基于多目标粒子群算法的城市固废焚烧过程智能操作优化  

Intelligent Operational Optimization of Municipal Solid Waste Incineration Process Based on Multi-objective Particle Swarm Algorithm

作　　者：蒙西 侯启正 乔俊飞 MENG Xi;HOU Qi-Zheng;QIAO Jun-Fei(Faculty of Information Technology,Beijing University of Technology,Beijing 100124;Beijing Laboratory of Smart Environmental Protection,Beijing 100124;Engineering Research Center of Intelligent Perception and Autonomous Control,Ministry of Education,Beijing 100124)

机构地区：[1]北京工业大学信息学部,北京100124 [2]智慧环保北京实验室,北京100124 [3]智能感知与自主控制教育部工程研究中心,北京100124

出　　处：《自动化学报》2024年第12期2462-2473,共12页Acta Automatica Sinica

基　　金：国家自然科学基金(62273013,62021003);北京市科技新星计划(20230484310);科技创新2030“新一代人工智能”重大项目(2021ZD0112301)资助。

摘　　要：城市固废焚烧(Municipal solid waste incineration,MSWI)技术因兼具减量化、无害化、资源化等特点,已成为治理固废污染的主要方式.由于城市固废成分复杂,含水率、热值动态波动,固废燃烧、余热利用、烟气净化等环节耦合冲突,实际工业过程难以高效运行.为此,本文提出了一种基于多目标粒子群算法的城市固废焚烧过程智能操作优化方法,以期实现燃烧效率和烟气净化效率的协同优化.首先,设计自组织径向基函数(Self-organizing radial basis function,SORBF)神经网络建立运行指标模型,实现城市固废焚烧过程运行性能的在线评价;其次,引入区域拥挤度指标提出了一种改进的多目标粒子群优化算法,以获取操作变量的Pareto解集;然后,利用熵权法确定操作变量最佳设定值,实现城市固废焚烧过程高效运行;最后,通过北京某城市固废焚烧厂的实际运行数据对所提方法进行验证,实验结果表明基于多目标粒子群算法的智能操作优化方法可以实现燃烧效率与脱硝效率的协同提升.Municipal solid waste incineration(MSWI)technology has become the main way to address solid waste pollution due to its characteristics of reduction,harmlessness,and resource utilization.However,it is difficult for actual industries to operate efficiently due to the complex composition of municipal solid waste,dynamic fluctuations in moisture content and calorific value,coupling conflicts in solid waste combustion,waste heat utilization and flue gas purification.To enhance combustion efficiency and flue gas purification efficiency,this paper proposes an intelligent operational optimization method of MSWI process based on multi-objective particle swarm algorithm.First,operational index models are established by designing self-organizing radial basis function(SORBF)neural networks to achieve online evaluation of operational performance in MSWI process.Second,an improved multi-objective particle swarm optimization algorithm is presented by incorporating regional congestion degree index to obtain the Pareto solutions of operating variables.Then,the entropy weight method is employed to determine the optimal set value of operating variables,achieving efficient operation of MSWI process.Finally,the proposed method is verified through actual operational data from a MSWI plant in Beijing,and the experimental results demonstrate that the intelligent operational optimization method based on multi-objective particle swarm algorithm can improve combustion efficiency and reduce nitrogen oxide emissions.

关 键 词：城市固废焚烧过程 燃烧效率 NOX排放 运行指标模型 多目标粒子群算法 

分 类 号：X705[环境科学与工程—环境工程] TP18[自动化与计算机技术—控制理论与控制工程]