白云鄂博西矿台阶爆破参数优化研究

发布时间：2018-05-11 09:52

本文选题：爆破 + 台阶　；参考：《内蒙古科技大学》2015年硕士论文

【摘要】：在白云鄂博西矿生产过程中，台阶爆破是一个非常重要的生产环节，爆破效果影响着生产过程中的采装、运输等后续工序的效率和总的经济效益。不同爆区之间的岩石性质的差异使原有爆破参数与分区后的爆区岩性不相匹配，结果导致爆破效果发生变化。大块率、炸药单耗偏高，残留根底，直接影响着铲装、运输、破碎等后续工序和采矿总成本。在实际施工中，当地质条件和环境发生变化，爆破效果也会发生相应的改变，实践表明，，工程爆破效果与很多因素有关，比如矿岩岩性、炸药性能以及爆破参数等，且这些因素之间存在着一定的客观规律性。爆破工作所要解决的重要课题就是了解、描述这种客观的规律性，并在此基础上形成较为成熟的爆破经验并使之得以积累。本文根据各岩石的物理力学性质，通过分析各岩石可爆性将白云鄂博西矿台阶分为易爆区、中等难度爆破区和难爆区三个区域进行分区爆破。充分利用人工神经网络具有的自学习、自适应、自组织和非线性动力学特性等特点，建立了面向MATLAB语言的爆破参数BP神经网络预测模型，对爆破参数进行优化。对白云鄂博西矿爆破现场实验数据进行统计、分析、研究，利用BP神经网络对爆破参数进行合理优化，得出适合白云鄂博西矿的爆破参数。易爆区域孔网参数约为117m、最小抵抗线为6.5m、炸药单耗降至0.548kg/m3；中等爆破难度区域孔网参数约为107m、最小抵抗线为6.0m、炸药单耗降0.578kg/m3；难爆区域孔网参数约为106.5m、最小抵抗线为5.5m、炸药单耗降0.657kg/m3。爆破后大块率控制在2%以内、基本无根底出现，有效的解决了之前矿山大块率高、根底较多的问题，满足矿山对精细化爆破的要求，爆破效果明显提高。通过分析和研究白云鄂博西矿爆破参数，提出利用BP神经网络来建立符合实际情况的爆破预测模型并进行训练，经检验，模型的建立合理、精度符合要求，指导矿山的爆破生产效果良好，降低了采矿成本，提高了该矿的经济效益。
[Abstract]:Bench blasting is a very important production link in the production process of Bayan Obo West Mine. The blasting effect affects the efficiency and overall economic benefit of the subsequent processes such as mining and loading, transportation and so on. The difference of rock properties between different blasting areas makes the original blasting parameters do not match the lithology of the blasting area after the division, which results in the change of blasting effect. Bulk rate, high unit consumption of explosive, and residual base directly affect shovel loading, transportation, crushing and other follow-up processes and total mining costs. In the actual construction, when the geological conditions and environment change, the blasting effect will also change. The practice shows that the blasting effect is related to many factors, such as rock and ore properties, explosive properties and blasting parameters, etc. And there are some objective laws between these factors. The important task to be solved in blasting work is to understand and describe this objective regularity, and on this basis to form more mature blasting experience and make it accumulate. According to the physical and mechanical properties of each rock, the bench of Bayan Obo West Coal Mine is divided into three zones: explosive area, middle difficulty blasting area and difficult blasting area by analyzing the blasting property of each rock. Taking full advantage of the self-learning, self-adaptive, self-organizing and nonlinear dynamic characteristics of artificial neural network, a BP neural network prediction model of blasting parameters for MATLAB language is established, and the blasting parameters are optimized. The blasting field experiment data of Bayan Obo West Mine were analyzed and analyzed. The blasting parameters were optimized by BP neural network and the blasting parameters suitable for Bayan Obo West Mine were obtained. The parameters of hole net in explosive region are about 117 m, the minimum resistance line is 6.5 m, the unit consumption of explosive is decreased to 0.548 kg 路m ~ (3), the parameter of hole net is about 107 m, the minimum resistance line is 6.0 m, the unit consumption of explosive is 0.578kg / m ~ (3), the parameter of hole net is about 106.5 m, the minimum resistance line is 5.5 m, and the unit consumption of explosive is 0.657 kg 路m ~ (3). After blasting, the boulder rate is controlled within 2% and there is basically no root bottom, which effectively solves the problems of high boulder rate and more root and bottom before, and meets the requirements of fine blasting in mines, and the blasting effect is obviously improved. Based on the analysis and study of blasting parameters in Bayan Obo West Mine, a BP neural network is put forward to establish and train the blasting prediction model in accordance with the actual situation. After testing, the establishment of the model is reasonable and the precision meets the requirements. The blasting effect of the mine is good, the mining cost is reduced and the economic benefit of the mine is improved.
【学位授予单位】：内蒙古科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD235

【参考文献】