Forecast of air temperature based on BP neural network

The change of temperature is closely related to people's life, and the drastic change of the next day's temperature will affect people's normal life, so it is very important to accurately predict the next day's temperature. Information fusion technology is a process of automatic analysis and comprehensive processing of multi-source information in order to complete the required decision-making and evaluation tasks. BP neural network is one of the information fusion algorithms, which can predict the data collected by various sensors. Therefore, the data collected by Canberra sensor, such as maximum temperature, minimum temperature, rainfall and maximum wind speed, are processed, and the BP neural network is constructed to predict the maximum and minimum temperature of the next day. The experimental results show that this method can well predict the maximum and minimum temperature of the next day.