摘要:
液压支架推移系统的控制精度不足会导致综采工作面直线度产生偏差，严重制约开采效率，甚至埋下安全隐患。现有控制方法受非线性摩擦、动态扰动耦合及未建模动态特性影响，难以实现毫米级定位。针对此问题，首先，建立了融合液压缸动力学、流量特性及扰动耦合的非线性模型。其次，设计了扩张状态观测器，实时估计并补偿系统未建模动态与扰动总和。同时，改进了准滑动模态控制器，结合双幂次趋近律与新型非线性饱和函数，通过边界层内非线性反馈抑制抖振并加速收敛。最后，搭建了液压支架推移控制系统仿真模型，验证了本文所提方法的可行性和有效性。结果表明：相比传统滑模控制，所提方法在阶跃响应中稳态时间缩短47.6%，误差趋近于零；正弦跟踪峰值误差降低94.7%，频带特性更优；方波输入下实现平滑切换，鲁棒性更强。
Abstract:
The insufficient control accuracy of hydraulic support pushing systems can lead to deviations in the straightness of the fully-mechanized mining face, which severely restricts mining efficiency and even poses potential safety hazards. Existing control methods struggle to achieve millimeter-level positioning due to nonlinear friction, dynamic disturbance coupling, and unmodeled dynamic characteristics. To address this issue, this study first establishes a nonlinear model integrating hydraulic cylinder dynamics, flow characteristics, and disturbance coupling. Next, an extended state observer (ESO) is designed to estimate unmodeled dynamics and total disturbances in real time. Furthermore, a quasi-sliding mode controller is improved by combining a dual-power reaching law with a novel nonlinear saturation function, suppressing chattering and accelerating convergence through nonlinear feedback within the boundary layer. Finally, a simulation model of the hydraulic support pushing control system is developed to validate the feasibility and effectiveness of the proposed method. Results demonstrate that, compared to traditional sliding mode control, the proposed method reduces steady-state time by 47.6% in step response with near-zero error, decreases peak tracking error by 94.7% in sinusoidal tracking with superior frequency bandwidth characteristics, and achieves smooth switching under square-wave input, exhibiting enhanced robustness.
下载:
