The municipal solid waste incineration process is a complex nonlinear dynamic system and the modeling and control of key variable furnace temperature is an urgent problem to be solved. For this problem, the furnace temperature model is established based on the actual data and its control problem is studied. Firstly, the operating variable with the high correlation to the furnace temperature is selected as the model input variable based on the actual operating conditions and expert knowledge. Secondly, the data-driven furnace temperature model is established by Tkagi-Sugeno (TS) fuzzy neural network. Finally, the furnace temperature tracking error and error change rate are taken as the input of radial basis function (RBF) neural network PID controller (RBF-PID controller), and the primary air flow is taken as the output in order to realize the accurate control of furnace temperature. The experimental results show that the furnace temperature model based on TS fuzzy neural network can fit the actual value well, the RBF-PID controller can achieve real-time and accurate control of furnace temperature with high control accuracy.