Enhanced combination modeling method for combustion efficiency in coal-fired boilers

► A combination modeling method composed of ELM, ANFIS and PS-ABC is proposed. ► Extreme learning machine (ELM) is used to build a global model. ► Adaptive neuro-fuzzy inference system (ANFIS) is applied to compensate output errors of ELM model. ► PS-ABC, a modified hybrid artificial bee colony algorithm, is adopted to optimize input weights and bias of ELM. ► The proposed method is applied to a 300 WM coal-fired boiler.