论文部分内容阅读
矿岩可爆性的准确判定对采矿设计、安全生产等具有重要意义.将马氏距离判别法引入到矿岩可爆性分级中,建立了矿岩可爆性等级分类的距离判别模型.选用岩石抗压强度、岩石容重、岩石完整性系数和炸药单耗4项指标作为判别因子,将35个矿岩可爆性实例作为学习样本进行训练,建立了相应的判别函数对待判样本进行分类,结果表明经过训练后的模型误判率为零.将判别模型应用于工程实例,判别结果也与BP神经网络方法相符,表明该模型具有良好的判别功能,可以在实际工程中进行推广应用.
The accurate determination of the ore blastability is of great significance for mining design and safety production, etc. The Mahalanobis distance discriminant method is introduced into the blastability classification of ore rock, and the distance discriminant model of the ore blastability grade classification is established. Rock compressive strength, rock bulk density, rock integrity coefficient and explosive unit consumption as the discriminant, 35 examples of blasting of ore rock are trained as learning samples, corresponding discriminant functions are established to classify the samples to be adjudged, The result shows that the misjudgment rate of the trained model is zero.The discriminant model is applied to the engineering example and the discriminant result is also consistent with the BP neural network method, which shows that the model has a good discriminating function and can be popularized and applied in practical engineering.