粒子群优化算法在源强反算问题中的应用研究

在危险化学品泄漏事故中泄漏源强是预测事故后果的主要影响参数,也是事故应急救援决策的基础。为了在化学品泄漏事故过程中快速准确地获取泄漏源强数据,将粒子群优化(PSO)算法应用于危险化学品泄漏源强的反算中。利用高斯烟羽扩散模型和下风向浓度测量数据,将计算浓度与测量浓度的误差平方和作为目标函数,采用粒子群算法来优化,以确定源强并通过模拟的测量浓度数据进行算法有效性验证。结果表明,PSO算法及其参数改进算法不依赖于初值的选择,计算速度快,能满足事故应急响应救援的需要。

PSO Algorithm for Back-calculation of Source Intensity

In the leakage accident of hazardous chemicals,the leakage source intensity is the main parameter for accident consequence prediction and the basis of emergency-rescue decision.In order to obtain source intensity information,the particle swarm optimization(PSO) algorithm was applied to the back-calculation of leakage source intensity.With Gaussian plume dispersion model and downwind measurement data,and by taking the error square sum of calculated concentration and measured concentration as the objective function,an optimization on the function was conducted by using PSO algorithm so as to determine the source intensity.The effectiveness of PSO algorithm is verified with the measurement data of concentration by simulation.The result indicates that the proposed algorithm can satisfy the needs of emergency rescue due to its independence of the selection of initial values and its rapid calculation speed.