基于极限学习机的采煤机记忆截割调高控制算法研究

发布时间：2018-10-31 08:28

【摘要】：采煤机的自动调高是实现采煤机工作面少人化和无人化的关键技术之一。采煤机自动调高控制对提升采煤系统可靠性、延长设备寿命、提高生产效率以及保障生产安全具有重要的现实意义。本文对基于极限学习机的采煤机记忆截割调高控制算法进行研究,完成的主要工作如下:通过对采煤机组成结构和工作原理分析,结合采煤机记忆截割系统研究,进而确定了记忆截割的采集参数和采样周期;考虑到现有的采煤工艺,建立了人工示范刀轨迹规划模型,并分析粒子群算法和遗传算法优缺点,利用两者结合的混合优化算法对示范刀轨迹进行规划,并对规划路径进行误差分析,结果表明混合优化算法能够满足采煤机的工作需求;对传统神经网络算法及一种新型的单隐含层前馈神经网络算法——极限学习机进行了深入研究,表明极限学习机算法在运算速度、计算精度、泛化能力上更具优势;通过对采煤机液压伺服调高系统的分析,针对采煤机调高系统的非线性、延时性等特性,进而设计了基于极限学习机PID控制策略,并利用Matlab/Simulink软件对截割轨迹进行跟踪控制,给出误差曲线。通过对仿真结果和跟踪精度分析,表明极限学习机PID控制能够适用于非线性、延时的采煤机调高系统,能够有效提高跟踪轨迹的精度,对实现采煤无人工作面具有深远影响。
[Abstract]:Automatic height adjustment of shearer is one of the key technologies to realize less humanization and dehumanization of shearer working face. The automatic height control of shearer has important practical significance in improving the reliability of coal mining system, prolonging the life of equipment, improving production efficiency and ensuring production safety. The main work of this paper is as follows: by analyzing the structure and working principle of shearer, combining with the memory cutting system of shearer, the paper studies the control algorithm of shearer's memory cutting height based on extreme learning machine, and the main work is as follows: by analyzing the structure and working principle of shearer, combining with the research of shearer's memory cutting system, Then, the acquisition parameters and sampling period of memory cutting are determined. Considering the existing coal mining technology, the manual demonstration tool path planning model is established, and the advantages and disadvantages of particle swarm optimization algorithm and genetic algorithm are analyzed. The error analysis of the planning path shows that the hybrid optimization algorithm can meet the needs of the shearer. The traditional neural network algorithm and a new single hidden layer feedforward neural network algorithm, the ultimate learning machine, are studied in detail. The results show that the ultimate learning machine algorithm has more advantages in computing speed, calculation accuracy and generalization ability. Through the analysis of the hydraulic servo height adjusting system of the shearer, aiming at the nonlinear and delay characteristics of the height adjusting system of the shearer, this paper designs the PID control strategy based on the ultimate learning machine, and uses Matlab/Simulink software to track and control the cutting path. The error curve is given. Through the analysis of simulation results and tracking accuracy, it is shown that the PID control of the ultimate learning machine can be applied to the nonlinear and delayed height adjustment system of the shearer, which can effectively improve the precision of tracking the track, and has a profound influence on the realization of the unattended coal mining face.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD632.1;TP18

【相似文献】