摘要
在研究支架支护位姿的基础上对支架支护位姿进行了运动学研究,认为支架支护异常位姿可分为俯视采场方向的异常支护状态和与煤壁垂直的异常支护状态。通过BP神经网络法实现了支护位姿的非线性拟合,同时通过支护位姿训练集对神经网络模型进行了训练。结果显示,支架支护位姿的非线性拟合模型可较为精确地实现运动学计算,拟合误差极小。
Based on the research of the support posture, this paper carried out kinematics research on the support posture of the support. It is concluded that the abnormal support posture of the support can be divided into the abnormal support state overloo-king the direction of the stope and the abnormal support state perpendicular to the coal wall. The BP neural network method was used to realize the nonlinear fitting of the support posture, and the neural network model was trained by the training set of the su-pport posture. The results showed that the nonlinear fitting model of support posture can be used to realize kinematics calculation accurately with minimal fitting error.
作者
杨鹏
YANG Peng(Xinju Coal Mining Machinery Equipment Manufacturing Co., Ltd., Huozhou Coal Electricity Group, Huozhou 031400, Shanxi, China)
出处
《能源与节能》
2019年第1期16-18,共3页
Energy and Energy Conservation
关键词
液压支架
神经网络
支护位姿
运动学
hydraulic support
neural network
support posture
kinematics