Short-term irradiance prediction method and system based on historical data analysis

A technology of historical data and forecasting methods, applied in the field of short-term irradiance forecasting methods and forecasting systems, can solve problems such as heavy workload and complicated forecasting steps, and achieve the effect of improving forecasting accuracy

Pending Publication Date: 2022-01-21
HUANENG LANCANG RIVER HYDROPOWER +3
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The method in this application can effectively improve the accuracy of irradiance prediction, but it needs to predict and superimpose multiple sets of data, the prediction steps are complicated and the workload is heavy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Short-term irradiance prediction method and system based on historical data analysis
  • Short-term irradiance prediction method and system based on historical data analysis
  • Short-term irradiance prediction method and system based on historical data analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0071] The short-term irradiance prediction method based on historical data analysis of this embodiment, such as Figure 11 shown, including the following steps:

[0072] S10, historical data acquisition,

[0073] Obtain the historical solar irradiance data of the target area for one year;

[0074] S20, data preprocessing,

[0075] Firstly, the solar irradiance data obtained by S10 are divided into four types of seasonal data series: spring, summer, autumn and winter according to their time periods. Among them, the data from March to May are divided into spring data series; The data is divided into summer data series; the data from September to November is divided into autumn data series; the data series from December to February is divided into winter data series;

[0076] Then according to the irradiance of the solar irradiance data acquired by S10, all kinds of seasonal data series are divided into three types of weather data series: sunny day, cloudy day and rainy day, ...

Embodiment 2

[0117] The short-term irradiance prediction method based on historical data analysis of this embodiment, such as Figure 11 shown, including the following steps:

[0118] S10, historical data acquisition,

[0119] Obtain the historical solar irradiance data of the target area for one year;

[0120] S20, data preprocessing,

[0121] Firstly, the solar irradiance data obtained by S10 are divided into four types of seasonal data series: spring, summer, autumn and winter according to their time periods. Among them, the data from March to May are divided into spring data series; The data is divided into summer data series; the data from September to November is divided into autumn data series; the data series from December to February is divided into winter data series;

[0122] Then, according to the irradiance of the solar irradiance data acquired by S10, all kinds of seasonal data sequences are divided into three types of weather data sequences: sunny day, cloudy day and rain...

Embodiment 3

[0168] This embodiment utilizes the method of embodiment 2 to predict the weather on a certain day in a certain area, classify the spring irradiance in a certain area, and use the total 20 days of the weather type data sequence corresponding to the forecast day, and the data sequence of 20×24h=480h as The LSTM training and test data predict the 24h solar irradiance of the target day, the first 19 days are the training data, and the last day is the test data, and are compared with the actual solar irradiance data sequence of the target day.

[0169] In the LSTM neural network of this embodiment, the number of hidden layer units n=96*3, the maximum number of iterations m=250 times, m 1 = 125, the initial learning rate r = 0.005, the learning drop factor is f = 0.2, and the gradient threshold g = 1.

[0170] Depend on Figure 7 It can be seen that the fluctuation range of the sunny type irradiance data series is smaller than that of the actual data series, and is similar to the ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a short-term irradiance prediction method and prediction system based on historical data analysis, and belongs to the technical field of photovoltaic power generation. The prediction method comprises the following steps: S10, collecting historical data; s20, data preprocessing: dividing the solar irradiance data obtained in the step S10 into four seasonal data sequences of spring, summer, autumn and winter, and then dividing each seasonal data sequence into three weather data sequences of sunny days, cloudy days and rainy days; s30, training data matching: obtaining weather type data of the target prediction day according to the weather forecast, and taking a corresponding weather data sequence of the season corresponding to the weather type data as training and testing data; and S40, prediction: inputting the data sequence matched in the step S30 into the LSTM neural network model, and predicting the 24-hour solar irradiance of the target prediction day. According to the prediction method, the LSTM neural network is utilized to track the volatility of irradiance, the training data is preprocessed, and the prediction precision is effectively improved on the basis of not adopting various coupling prediction model methods.

Description

technical field [0001] The invention belongs to the technical field of photovoltaic power generation, and more specifically relates to a short-term irradiance prediction method and a prediction system based on historical data analysis. Background technique [0002] As clean and mature forms of renewable energy power generation, photovoltaic power generation and hydropower generation have attracted more and more attention. According to statistics from the International Renewable Energy Agency, the total installed capacity of photovoltaics in the world increased by 580GW in 2019, and the total installed capacity of hydropower reached 1308GW. Accurate runoff prediction can provide important reference data for hydropower station dispatching and reservoir flood control. Due to the intermittent and fluctuating photovoltaic output power, it is more difficult to connect to the grid, and irradiance has the greatest impact on photovoltaic output power. Therefore, Accurate irradiance p...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06Q10/04G06Q50/06G06N3/04G06N3/08
CPCG06Q10/04G06Q50/06G06N3/08G06N3/048G06N3/044Y04S10/50
Inventor 周建郭苏黄文波李旭吴峰项华伟江薇许昌李大成田耘阿依努尔张艳青段兴林郑堃吴迪吕艳军
Owner HUANENG LANCANG RIVER HYDROPOWER
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap