WOA-GSK-ELMAN: An Intelligent Atmospheric Temperature Prediction Model

Atmospheric temperature is an essential guide for human beings to engage in production and life, so how to achieve a more accurately predict atmospheric temperature has become a hot research topic in recent years. Due to the complex factors affecting temperature transformation, it is impossible to accurately predict the future temperature change simply by using daily temperature data. Therefore, this paper uses 14 meteorological factors such as air humidity, light intensity, and carbon dioxide concentration to predict atmospheric temperature based on ELMAN. To improve the accuracy of ELMAN prediction, we use the WOA to optimize the weight and bias of ELMAN. To solve the problem of WOA which easily to fall into the local optimum. So we use the GSK to hybridize and propose a WOA variant of WOA-GSK. In the algorithm benchmark test, the proposed WOA-GSK achieved outstanding results; in the simulation test, the proposed WOA-GSK- ELMAN achieved much higher than WOA-ELMAN and ELMAN. In addition, to ensure the practicability of the algorithm in real life, this paper uses the self-made collected data set for temperature prediction. The R ² = 0. 9999 shows that the WOA-GSK-ELMAN proposed in this paper has high accuracy and practicability.