltaic module, U mpp The output voltage of the photovoltaic module.

[0038] In a second aspect, the present invention further provides a new energy power prediction system based on meteorological conditions and cluster division, comprising:

[0039] The processing module is configured to obtain historical operation data and obtain measured meteorological data through a weather forecast system; divide the power stations in the area into clusters and output a number of clusters;

[0040] The model prediction block is configured to establish a benchmark power station screening model based on historical operation data, and screen out benchmark power stations from a number of clusters through the benchmark power station screening model; obtain historical operation data and measured meteorological data of the benchmark power station, establish a cluster prediction model, and output the cluster power prediction curve of the benchmark power station through the cluster prediction model; calculate the confidence interval curve through the electric energy model, and correct the cluster prediction data;

[0041] The main control module is connected to the processing module and the model prediction block, and is used to execute the above-mentioned new energy power prediction method based on meteorological conditions and cluster division.

[0042] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0043] By adopting the method and system provided by the present invention, digital weather forecast data of multiple spatial and temporal scales in the region and digital weather forecast data reported by the stations, historical operation data and measured meteorological data are collected, and cluster division principles, benchmark power station screening models, cluster prediction models, electric energy calculation models and confidence intervals are established through big data analysis technology; benchmark power stations are selected and cluster power prediction curves are calculated, and the confidence interval curves are calculated using the electric energy model, and meteorological conditions are integrated with power predictions to achieve more accurate correction of cluster prediction data. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A schematic diagram of a process for establishing a benchmark power plant screening model for the present invention;

[0046] Figure 2 A schematic diagram of a process for establishing a cluster prediction model in the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, 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 in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0048] The division of modules in this application is a logical division. There may be other division methods when implemented in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0049] The modules or submodules described independently may be physically separated or not: they may be implemented by software or hardware, and some modules or submodules may be implemented by software, and the processor may call the software to implement the functions of these modules or submodules, and other modules or submodules may be implemented by hardware, such as by hardware circuits. In addition, some or all of the modules may be selected according to actual needs to achieve the purpose of the present application.

[0050] Please refer to Figure 1-Figure 2 The present invention provides a new energy power prediction method based on meteorological conditions and cluster division, including:

[0051] S101: Acquire historical operation data and obtain measured meteorological data through a weather forecast system;

[0052] The following methods can be used to collect digital weather forecast data of multiple spatial and temporal scales in the region, as well as digital weather forecast data reported by stations, historical operation data and measured meteorological data:

[0053] Official meteorological data service websites: Visit official websites such as the China Meteorological Data Network and the National Meteorological Science Data Center. These websites provide weather forecast data and historical data covering all parts of the country. Users can choose data of different time and space scales as needed.

[0054] Meteorological information sharing platform: Join a meteorological information sharing platform, such as the China Meteorological Administration's information sharing platform, through which you can obtain official weather forecast data and data shared by other agencies.

[0055] Cooperation with local meteorological bureaus: Establish cooperative relations with local meteorological bureaus and obtain digital weather forecast data and related measured data directly from local meteorological bureaus.

[0056] Station data reporting system: Establish or use the existing station data reporting system, through which the station can report real-time digital weather forecast data and historical operation data.

[0057] Automatic weather station data: Install an automatic weather station to collect real-time meteorological data such as temperature, humidity, wind speed, and precipitation.

[0058] S102: Divide the power stations in the area into clusters and output a number of clusters;

[0059] S103: establishing a benchmark power plant screening model based on historical operation data, and screening out benchmark power plants from a number of clusters using the benchmark power plant screening model;

[0060] S104: Acquire historical operation data and measured meteorological data of the benchmark power station, establish a cluster prediction model, and output a cluster power prediction curve of the benchmark power station through the cluster prediction model;

[0061] S105: Calculate the confidence interval curve through the electric energy model and correct the cluster prediction data.

