基于粒子群算法优化的BP神经网络在海水水质评价中的应用

针对目前存在的海水水质受多因素影响、评价难的现状，提出了一种基于粒子群算法（PSO）优化误差反向传播（BP）神经网络的海水水质评价模型。该模型通过PSO得到BP神经网络最优的权值和阈值，结合青岛东部海域10个监测站点的数据得到水质评价结果。实验证明，该模型和单因子评价、传统的BP神经网络评价相比较，具有训练时间短、预测精度高的特点，在海水水质评价中具有良好的应用价值。

Application of BP neural network based on particle swarm optimization in seawater quality assessment

To address the current situation in which seawater quality is affected by many factors and is difficult to evaluate, a seawater quality assessment model based on particle swarm optimization (PSO) optimized-error backpropagation (BP) neural network is proposed. The model uses the optimal weight and threshold of a BP neural network through PSO to obtain water quality evaluation results based on data from 10 monitoring stations in the eastern sea area of Qingdao. Experiments show that the model has a shorter training time and higher prediction accuracy compared with single-factor evaluation and traditional BP neural network evaluation. Overall, the proposed model has good application value in seawater quality assessment.