论文部分内容阅读
混凝土坝温控措施的制定是一个复杂的多因素系统优选问题。为了对混凝土坝的温控措施进行优选,以混凝土最高温度和最大日降温速率作为输入,通水流量、通水时间、通水温度和水管间距作为输出建立混凝土坝温控措施神经网络智能优选模型。采用均匀设计方法进行温控措施组合设计,并对其进行温度场仿真分析以获得学习样本,进而得到训练好的神经网络优选模型,输入实测最高温度和最大日降温速率优选出对应的温控措施。实践结果表明,该温控措施神经网络优选模型合理可行。
The formulation of concrete dam temperature control measures is a complex multi-factor system optimization problem. In order to optimize the temperature control measures of concrete dams, the intelligent optimal selection model of temperature control measures of concrete dam is established based on the maximum temperature of concrete and the maximum daily cooling rate as input, water flow rate, water flow time, water temperature and water pipe spacing as output . The uniform design method is used to design the combination of temperature control measures and the temperature field simulation analysis is conducted to obtain the learning samples so as to obtain the trained neural network optimization model. Corresponding temperature control measures are obtained by inputting the measured maximum temperature and the maximum daily cooling rate . The practical results show that the neural network optimization model of temperature control measures is reasonable and feasible.