摘要
快速准确地进行工程成本估算对建筑企业至关重要。传统的工程成本估算方法工作量大、估算速度慢;难以满足估算精度的要求。为符合实际,文章将影响成本的特征因素分为精确量和模糊变量,利用模糊神经网络(FNN)的自组织和自学习,对模糊网络的隶属度和推理规则进行学习和优化。提出了基于组合模糊神经网络的方法,进行建设工程成本估算。通过计算实例表明该方法是有效的,为工程成本估算提供了有价值的参考依据。
To estimate building cost rapidly and accurately is very important for construction enterprises.Traditional methods use construction quota to compute building cost,which costs too much manual labor and time but can not satisfy the demand of enterprises.By constructing a composite fuzzy neural network,which compose the precise and fuzzy inputs coming from the characteristic factors of a project into one fuzzy neural network,this paper puts forward a new method to estimate building cost.It overcomes the shortcoming that only precise values or fuzzy values are used in one network.Takagi-Sugeno inference rulers and genetic algorithm are applied in setting up and training network.Finally a practice instance shows the validity of the method.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第28期174-176,共3页
Computer Engineering and Applications
基金
国家863高技术研究发展计划(编号:2001AA412150)
陕西省教育厅专项科研计划资助(编号:04JK213)