在建设项目管理中,前期工程估价是一项十分重要的工作,根据工程造价的提点,分解研究工程造价的构成,确定了7个主要影响因素,作为神经网络的输入变量,利用BP神经网络的基本原理建立工程估价模型,收集10组工程估价实例作为训练样本,并另取2组实例用作检测样本,经测算其精度,符合工程投资估算的要求。因此利用BP神经网络估算工程造价是可行的。 In the construction project management, the pre-project valuation is a very important work. According to the construction cost, it disaggregates the composition of the research project cost and determines the seven main influencing factors. As the input variables of the neural network, the BP neural network The basic principles of the establishment of engineering valuation model, the collection of 10 examples of project valuation as a training sample, and another two examples used as a test sample, the accuracy of the calculation, in line with the requirements of the project investment estimates. Therefore, using BP neural network to estimate the project cost is feasible.