改进粒子群优化BP神经网络的心理压力识别算法

为提高心理压力的识别率,提出一种改进的粒子群优化BP(back propagation)神经网络的压力识别算法。该算法在基本粒子群(particle swarm optimization,PSO)模型的基础上,引入了收缩因子,在收缩因子的作用下,使速度的边界限制消失,选取适当的参数来保证PSO算法的有界和收敛特性,实现对BP神经网络的优化。利用心算任务进行压力诱发,采集高压、低压状态下的心电信号,提取了与心理压力相关的心率变异性特征值,并对特征数据对比分析;建立了心理压力程度的分类模型,通过改进的PSO模型优化BP神经网络以识别心理压力。结果表明:改进的粒子群优化BP神经网络算法与BP神经网络相比收敛速度快、误差小且识别率高,该算法对心理压力的识别率可达94.83%,识别效果优于未优化的BP神经网络算法。

An Improved Particle Swarm Optimization Bp Neural Network Algorithm for Psychological Stress Identification

In order to improve the recognition rate of psychological pressure, an improved pressure recognition algorithm based on particle swarm optimization (BP) neural network is proposed. On the basis of the basic particle swarm optimization (PSO) model, the shrinkage factor is introduced. Under the action of the shrinkage factor, the boundary limit of the velocity disappears, and the appropriate parameters are selected to ensure the boundary and convergence characteristics of the PSO algorithm. The optimization of BP neural network is realized. In this paper, the ECG signals under high pressure and low voltage are collected by using mental arithmetic task to induce pressure, and the eigenvalues of heart rate variability related to psychological pressure are extracted, and the characteristic data are compared and analyzed, and the classification of psychological pressure degree is established. The improved PSO model is used to optimize the BP neural network to identify psychological stress. The results show that compared with BP neural network, the improved particle swarm optimization BP neural network algorithm has faster convergence speed, smaller error and higher recognition rate. The recognition rate of the improved particle swarm optimization algorithm for psychological pressure can reach 94.83%. The recognition effect is better than the unoptimized BP neural network algorithm.