Application of Fuzzy Predictive Control for Smoke Sensing Alarm in Automatic Train Driving Control

The efficacy of traditional smoke detection methods is often constrained by the use of fixed threshold judgments, which impedes the ability of these methods to accurately and timely detect smoke, thereby increasing the risk of train operation. The objective of this study is to enhance the precision and reliability of automated driving control, thereby enabling vehicles to adapt to diverse and challenging environments more effectively. Based on fuzzy predictive control and granular computing, the algorithm is improved to solve the problem of rule explosion. Combining the annealing algorithm and particle swarm optimization algorithm to optimize the system parameters, a new fuzzy predictive control system for smoke sensing alarms is designed. The experimental results showed that the proposed algorithm on the Rosenbrook algorithm was concentrated between 45 and 51, demonstrating good stability and GSC, and having an excellent ability to jump out of local optima. Furthermore, the model demonstrated a rapid and stable convergence of the prediction accuracy curve, with an average prediction accuracy of 95.1% and a final value of 97.8%. The proposed smoke sensing alarm fuzzy predictive control has the advantages of being less prone to local optima, high accuracy, short time consumption, and fast response in automatic driving control. It provides new solutions and methods for other similar problems in the field of intelligent transportation.