[0062] By adopting the method and system provided by the present invention, digital weather forecast data of multiple spatial and temporal scales in the region and digital weather forecast data reported by the stations, historical operation data and measured meteorological data are collected, and cluster division principles, benchmark power station screening models, cluster prediction models, electric energy calculation models and confidence intervals are established through big data analysis technology; benchmark power stations are selected and cluster power prediction curves are calculated, and the confidence interval curves are calculated using the electric energy model, and meteorological conditions are integrated with power predictions to achieve more accurate correction of cluster prediction data.

[0063] In an exemplary embodiment of the present invention, dividing the power stations in the region into clusters includes:

[0064] The geographical location information of all power stations connected to the power system is obtained, and the coverage area of ​​a single cluster is set; the area of ​​the region to which the power station belongs is obtained, and the area of ​​the region to which the power station belongs is divided according to the coverage area of ​​the set single cluster; and the power stations located within the coverage area of ​​the single cluster are divided into one cluster.

[0065] Specifically, power stations with similar geographical characteristics, within 30 kilometers of each other, and in the same section of the power grid are divided into clusters to improve the accuracy of fitting the statistical characteristics of output power. The division dimensions include at least:

[0066] Based on the geographical distribution of power stations; based on the location where the power stations are connected to the power system; based on the characteristics of the cluster resources.

[0067] In an exemplary embodiment of the present invention, establishing a benchmark power plant screening model includes:

[0068] S201: Acquire real-time power data of all power stations and cluster power data, calculate the correlation coefficient between power stations and cluster power, and sort them;

[0069] Specifically, the real-time power data X of the power plant and the cluster power data Y are input, and the correlation coefficient between the power plant and the cluster power is calculated and sorted by the Wilson correlation coefficient.

[0070] The correlation coefficients for calculating power plant and cluster power include:

[0071]

[0072] In the formula, r XY is the correlation coefficient, E is the mathematical expectation, cov(X,Y) is the covariance of random variables X and Y, σ X is the standard deviation of the random variable X, σ Y is the standard deviation of the random variable Y, μ X is the mean of X, μ Y is the mean of Y.

[0073] S202: Setting a base point value of the correlation coefficient, and retaining the first power station corresponding to the correlation coefficient greater than the base point value;

[0074] In this embodiment, the first power station corresponding to the correlation coefficient ranked in the top 30% is selected.

[0075] S203: Acquire data of the first power station, establish a data matrix, and calculate a correlation coefficient in the data matrix;

[0076] Among them, the data matrix is ​​established by using the cluster division principle to select the benchmark power station within the cluster using the grey correlation between cluster power and power station power, that is, the benchmark power station is selected through the time series analysis of each power station and cluster power generation. m time series of candidate power stations;

[0077]

[0078] Among them, n represents the length of the power station data sequence, m represents the number of candidate power stations, and x is the specific data.

[0079] Calculating the correlation coefficient in a data matrix involves:

[0080]

[0081] Where, L oi (k) is the correlation coefficient, Δ min is the minimum value of the absolute difference between the power generation of the cluster and the power plant at each moment, Δ max The maximum value of the absolute difference of the values ​​at each moment in the comparison sequence of the power generation of the cluster and the power station. P is introduced to avoid the distortion caused by the maximum and minimum values ​​of the absolute difference of the comparison sequence. The value is between 0-1, usually between 0.1-0.5. oi (k) is the absolute difference between the power generation sequence of the cluster in year k and the power generation sequence of each power station in the ith year.

[0082] S204: Calculate the correlation between each power station power generation series and the cluster power generation series, and sort the correlations;

[0083] Specifically, calculating the correlation between each power station power generation series and the cluster power generation series includes:

[0084]

[0085] In the formula, r oi is the correlation degree, n and k are constants.

[0086] S205: Setting a correlation selection threshold, retaining the second power stations whose correlation is greater than the correlation selection threshold, and selecting the power station with the largest installed capacity among the second power stations as the benchmark power station.

[0087] In this embodiment, 50% of the standby power stations m are selected as reference power stations, and the power station with the largest installed capacity is selected as the benchmark power station.

