Multiple Input Mulitple Output Control Method Based on Single Neuron Adaptive PID for Municipal Solid Waste Incineration Process

The automatic combustion control (ACC) system of municipal solid waste incineration (MSWI) processes realizes the stable operation through the coupling control among multiple loops. However, due to the difference in municipal waste solid components and operating level, the introduced ACC system is difficult to be directly applied in China. The research on MSWI process controlling is still the key to solve its stable running. Aiming at the strong coupling and nonlinear characteristics of MSWI processes, this paper proposes a multiple input multiple output control method based on single neuron adaptive PID. Firstly, multi-input multi-output controlled object models in terms of furnace temperature (FT), boiler steam flow (BSF), and oxygen (OX) content are established, whose input and output are determined on the basis of MSWI process flow and ACC system’s mechanism. Then, based on the experience of experts in the field, the multi-loop controller of the MSWI process is established by using single neuron adaptive PID controller. Finally, the accuracy and effectiveness of the proposed framework are verified by actual industrial process data.

I. Introduction

The increasing amount of municipal solid waste (MSW) has received extensive attention. MSW incineration (MSWI) technology has become the main means of MSW treatment because of its advantages of harmlessness, reduction, and resource utilization[1],[2]. As a typical process industry, the MSWI process involves complex physical and chemical reactions, which has the characteristics of multi-input, multi-output, strong coupling, and nonlinearity. The safe and stable operation needs to comprehensively consider the coupling relationship between multiple controlled variables. However, due to the difference in waste components and operating level, the introduced ACC system is difficult to be directly applied in China, mostly MSWI industry relies on manual control by domain experts[3]. The long-term stable operation of the MSWI process is difficult to be guaranteed because of the differences in the experience of field experts and the subjectivity and randomness of manual operation. Therefore, the study of intelligent control of multiple controlled variables in the MSWI process is the key to ensure its stable operation.