建筑工程全过程成本高精度预测方法仿真

Simulation of high-precision cost prediction method for the entire process of construction engineering

为了提高建筑工程全过程成本预测的精度,保证成本的有效控制,提出改进神经网络的建筑工程全过程成本预测方法。基于系统工程学建立建筑工程全过程成本标准体系,针对建筑工程成本标准体系,利用灰色关联分析法计算每个成本预测指标的权重,保留权重较大的指标数据作为训练样本;引入AdaBoost算法改进神经网络,建立强预测器对样本展开训练,以此实现高精度的建筑工程全过程成本预测。仿真实验结果表明,所提方法能够在有效选取对建筑工程成本管理影响较大的成本预测指标的基础上,具有较高的成本预测精度和训练拟合度,提高了预测结果的可靠性。

In order to improve the accuracy of cost prediction throughout the entire construction project and ensure effective cost control,an improved neural network-based method for predicting the cost of the entire construction project process was proposed.Based on systems engineering,a standard system for the cost of the entire construction project process was constructed at first.On the basis of the standard system,the grey relational analysis method was employed to calculate the weight of each cost prediction index,and then the index data with bigger weights were retained as training samples.Next,the AdaBoost algorithm was introduced to improve the neural network.Moreover,a strong predictor was set up to train the samples,thus achieving high-precision cost prediction for the entire construction project process.Simulation results show that the proposed method can effectively select the cost prediction indexes with greater impact on construction project cost management,while achieving high-cost prediction accuracy and training fitting degree.Therefore,this method can improve the reliability of prediction.