[0088] An exemplary implementation of the present invention is that simply doubling the benchmark power station does not conform to the actual situation and ignores the resource differences within the cluster. The nonlinear relationship between the power station output and the cluster output is calculated by combining statistical methods and physical methods, and finally the cluster power prediction and confidence interval prediction are calculated by superimposing the predicted output of the cluster.

[0089] Building a cluster prediction model includes:

[0090] S301: Obtain the measured annual meteorological data of the reference power station, establish a first time series, obtain the forecast annual meteorological data, establish a second time series, and establish a function model based on the first time series and the second time series;

[0091] Specifically, considering the errors in the digital weather forecast data, in order to more accurately reflect the meteorological changes of the benchmark power station, the meteorological data of the digital weather forecast is corrected. First, the meteorological correlation equation between the measured meteorology and the digital weather forecast is established by using the time series y of the benchmark power station's annual meteorological formation and the time series x of the benchmark power station's annual meteorological formation.

[0092] y=f(x)

[0093] Among them, regarding the specific equation or function model, a linear relationship can be obtained according to the error rate between the actual weather at the local location and the forecasted weather.

[0094] S302: Correcting the forecasted annual meteorological data through a function model, and establishing a probability distribution diagram based on the corrected annual meteorological data;

[0095] The maximum probability weather, maximum weather value and minimum weather value can be obtained through the probability distribution diagram.

[0096] S303: Calculate the cluster prediction power value through the probability distribution diagram of the annual meteorological data, and obtain the cluster prediction curve through the cluster prediction power.

[0097] The cluster prediction curve, maximum output curve and minimum output curve are output through the maximum probability meteorological value, maximum meteorological value and minimum meteorological value in the probability distribution diagram.

[0098] In an exemplary embodiment of the present invention, the computing cluster predicting power value includes:

[0099] P=0.5C p ρAv 3 +KI mpp U mpp

[0100] Where P is the cluster prediction power value, C p is the wind energy utilization coefficient, ρ is the air density, A is the total area of ​​the wind turbine swept by the airflow, v is the wind speed in the forecast annual meteorological data, K is the photovoltaic module utilization coefficient, I mpp is the output current of the photovoltaic module, U mpp The output voltage of the photovoltaic module.

[0101] In a second aspect, the present invention further provides a new energy power prediction system based on meteorological conditions and cluster division, comprising:

[0102] The processing module is configured to obtain historical operation data and obtain measured meteorological data through a weather forecast system; divide the power stations in the area into clusters and output a number of clusters;

[0103] The model prediction block is configured to establish a benchmark power station screening model based on historical operation data, and screen out benchmark power stations from a number of clusters through the benchmark power station screening model; obtain historical operation data and measured meteorological data of the benchmark power station, establish a cluster prediction model, and output the cluster power prediction curve of the benchmark power station through the cluster prediction model; calculate the confidence interval curve through the electric energy model, and correct the cluster prediction data;

[0104] The main control module is connected to the processing module and the model prediction block, and is used to execute the above-mentioned new energy power prediction method based on meteorological conditions and cluster division.

[0105] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0106] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0107] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A new energy power prediction method based on meteorological conditions and cluster division, characterized in that: include: Obtain historical operating data and obtain measured meteorological data through the weather forecast system; Divide the power stations in the area into clusters and output several clusters; A benchmark power plant screening model is established based on historical operation data, and benchmark power plants are screened out from a number of clusters through the benchmark power plant screening model; Obtain historical operating data and measured meteorological data of the benchmark power station, establish a cluster prediction model, and output the cluster power prediction curve of the benchmark power station through the cluster prediction model; The confidence interval curve is calculated through the electric energy model to correct the cluster prediction data.

2. The new energy power prediction method based on meteorological conditions and cluster division according to claim 1 is characterized in that: The clustering of the power stations in the area includes: Obtain the geographical location information of all power stations connected to the power system and set the coverage area of ​​a single cluster; Get the area of ​​the region to which it belongs, and divide the area of ​​the region according to the coverage area of ​​the set single cluster; The power stations located within the coverage area of ​​a single cluster are divided into one cluster.

