Robust Output Feedback Position Control of Hydraulic Support with Neural Network Compensator

Hydraulic support is important equipment in the fully mechanized mining face, and control performance of hydraulic multi-cylinder system directly affects smooth progress coal process, which basis for continuous advancement face. However, friction force process pulling frame complex due to underground environmental load. Moreover, parameters moving cylinder are uncertain, state cannot be measured, increases difficulty control. A proportional-integral-derivative controller usually used electro-hydraulic closed-loop systems because its computational complexity, but robustness poorly adapted variable load conditions mine. Therefore, a robust output feedback position proposed this paper improve accuracy with only signal. The modeled as standard type, then high-order differentiator estimate immeasurable states using neural network compensator applied compensate external disturbance cylinder. Furthermore, ZY3200/08/18D adopted analyze effectiveness designed simulations. Finally, real-time built, experimental results show that novel has improved by 47.2% 30.6% tracking compared PI controller.