Parameters of a proportion-integral-derivative（PID） controller were adjusted online with fuzzy inference and a particle swarm optimization（PSO） algorithm were applied to the fuzzy PID controller to adjust the inertia weight coefficient and optimize quantitative factors and controller parameters. Then the fuzzy PID controller, optimized by the PSO algorithm, was applied to the barrel temperature control of the extruders for wood-plastic composites（WPC）. Results showed that the optimized fuzzy PID controller could increase heating rate, reduce overshoot, shorten adjustment time, enhance anti-jamming capability and effectively improve control effects of the barrel temperature. Tensile strength and production efficiency of the WPC panels were improved significantly.