Wind and photovoltaic generation power prediction system with multiple prediction modes

Abstract

The invention discloses a wind and photovoltaic generation power prediction system with multiple prediction modes. The system is characterized by comprising a numerical value weather prediction data acquisition module for inputting numerical value weather forecast data serving as basic data of a prediction algorithm, a prediction mode selection module used for selecting a proper prediction algorithm to predict the power of a wind power plant or a photovoltaic power plant, a prediction algorithm module used for predicting the power of the wind power plant or the photovoltaic power plant and comprising a physical model, an adaptive logic network algorithm model and a BP neural network model, and a prediction result storage module for storing result data predicted by the prediction algorithm module. According to the method, the numerical value weather forecast data is used as input data of the prediction algorithm, the most proper prediction method is selected through the mode selection module, and a power prediction result is obtained.

Description

Technical field [0001] The invention relates to wind power prediction, photovoltaic power prediction, and microgrid power prediction. The invention particularly relates to a multi-prediction mode wind power photovoltaic integrated power generation power prediction system, which belongs to the technical field of wind power photovoltaic power generation. Background technique [0002] The wind power photovoltaic integrated power generation prediction system analyzes the influence of wind speed, wind direction, temperature, humidity and other factors on the output power of wind farms, as well as the influence of solar radiation, temperature, cloud cover, wind speed and other factors on the output power of photovoltaic power plants, combined with numerical weather The forecast data is analyzed and calculated; according to the actual operating conditions and topographic characteristics, a wind power prediction model is established for each wind farm, and a photovoltaic power prediction ...

AI Technical Summary

Problems solved by technology

Therefore, within a period of time after being put into operation, power prediction cannot be realized through artificial intelligence algorithms

[0004] However, in actual operation, it is found that various artificial intelligence algorithms have certain limitations. For example, the adaptive logic network ALN has high requirements for the integrity of historical data, and few wind farms or photovoltaic power stations in my country have Complete historical data; the disadvantage of BP neural network is that the learning rate is fixed, so the convergence speed of the network is slow and requires a long training time

For some complex problems, the training time required by the BP neural network algorithm may be very long

Images