3. The new energy power prediction method based on meteorological conditions and cluster division according to claim 1 is characterized in that: The establishment of the benchmark power station screening model comprises: Obtain the real-time power data of all power plants and cluster power data, calculate the correlation coefficient between power plants and cluster power, and sort them; Set a base point value of the correlation coefficient, and retain the first power station corresponding to the correlation coefficient greater than the base point value; Obtain data from the first power station, establish a data matrix, and calculate the correlation coefficient in the data matrix; Calculate the correlation between each power station's power generation series and the cluster's power generation series, and sort the correlations; A correlation selection threshold is set, the second power stations with correlations greater than the correlation selection threshold are retained, and the power station with the largest installed capacity among the second power stations is selected as the benchmark power station.

4. The new energy power prediction method based on meteorological conditions and cluster division according to claim 3 is characterized in that: The calculation of the correlation coefficient between the power station and the cluster power includes: In the formula, r XY is the correlation coefficient, E is the mathematical expectation, cov(X,Y) is the covariance of random variables X and Y, σ X is the standard deviation of the random variable X, σ Y is the standard deviation of the random variable Y, μ X is the mean of X, μ Y is the mean of Y.

5. The new energy power prediction method based on meteorological conditions and cluster division according to claim 3 is characterized in that: The calculation of the correlation coefficient in the data matrix includes: Where, L oi (k) is the correlation coefficient, Δ min is the minimum value of the absolute difference between the power generation of the cluster and the power plant at each moment, Δ max is the maximum absolute difference of the values ​​at each moment in the comparison sequence of the power generation of the cluster and the power plant, p is the resolution coefficient, Δ oi (k) is the absolute difference between the power generation sequence of the cluster in year k and the power generation sequence of each power station in the ith year.

6. The new energy power prediction method based on meteorological conditions and cluster division according to claim 3 is characterized in that: The calculation of the correlation between each power station power generation series and the cluster power generation series includes: In the formula, r oi is the correlation degree, n and k are constants.

7. The new energy power prediction method based on meteorological conditions and cluster division according to claim 3 is characterized in that: The establishing of the cluster prediction model comprises: Obtaining the measured annual meteorological data of the benchmark power station, establishing a first time series, obtaining the forecast annual meteorological data, establishing a second time series, and establishing a function model based on the first time series and the second time series; Correct the forecasted annual meteorological data through the function model, and establish a probability distribution diagram based on the corrected annual meteorological data; The cluster prediction power value is calculated through the probability distribution diagram of the annual meteorological data, and the cluster prediction curve is obtained through the cluster prediction power.

8. The new energy power prediction method based on meteorological conditions and cluster division according to claim 3 is characterized in that: The computing cluster predicts the power value including: P=0.5C p ρAv 3 +KI mpp HE mpp Where P is the cluster prediction power value, C p is the wind energy utilization coefficient, ρ is the air density, A is the total area of ​​the wind turbine swept by the airflow, v is the wind speed in the forecast annual meteorological data, K is the photovoltaic module utilization coefficient, I mpp is the output current of the photovoltaic module, U mpp The output voltage of the photovoltaic module.

9. A new energy power prediction system based on meteorological conditions and cluster division, characterized in that: include: a processing module configured to obtain historical operation data and obtain measured meteorological data through a weather forecast system; Divide the power stations in the area into clusters and output several clusters; A model prediction block is configured to establish a benchmark power plant screening model based on historical operation data, and screen out benchmark power plants from a number of clusters through the benchmark power plant screening model; Obtain historical operating data and measured meteorological data of the benchmark power station, establish a cluster prediction model, and output the cluster power prediction curve of the benchmark power station through the cluster prediction model; The confidence interval curve is calculated through the electric energy model to correct the cluster prediction data; The main control module is connected to the processing module and the model prediction block, and is used to execute the new energy power prediction method based on meteorological conditions and cluster division as described in any one of claims 1-8.

Images