Photovoltaic power station super-short-time power prediction correcting method

Abstract

The invention discloses a photovoltaic power station super-short-time power prediction correcting method, which comprises the steps of: step 1, establishing four kinds of day type sets for photovoltaic power station historical data on a monthly basis; step 2, selecting the photovoltaic power station historical data of 6 workdays every week from each data set; step 3, calculating a temperature change value delta TN, a humidity change value delta SN, an irradiance change value delta FN between two continuous acquisition points on each workday, and a difference value delta PN between photovoltaic power station actual measured power and photovoltaic power station predicted power at the current acquisition point; step 4, calculating a temperature influence coefficient KtN, a humidity influence coefficient KsN, an irradiance influence coefficient KfN of two continuous acquisition points in each data set monthly; step 5, and correcting the photovoltaic power station predicted power. The photovoltaic power station super-short-time power prediction correcting method introduces meteorological parameters such as the temperature change value, the irradiance change value and the humidity change value to measure real-time weather variations, and corrects a photovoltaic power station predicted power value of an original power prediction model.

Description

technical field [0001] The invention relates to a method for power prediction and correction of a photovoltaic power station. Background technique [0002] Large-scale photovoltaic grid-connected power generation is an effective way to make full use of solar energy. However, the large-scale connection of photovoltaic power generation systems to the power grid will bring many problems to the reliable and stable operation of the power grid, increase difficulties in power grid dispatching, and affect the traditional start-up and dispatching methods of the power grid. The power prediction of the photovoltaic power generation system plays an important role in the investment planning and optimal design of the photovoltaic power generation system. Especially the ultra-short-term power forecasting is helpful to judge the change trend of the output power of the photovoltaic power generation system, thereby reducing the impact of the fluctuation of the output power of the photovoltai...

AI Technical Summary

Problems solved by technology

The problem is that the daily models used in the above power prediction methods are all simple daily models. For example, the daily models are simply divided into several types, and there is no analysis of the influence of various effective meteorological factors on the power generation of photovoltaic power generation systems. Comprehensive investigation, the effectiveness of the established power prediction model under various meteorological conditions has yet to be verified

In particular, the predicted power of photovoltaic power plants obtained by these methods is not enough for the real-time change information of various meteorological elements on the forecast day. For example, it is difficult to achieve ultra-short-term accurate forecasts for sudden weather such as cloudy and rain

In addition, the power grid dispatching department requires a time resolution of 15 minutes for power forecasting, and the above power forecasting models and forecast time scales cannot meet the actual requirements of power grid dispatching

Images