Furnace temperature control based on interval type-2 fuzzy broad learning system for municipal solid waste incineration process

To address the critical challenges in furnace temperature (FT) control during municipal solid waste incineration (MSWI), such as control uncertainties, dynamic time-varying characteristics, frequent adjustments, equipment wear, pollution emissions, and high energy consumption, we propose a new adaptive FT control strategy based on artificial intelligence (AI). This research is significant as it aims to enhance control efficiency and adaptability in complex operational environments. An interval type-2 fuzzy broad learning system (IT2FBLS) controller is developed based on an analysis of FT control characteristics. A dynamic-static switching event-triggering mechanism (DSSETM) is designed to adaptively switch between different event-triggering mechanisms, thereby reducing the update frequency of the manipulated variable (MV). Additionally, a multi-type eventtriggering mechanism (METM) is introduced to dynamically update both the controller structures and control laws, further enhancing performance. Stability analysis demonstrates the robustness of the proposed control strategy across various operational stages. The proposed AI approach is implemented through experiments using real data from municipal solid waste incineration (MSWI) plants and a hardware-in-the-loop platform (HILP). This work contributes to the application of AI in optimizing FT control within MSWI processes, aiming to effectively address uncertainties in the FT control process and improve control performance while reduce mechanical wear and energy loss, demonstrating both its effectiveness and practical applicability.