基于改进简化粒子群算法的活性污泥过程建模

Modeling of activated sludge process based on improved simplified particle swarm algorithm

提出一种改进简化粒子群算法,并应用于某氧化沟工艺的活性污泥1号模型(activated sludge model No.1,ASM1)与活性污泥3号模型(activated sludge model No.3,ASM3),对模型进行参数优化,提高模型准确性,并比较模型校正后的准确度。改进简化粒子群算法在简化粒子群算法基础上,对决策变量进行单维度搜索,提高算法的全局搜索能力。利用典型测试函数与基本粒子群算法、简化粒子群算法以及帝国竞争算法进行对比,其结果表明,该算法在函数优化上具有较明显优势。将改进简化粒子群算法应用到ASM1和ASM3模型中,其结果表明,该算法在模型参数优化上具有良好效果,模型对比上得到较清晰结果。

An improved simplified particle swarm optimization algorithm (ISPSO) was applied to improve the accuracy, optimize the parameters and analyze the feasibility of activated sludge model No.1 (ASM1) and activated sludge model No.3 (ASM3) for the Orbal process. Based on simplified particle swarm optimization algorithm (SPSO), single dimensional search was used to enhance the global searching ability. ISPSO algorithm was compared with imperial competition algorithm (ICA), particle swarm optimization algorithm (PSO), SPSO algorithm and ISPSO algorithm with 7 typical test functions. Results demonstrate that ISPSO has better performance. ASM1 and ASM3 model were applied to ISPSO algorithm. Application shows good results in model optimization and a distinct conclusion in model comparison is obtained.