摘要
在盾构掘进系统中,因为盾构刀具磨损速率直接关系到施工工期的长短,而施工工期是评价工程项目综合经济效益最重要的指标。盾构刀具磨损速率的误差过大会导致施工企业对施工周期的错误预估,加大施工企业的经营成本,严重时会导致整个盾构系统故障,因此,盾构刀具磨损速率是盾构掘进过程中需要监测的重要参数之一。针对盾构掘进工艺进行研究,利用盾构掘进系统中的围岩强度、耐磨性、刀盘转矩、刀盘转速等主要运行指标作为输入变量,盾构刀具磨损速率作为输出变量分别建立模拟退火优化的支持向量机模型(SA-SVM)以及遗传优化的最小二乘支持向量机模型(GALS-SVM)。用建立的两种人工智能模型对神华神东补连塔煤矿斜井盾构刀具磨损速率进行预测,通过对比刀具磨损速率预测值与实际值来验证人工智能预测模型的可行性与精确度。
In the shield driving system,because the shield tool wear rate is directly related to the length of the construction period, and the construction period is the most important index to evaluate the comprehensive economic benefits of the project, shield cutter wear rate of error over the general assembly in construction enterprises to the construction cycle error estimates, increase the operating costs of construction enterprises, serious when can cause the failure of the entire shield, so the shield cutter wear rate is one of the important parameters of shield construction process need to be monitored. The shield tunneling technology research, using shield boring system of rock strength, wear resistance, knife wheel torque, rotation speed of the cutter head, the main operating indicators as input variables and shield cutter wear rate as the output variables are established respectively simulated annealing optimization support vector machine model SA-SVM and genetic optimization of least squares support vector machine model (GALS-SVM). Established two kinds of artificial intelligence model of Shenhua Shendong fill even tower coal mine inclined shield cutter wear rate prediction,by contrast tool wear rate prediction value and actual value to verify the artificial intelligence feasibility and the accuracy of the model prediction.
出处
《施工技术》
CAS
北大核心
2016年第22期25-30,共6页
Construction Technology
基金
国家科技支撑计划项目(2013BAB10B02)
关键词
盾构
刀具磨损速率
人工智能模型
预测
shields
tool wear rate
artificial intelligence model
prediction