PSO-BP神经网络在船舶二级脱硝系统中的应用

随着国际海事组织（IMO）标准Tier III标准的实施，船舶柴油机尾气中的氮氧化物（NOx）的排放受到了更严格的限制。选择性催化还原脱硝技术因其高效、可靠等优势成为船舶柴油机NOx减排的主要选择。本文先根据SCR反应机理在MATLAB中搭建一级SCR脱硝系统和二级SCR脱硝系统模型，通过仿真结果对比，发现加入二级脱硝系统后，脱硝率可以得到提高。鉴于BP网络预测精度较低，本文将粒子群（PSO）算法加入到BP权值训练过程中，利用PSO-BP 神经网络预测 SCR-DeNOX系统出口处NOX浓度。结果表明：PSO-BP 神经网络不但可以预测SCR-DeNOX系统出口处的NOX浓度，而且与传统BP 神经网络相比可以提高预测结果的精度，为有效控制喷氨量、降低氨气逃逸量提供帮助。

Application of PSO-BP Neural Network in Ship's Secondary Denitration System

With the implementation of International Maritime Organization (IMO) standard Tier III, the emission of nitrogen oxides (NOx) from marine diesel engine exhaust has been more severely restricted. Selective catalytic reduction denitration technology has become the main choice for NOx reduction of marine diesel engines due to its advantages of high efficiency and reliability. In this paper, according to the SCR reaction mechanism, the models of one-stage SCR denitration system and two-stage SCR denitration systemare built in MATLAB. Through the comparison of simulation results, it is found that the denitration rate can be improved by adding the two-stage SCR denitration system. In view of the low prediction accuracy of BP network, particle swarm optimization (PSO) algorithm is added to the BP weight training process and PSO-BP neural network is used to predict NOX concentration at the outlet of SCR-DeNOX system. The results show that PSO-BP neural network can not only predict NOX concentration at the outlet of SCR-DeNOX system, but also improve the accuracy of prediction results compared with traditional BP neural network, which provides help for effectively controlling ammonia injection and reducing ammonia escape.