CAJ | 학술논문

尾矿坝的变形监测是金属矿山企业生产管理极其重要的环节，针对目前尾矿坝变形预测模型存在不足的现状，论文采用了基因表达式编程（GEP）算法，以Eclipse为开发工具，通过选择函数集和终止符集、种群初始化、染色体解码、适应度评估、遗传操作等过程，建立了基于GEP-Deep Exca-vation的尾矿坝变形预测模型，并对某金属矿山尾矿坝监测点位移数据进行了预测；经与灰色GM （1，1）和BP神经网络2种模型试验对比分析，证实了基于GEP的尾矿坝变形预测模型的可行性和有效性，从而为金属矿山尾矿坝的变形预测提供了一种新方法。
미광패적변형감측시금속광산기업생산관리겁기중요적배절，침대목전미광패변형예측모형존재불족적현상，논문채용료기인표체식편정（GEP）산법，이Eclipse위개발공구，통과선택함수집화종지부집、충군초시화、염색체해마、괄응도평고、유전조작등과정，건립료기우GEP-Deep Exca-vation적미광패변형예측모형，병대모금속광산미광패감측점위이수거진행료예측；경여회색GM （1，1）화BP신경망락2충모형시험대비분석，증실료기우GEP적미광패변형예측모형적가행성화유효성，종이위금속광산미광패적변형예측제공료일충신방법。
Deformation monitoring of tailings dam is a very important part for metallic mine enterprises in pro-duction management. In view of the existing defects of tailings dam deformation prediction model, this paper has established the prediction model of tailings dam deformation based on GEP-Deep Excavation, and made a prediction for observation displacement in a certain metal mine tailings dam by GEP (Gene Expression Programming) algorithm with Eclipse as a development tool, through the process of selecting a set of functions, terminating character sets, population initialization, the chromosome decoding, fitness evaluation and genetic operation. By contrastive analysis of the gray GM(1,1) and BP neural network, empirical studies confirm the feasibility and effectiveness of the prediction model of tailings dam deformation based on GEP. This paper provides a new method for tailings dam deformation prediction of metal mine.