Interval Type-2 Fuzzy Neural Network Control for Furnace Temperature in Municipal Solid Waste Incineration Processes

Furnace temperature plays a pivotal role in ensuring the stability of municipal solid waste incineration (MSWI), influenced by variable factors such as material composition, feeding patterns, and equipment operation. We introduce an interval type-2 fuzzy neural network (IT2FNN) control method. Initially, we analyze furnace temperature control characteristics to identity crucial operating variables. Subsequently, we construct a multiple-input single-output (MISO) model for furnace temperature using the linear regression decision tree (LRDT) algorithm. Finally, we develop an IT2FNN-based controller for furnace temperature. The effectiveness of our proposed method is confirmed through control experiments using actual MSWI process data.

I. Introduction

Economic advancement and urbanization have markedly increased the production of municipal solid waste (MSW) [1],[2]. This increase presents significant challenges for environmental management and sustainable development [3]. Among waste treatment technologies, MSW incineration (MSWI) is notable for its waste reduction and recycling capabilities, offering a pollution-free solution [4]. Yet, in China, MSWI operations frequently depend on manual control, leading to variability in performance and difficulties in maintaining stable operations over the long term.