煤矿井下掘进机电气设备的节能研究

发布时间：2018-02-23 19:41

本文关键词： 煤矿节能掘进机 BP神经网络预测　出处：《北京建筑大学》2017年硕士论文　论文类型：学位论文

【摘要】：煤炭是现代工业生产不可或缺的能源。在煤炭开采过程中,大量的能源消耗来自井下掘进电气设备。为了绿色开采和节能减排,应采取措施降低能耗。因此,研究煤矿井下掘进电气设备的节能有着十分重要的理论和实际意义。在煤矿掘进电气设备中,电力变压器、煤炭输送机、电动机和掘进机能耗较大。因此,本文主要对以上四种电气设备进行节能研究。通过分析电力变压器的谐波产生原因和能耗组成,提出了谐波治理方法和节能措施。根据所提出的节能方案,通过对综掘工作面负荷量的计算选择了变压器的类型。煤炭带式输送机的带宽、带速和输送带覆盖层厚度对电能的消耗有着较大的影响,依据煤炭运量和节能的要求,对带式输送机的参数和能耗之间的关系进行了分析,并给出了最后的设备选型。针对电动机的高能耗问题,根据变频调速原理对电动机进行节能控制。结果表明,采用变频调速方法能够有效地降低电动机产生的高耗能的问题。针对矿井掘进工作面中的高功率掘进机电压波动是导致能耗的主要原因,提出了一种改进的BP神经网络电压预测方法。利用均方误差构造一个可以自适应的学习步长,有效地解决了传统学习方法中步长选择的试凑问题。通过随机选择的归一化样本对提出的方法进行验证,仿真结果显示预测值和实际值之间的误差在±0.04之内,达到了预测电压和后续稳定输出的要求。
[Abstract]:Coal is an indispensable source of energy for modern industrial production. In the process of coal mining, a large amount of energy consumption comes from underground excavating electrical equipment. For the sake of green mining and energy saving and emission reduction, measures should be taken to reduce energy consumption. It is of great theoretical and practical significance to study the energy saving of electrical equipment for underground tunneling in coal mine. In the electrical equipment of coal mine tunneling, power transformer, coal conveyer, motor and roadheader consume more energy. Based on the analysis of harmonic generation and energy consumption composition of power transformer, the harmonic treatment method and energy saving measures are put forward. The type of transformer is selected by calculating the load of fully mechanized working face. The bandwidth, speed and thickness of belt cover of coal belt conveyor have great influence on the consumption of electric energy, according to the requirement of coal transportation and energy saving, The relationship between the parameters and energy consumption of belt conveyor is analyzed, and the final equipment selection is given. In view of the high energy consumption of the motor, the energy-saving control of the motor is carried out according to the principle of variable frequency speed regulation. The problem of high energy consumption produced by motor can be effectively reduced by using frequency conversion method. The main cause of energy consumption is voltage fluctuation of high power roadheader in mining face. An improved BP neural network voltage prediction method is proposed, which uses mean square error to construct an adaptive learning step. The problem of step size selection in traditional learning method is effectively solved. The simulation results show that the error between the predicted value and the actual value is within 卤0.04 by the random selection of normalized samples. The predicted voltage and subsequent stable output are achieved.
【学位授予单位】：北京建筑大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD632.2

【参考文献】