Photovoltaic power generation prediction method based on kmeans clustering

Abstract

The invention belongs to the technical field of photovoltaic power generation, and in particular relates to a method for predicting photovoltaic power generation based on KMeans clustering. It uses the kernel density function to fit the probability density function estimation of each type of data, so as to give the distribution law in the statistical sense of power generation data. It includes the following steps: Step 1. Obtain power generation data from the photovoltaic power plant and clean the data; Step 2. Perform hypothesis testing on the sample data, obtain the distribution law of power generation data through hypothesis testing, and divide the data into two types of Beta distribution and Weibull distribution Step 3, use kernel density function to fit Beta distribution, obtain the shape parameter a, b of Beta distribution; Step 4, carry out KMeans clustering analysis to shape parameter a, obtain clustering result; Step 5, according to the clustering result of step 5 Class results Carry out kernel density function fitting to the Beta distribution of each class, and obtain the confidence interval of each class of Beta distribution; step 6, predict the power generation output.

Description

technical field [0001] The invention belongs to the technical field of photovoltaic power generation, and in particular relates to a method for predicting photovoltaic power generation based on KMeans clustering. Background technique [0002] The over-exploitation of fossil energy has led to the rapid depletion of the earth's energy, which has led to a global power market reform. Renewable energy power generation technology has become a research hotspot in the power system. Incorporating new energy power generation units into the power grid must fully consider the impact of its power generation uncertainty and discontinuity on the power grid, such as causing voltage deviation, voltage fluctuation and flicker, harmonic distortion, three-phase unbalance and frequency fluctuation, etc. The problem brings intermittent and random fluctuations to the power grid. Since the power quality is an important factor affecting the safe and stable operation of the power system, it is neces...

AI Technical Summary

Problems solved by technology

Because the mapping relationship that the photovoltaic power prediction model needs to fit is significantly different under different weather conditions, and the weather forecast data is usually biased, so it is very difficult for the photovoltaic prediction model established based on the weather forecast data to accurately predict the power generation.

In the actual power grid, affected by weather changes and seasonal changes, the output power of photovoltaic power sources changes randomly. The existing photovoltaic power generation forecasting and analysis methods mainly classify existing power generation data according to weather data, so as to calculate the corresponding weather conditions. The output power of the photovoltaic power generation data has failed to dig out the law of the power generation data itself, and the distribution type and statistical significance of the photovoltaic power generation data have not been given.

Images