摘要
To address the challenges of high input dimensionality,high computational costs,and insufficient prediction accuracy in the safety early warning process of hoisting operations,this study proposes a safety early warning model based on Random Forest(RF),Genetic Algorithm(GA),and Support Vector Machine(SVM).First,the RF algorithm is employed to assess the importance of indicators involved in the hoisting safety process,thereby reducing data dimensionality and improving the operational efficiency of the model.Next,the GA is used to optimize the parameters of the SVM to enhance its generalization capability.Finally,an integrated safety early warning model is constructed by combining RF and GA-optimized SVM.Experimental comparisons using randomly selected case data demonstrate that the proposed model offers significant advantages in early warning accuracy.Compared with traditional models,the classification warning accuracy improves by 11%,confirming the feasibility of the model.
