Research on Floor Damage Classification of Optimized BP Neural Network Using Genetic Algorithm

Abstract

Floor water inrush accidents are a significant component of mine water disasters, posing a serious threat to coal mine safety and production. Floor water inrush accidents are a significant component of mine water disasters, posing a serious threat to coal mine safety and production. The damage classification serves as a crucial indicator of the severity of floor water inrush accidents. This study outlines the primary control factors influencing floor damage classification. Utilizing a genetic algorithm, the BP neural network is optimized for floor damage classification. Nonlinear equations for floor damage depth are derived for short mining faces with shallow burial depths, long mining faces with shallow burial depths, short mining faces with deep burial depths, and long mining faces with deep burial depths based on a burial depth of 500m and an inclined length of 120m. The application of these formulas aligns closely with on-site measurements.