基于智能监测系统的给水管网调度方法研究

发布时间：2018-01-11 05:01

本文关键词：基于智能监测系统的给水管网调度方法研究　出处：《清华大学》2015年硕士论文　论文类型：学位论文

【摘要】：智能监测系统的发展使得管网的各种运行数据的获取更加完整。尽管数据信息越来越庞大,但是管网的优化调度仍然以人工经验为主,信息数据的利用率较低。针对上述情况,本论文开展了城市给水管网优化调度方法研究,旨在通过实时监测数据建立起供水管网优化调度的新方法。通过对水力学基本理论进行分析,本文提出了数字管网模型的管网压力计算方法,即利用智能监测系统的采集数据,结合空间插值理论对于管网的压力状态进行计算。通过建立了实验管网系统验证了模型的鲁棒性,因此解决了优化调度过程中管网压力的实时模拟问题。监测点的布置方法是数字管网模型准确建模的基础。监测点优化模型以数字管网模型的预测误差最小化为目标函数,并根据优化问题本身特点采用粒子群算法进行模型求解。针对管网中特殊结构区域,提出了理论-经验布置方法进行精度的提高,实现了对于管网中未监测节点水压的高精度计算,保证所有未监测点的计算水压误差小于3m。根据监测点的历史数据中呈现的周期性规律,本文提出采用支持向量机建立压力预测模型。经过深圳市宝城管网的实际运行数据的实践验证表明,支持向量机模型对于水压预测具有较高的可靠性,80%以上的预测结果相对误差小于5%。同时,本文提出采用模糊c均值算法实现对摩阻系数相似的管道进行分类,实现微观模型的建立。将预测的水压输入到微观模型中,实现管网水量的准确预测,预测结果与实际水量变化曲线较为吻合,日总用水量误差仅为5.1%。在分析了含有变速泵的泵站运行工况的基础上,本文建立了含有变速泵的直接优化调度模型。通过分析优化调度模型中目标函数和约束条件的特点,提出了模型求解的预处理方法。针对优化问题中含有的多种混合变量,提出了粒子群算法的编码方式和具体的实现过程。通过深圳市宝城管网模型的仿真实验验证了该优化方法的有效性,经过优化后的管网运行压力得到了降低,压力变化更为平缓,最大压力变化值由2.9m降至1.13m。
[Abstract]:With the development of intelligent monitoring system, all kinds of operation data are more complete. Although the data information is more and more huge, the optimization of pipe network scheduling is still based on artificial experience. The utilization rate of information data is low. In view of the above situation, this paper carried out a study on the optimal dispatching method of urban water supply network. The purpose of this paper is to establish a new method for optimal dispatching of water supply network through real-time monitoring data. Based on the analysis of the basic theory of hydraulics, this paper puts forward a method for calculating the pressure of water supply network based on digital pipe network model. That is to use the intelligent monitoring system to collect data, combined with the spatial interpolation theory to calculate the pressure state of the pipe network. The robustness of the model is verified by the establishment of the experimental pipe network system. Therefore, the problem of real-time simulation of pipe network pressure in the process of optimal dispatching is solved. The layout method of monitoring points is the basis of accurate modeling of digital pipe network model. The optimal model of monitoring points minimizes the prediction error of digital pipe network model. Objective function. According to the characteristics of the optimization problem, the particle swarm optimization algorithm is used to solve the model, and the theoretical and empirical layout method is proposed to improve the accuracy of the special structural area in the pipe network. The high precision calculation of the water pressure of the unmonitored nodes in the pipe network is realized, and the calculation water pressure error of all the unmonitored points is less than 3m. according to the periodicity of the historical data of the monitoring points. In this paper, the support vector machine (SVM) is used to establish the pressure prediction model. The actual operation data of Shenzhen Baochengguan network show that the SVM model has a high reliability for water pressure prediction. The relative error of the prediction results above 80% is less than 5. At the same time, the fuzzy c-means algorithm is proposed to classify the pipelines with similar friction coefficient. The predicted water pressure is input into the microscopic model to realize the accurate prediction of the pipe network water volume. The prediction results are in good agreement with the actual water volume change curve. The error of daily total water consumption is only 5.1. Based on the analysis of the operating conditions of the pump station with variable speed pump, the total water consumption error is only 5.1. In this paper, a direct optimal dispatching model with variable speed pump is established, and the characteristics of objective function and constraint conditions in the optimal scheduling model are analyzed. A preprocessing method for solving the model is proposed, aiming at the mixed variables contained in the optimization problem. The coding method and implementation process of PSO are proposed. The effectiveness of the optimization method is verified by the simulation experiment of Shenzhen Baochengguan network model, and the operation pressure of the optimized pipe network is reduced. The change of pressure is more gentle, and the maximum pressure change is decreased from 2.9 m to 1.13 m.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU991.3

【相似文献】