摘要
结合城市日用水量影响因素的特点和变化规律,建立了城市日用水量预测模型,采用粒子群优化算法优化BP人工神经网络的连接权值,以求解该预测模型。经优化后的BP人工神经网络运算速度快、泛化能力强、预测精度高。实例验证结果证明该日用水量预测模型和求解方法是可行的。
Combined with the characteristics and variation rule of the factors influencing municipal daily water consumption, a forecast model for municipal daily water consumption was set up, then the particle swarm optimization was used to optimize BP artificial neural network to solve the model. The optimized BP artificial neural network model has advantageous properties such as more rapid calculation, high generalization performance and high accuracy. The experimental results prove the forecast model for daily water consumption and the solution are feasible.
出处
《中国给水排水》
CAS
CSCD
北大核心
2007年第7期89-93,共5页
China Water & Wastewater
基金
国家自然科学基金资助项目(50278062
50578108)
关键词
城市日用水量
预测模型
粒子群优化算法
BP人工神经网络
泛化能力
municipal daily water consumption
forecast model
particle swarm optimization
BP artificial neural network
generalization performance
