Research On Construction Simulation Of Underground Cavities Considering Transportation Machinery Failure

The construction of underground cavities is a complex system engineering.It is complicated and influenced by many factors,that poses a great challenge for construction management.System simulation is effective to analyze it,the application of which can effectively improve the construction management efficiency.However,the existing study can not scientifically and systematically consider the influence of factors such as geology when quantifying the transportation machinery failure impact,and it is subjective and error-prone.The empirical formula or sampling time distribution were used to calculate the duration of construction unit,which causes a low-precision simulation calculation of key construction processes.Combining the advantages of M5 P model tree and Support Vector Regression(SVR),a transportation machinery failure prediction method based on M5P-SVR was proposed.A new simulation model based on slag construction loop modeling was proposed.Therefore,the simulation study considering transportation machinery failure was carried out.The main research results are as follows:(1)There is a problem of low precision when simulating the slag construction in existing research.The simulation model based on slag construction loop modeling was established.The slag transportation construction is a cyclical process between the working surface and the destination.It’s important for controlling the schedule.Therefore,based on the research conducted by our research group,this study improved the slag transportation simulation module in the Cycle Operation Network(CYCLONE).According to transportation plan,a slag construction loop was established to calculate its duration.Then coupling the improved CYCLONE model and the Critical Path Method(CPM),a model was established.This simulation model could analyze the traffic situation.The simulation precision was also effectively improved.(2)The method for predicting the of transportation machinery failure in existing research are subjectivity and error-prone.Based on the above simulation model,a M5P-SVR predicting method was proposed.In existing research,the engineering experience method and fuzzy theory for determining the failure probability of transportation machinery cannot systematically and scientifically consider the influence of geology and human factors.And the prediction error is large.Therefore,a M5P-SVR transportation machinery fault prediction method was proposed.The M5 P algorithm is a decision tree model,so it can use several factors as input attributes for modeling and regression prediction.It has the advantages of simple and effective training rules.The SVR algorithm has good generalization ability when solving small sample and nonlinear regression prediction problems.It can effectively avoid local disasters and over-learning.Therefore,using SVR algorithm replace regression model in M5 P algorithm can scientifically and reasonably consider the influence of some factors.Its training rules is effetivebut and the prediction accuracy is high.Therefore,it quantified the impact of transportation machinery failure effectively.(3)Taking the underground cavities as an example,the above-mentioned model was used for construction simulation analysis,and the accuracy and superiority of the model were verified by comparative analysis.Taking the underground cavities as an example,a simulation model of underground cavities construction considering transportation machinery failure was established,and the influence of transportation machinery failure on construction progress was analyzed.The simulation schedule and transportation density of the underground cavities were obtained.The accuracy and superiority of the model were verified by comparative analysis,which provided scientific theoretical guidance for real construction management.