Study On Earthwork Dynamic Allocation Based On MCTS Algorithm

The construction process of concrete face rock-fill dam is affected by hydrology,meteorology and geology,and there will be some deviation between actual progress and planned progress.It is the key to ensure the completion of the project on schedule to analyze the impact of schedule deviation on the follow-up work and the total construction period,and to control the earthwork allocation process in real time.At present,the commonly used earthwork dynamic allocation method is the earthwork optimization allocation system based on computer construction simulation technology,but this method has some disadvantages such as long calculation and analysis time,and requiring a lot of professional knowledge,etc.In recent years,the rapid development of artificial intelligence has had an important impact on various fields.According to the idea of AlphaGo-Zero,an intelligent earthwork dynamic allocation model(EDAM)based on Monte Carlo tree search(MCTS)is presented,which provides theoretical basis for real-time control of construction progress of rock-fill dam.The main research contents and results of this paper are as follows:(1)This paper systematically analyzes the filling process of rock-fill dam,the earthwork allocation process,and the adjacent month filling rise constraint.The state of EDAM is divided into the current cumulative filling volume,the filling volume last month and the current month.The action space of EDAM is restrained by the upper limit of filling intensity,which is decided by the filling working face.The reward function of EDAM is constructed by the completion of the joint construction period,the completion of the total construction period,the cost of construction machinery on the filling working face and the cost of earthwork allocation.EDAM organically combines the filling construction schedule control of rock-fill dam with the earthwork dynamic allocation.(2)For the specific problem of this paper,the upper confidence bound apply to tree algorithm(UCT)in the MCTS iterative process is improved.Combining the engineering examples,the traditional MCTS algorithm and the improved MCTS algorithm are compared.The research results show that the improved MCTS is superior to the traditional MCTS in terms of the quality and stability of the algorithm for the earthwork dynamic allocation.(3)Analysis of certainty earthwork dynamic allocation.In order to verify the feasibility of the EDAM,the comparison object selected in this paper is the computer simulation result.Since computer simulation uses the effective construction days to describe the environmental uncertainty,this paper first ignores the impact of environmental uncertainty and considers the state transition process in the EDAM as a deterministic decision process.The research results show that the filling schedules generated by the two models are basically consistent,which verifies the feasibility of the EDAM.Moreover,the calculation time of the EDAM is greatly reduced,and a new model and means is provided for the earthwork dynamic allocation.(4)Analysis of risk earthwork dynamic allocation.The study on the impact of environmental uncertainty shows that the difference between the filling schedule and the filling schedule after considering the risk is small,the schedule and the earthwork allocation strategy generated by the UEDA are reliable,and This is more consistent with the actual construction situation.