With the deepening of the “Belt and Road” initiative,the foreign trade and direct investment of China have maintained a steady growth trend.To diversify cooperation in infrastructure construction and energy project development,China should be grasping the opportunities of global development,simultaneously not neglecting the importance of risk prevention.Risk management is a key step and important content of engineering investment control.To improve the risk response capability of the project,and promote the scientific decision-making of investment,it is of great practical significance to research the risk assessment and control of engineering investment.This paper identifies the internal and external investment risks of the project through investment composition and meta-analysis.To analyze the main risks and build an investment risk system,Pareto principle is used.The least square support vector machine(LS-SVM),optimized by improved particle swarm optimization(PSO),is used to evaluate the risk factors of the investment risk system.For grade I high risks and grade II medium-high risks,they should be avoided and mitigated based on BIM-used whole process investment control.While for grade III medium risks,risk sharing should be based on the bargaining game model in which participants negotiate unfair status.And about grade IV medium-low risks and grade V low risks should be monitored online based on BIM risk early warning system.By studying on the assessment and control of investment risks,the main conclusions of this paper are as follows:(1)Through the investment composition,meta-analysis and Pareto principle,the risk is identified and analyzed,and an objective and comprehensive investment risk system is constructed,which fully considers the internal and external environment of the project and provides a reliable data source for risk assessment and control.(2)The LS-SVM with improved PSO optimization can accurately evaluate the risk level.PSO is used to optimize the regularization parameters and kernel function parameters in the LS-SVM objective function.And to avoid the PSO optimization process falling into local optimum and premature convergence,the average particle distance and fitness variance are introduced to improve the PSO.It can keep a better distribution of particles,reduce the randomness of search,and improve the dynamic and convergence characteristics of the algorithm.Therefore,compared with other algorithms,the optimized LS-SVM has higher prediction accuracy and generalization ability,which is suitable for small samples and non-linear risk evaluation.(3)BIM technology provides reliability guarantee for obtaining complete information in risk sharing game process.From the perspective of the unequal negotiation status,based on the complete information and incomplete information,the risk-sharing game model of project participants is constructed,and the conclusion of the game is that information plays a decisive role in risk sharing.BIM technology can highly integrate engineering information to ensure accurate,comprehensive and efficient information mastered by investors,which further illustrates the feasibility and importance of applying BIM to risk control.(4)BIM-based risk control provides a reference for dynamic management of investment risk.The whole process investment control and risk early warning system based on BIM can effectively monitor risks in real time,make the emergency response quickly,and realize the intelligent and information management of risks. |