TEMPERATURE CONTROL OF RUBBER COMPOSITES BY ADAPTIVE MULTI-DIMENSIONAL TAYLOR NETWORK DURING MICROWAVE HEATING PROCESS

Microwave technology has been widely used in rubber industry. In order to solve the problem of uneven temperature distribution, a novel control method of adaptive multi-dimensional Taylor network combined with Cuckoo Search is proposed in this paper. The adaptive multi-dimensional Taylor network control method is used to obtain the suitable output powers and phase difference under unknown system parameters. Cuckoo Search algorithm is utilized to optimize the whole situation and find the best fit input variables at sampling points. To verify the proposed control strategy, the dielectric permittivity of nitrile butadiene rubber composites is measured, and the control process is simulated based on measured values. The simulation results show that the proposed method can well control the temperature rising process with little difference between the average temperature and reference trajectory.