摘要
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 temperature control measures of concrete dam is a complex multi-factor system optimization problem. In order to optimize the temperature control measures of concrete dam,the concrete temperature control measures of the neural network intelligent optimization model was established by taking the highest temperature and the maximum cooling rate of the concrete as input and water flow,water temperature,water flow time and water pipe spacing as output. Uniform design method was used to the temperature control measures combination design,and carrying out simulation analysis of temperature field for obtaining the learning samples. Then the optimization neural network model was otained. The corresponding temperature control measures were selected by inputting the measured maximum temperature and the maximum daily cooling rate. The results show that the optimization neural network model of temperature control measures is reasonable and feasible.
出处
《水电能源科学》
北大核心
2017年第6期96-99,共4页
Water Resources and Power
基金
国家自然科学基金项目(51479103)
关键词
混凝土坝
温控措施
优选
神经网络智能优选模型
concrete dam
temperature control measure
optimization
neural network intelligent optimization model
