Short-period power combination prediction method for variable-weight-coefficient grid-connected photovoltaic power station

Abstract

The invention discloses a short-period power combination prediction method for a variable-weight-coefficient grid-connected photovoltaic power station, and the method comprises the steps: building a plurality of single-prediction models through historical data which is the nearest to a prediction time period, and solving the fitting value of each single-prediction model to each prediction sample point; calculating the weight coefficient of each single-prediction model at each prediction sample point through a gray scale correlation analysis method; carrying out the training and fitting of the fitting value of each prediction sample point and the corresponding weight coefficient through each single-prediction model, and obtaining a BPNN network model; obtaining a single power prediction value of each single-prediction model through the prediction values of the latest meteorological elements in a prediction time period, calculating a time-varying weight coefficient through the BPNN network model, and finally calculating the weighted power prediction value in the prediction time period; carrying out the above steps in a looped manner, and continuously updating the power prediction value in the prediction time period. Compared with the prior art, the method achieves the real-time changing of the weight coefficient of a combination prediction model, and is high in prediction precision.

Description

technical field [0001] The invention relates to the technical field of new energy power generation and power systems, in particular to a short-term power combination prediction method for grid-connected photovoltaic power plants with variable weight coefficients. Background technique [0002] In recent years, the scale of solar energy development and utilization has rapidly expanded, technological progress and industrial upgrading have been accelerated, and costs have been significantly reduced. It has become an important field of global energy transformation. During the "Twelfth Five-Year Plan" period, my country's photovoltaic industry system has been continuously improved, with remarkable technological progress, and the scale of photovoltaic manufacturing and application ranks among the top in the world. Great progress has been made in research and development of solar thermal power generation technology and equipment manufacturing, and a commercial test power station has...

AI Technical Summary

Problems solved by technology

[0004] At present, in the research on the meteorological factors affecting the output of photovoltaic power plants, due to the complex and changeable meteorological factors affecting the output of photovoltaic power plants, the actual weather changes are also much more complicated, and it is difficult to conduct a comprehensive analysis of the meteorological factors related to photovoltaic output. Most of the common photovoltaic prediction models consider a small number of meteorological factors, such as solar irradiance and air temperature as input variables, and rarely consider the introduction of relevant environmental factors such as atmospheric clarity index, sunshine hours, and temperature difference.

Even if all the meteorological factors mentioned above are taken into account when modeling, it will also increase the complexity of the forecast model. Moreover, there are multi-collinear relationships among the meteorological factors, which is not conducive to the establishment of the forecast model. The selected meteorological environment elements and how to reduce the multicollinear relationship between meteorological factors are particularly important for improving the prediction accuracy of the model

[0005] In addition to the meteorological factors considered above, the selection of the model algorithm is also important for the prediction accuracy when building the model. In the past, the power prediction of photovoltaic power stations was mostly based on a single prediction model, and for a single prediction model, it is necessary to improve its prediction. Accuracy is more difficult, and the accuracy and applicability of each single prediction model are different, and there is uncertainty in the prediction

[0006] The current common combination forecast is generally a combination of the forecast results obtained by multiple single forecast models, but the forecast results of different single models must be different, which requires the output results of each single forecast model Determine a weight, and the prediction model with equal weight has the same weight of the prediction result of a single prediction model, which will cause a large deviation in the prediction result of a single prediction model, which will affect the prediction result of the combined prediction model. Select the main weight of the appropriate weight The purpose is to eliminate the large bias that may exist in a single forecasting method and improve the accuracy of forecasting

[0007] In view of the above reasons, traditional photovoltaic power prediction methods are difficult to meet the practical, simple and predictive requirements of photovoltaic power plants and power systems.

Images