Neural network modeling of the annoyance perception of cabin noise in passenger cars with hybrid algorithm optimization

Addressing the issue of low accuracy in traditional quantitative models for the perception of annoyance due to noise in vehicles, an evaluation method is proposed using a neural network model optimized by a hybrid algorithm to predict the perceived annoyance of interior noise. This hybrid algorithm integrates the sparrow search algorithm (SSA) and genetic algorithm (GA) to optimize the back propagation (BP) neural network. Based on the subjective and objective sound quality evaluation data, an objective quantification model of interior noise annoyance using the SSA-GA-BP network is established and compared with BP, GA-BP, and SSA-BP models. The results show that the SSA-GA-BP model achieves higher prediction accuracy, closer to subjective evaluation figures, and stronger generalization capabilities, making it a viable alternative to traditional subjective experiments in sound quality evaluation.