摘要
刮板输送机的链条松紧状态是传动系统能否正常高效工作的前提。随着工作面煤层变化、采煤机位置变化和溜槽上物料变化,链条载荷受到多种因素的复合影响,链条受力状态具有随机性、复杂性、时变性和不确定性等特点,各种复杂变化特性造成了链条预紧力控制的困难。基于机器学习算法,将人工控制刮板输送机链条预紧力的历史数据作为训练集,建立了基于神经网络的刮板输送机链条预紧力动态控制算法模型。仿真结果表明,在经过大量数据训练后,模型能够较准确地实现链条预紧力控制,具有较好的适用性和先进性。
The tightness of the chain of the scraper conveyor is a prerequisite for the normal and efficient operation of the transmission system.With changes in the coal seam of the working face,the position of the shearer and the material on the chute,the chain load is affected by a combination of multiple factors.The force state of the chain has the characteristics of randomness,complexity,timevarying and uncertainty,and various complex changes make it difficult to control the chain pretensioning force.Based on the machine learning algorithm,took the historical data of manually controlled chain pretensioning force of scraper conveyor as the training set,and established the dynamic control algorithm model of chain pretensioning force of scraper conveyor based on neural network.The simulation results show that affter a large amount of data training,the model can accurately realize the chain pretensioning force control,and has good applicability and progressiveness.
作者
王天星
成敏
沈丰
曹伟
Wang Tianxing;Cheng Min;Shen Feng;Cao Wei(Scraper Machine Research Institute,Ningxia Tiandi Benniu Industrial Group Co.,Ltd.,Yinchuan 750000,China)
出处
《煤矿机械》
2024年第3期28-30,共3页
Coal Mine Machinery
基金
宁夏回族自治区重点研发项目(2021-TD-MS014)。
关键词
机器学习
神经元算法
刮板输送机
链条预紧力控制
machine learning
neuron algorithm
scraper conveyor
chain pretensioning force control
