首页> 外文会议>International conference on hydraulic engineering software >The water supply network analysis tool KANET
【24h】

The water supply network analysis tool KANET

机译:供水网络分析工具Kanet

获取原文

摘要

Statistically, the greater part of water supply systems' costs is due to distribution. Two classes of models are available for the analysis of distribution networks. First, simulation, the 'hydraulic network balance', treats existing or planned systems whose dimensions and input are known. Simulation has also been extended to include the system's dynamic hydraulic behaviour, tracing also the propagation of water quality. A second class of models utilizes optimization to design a new network. Cost optimal diameters, pumping heads, inflow rates are determined employing principles of Operations Research, observing constraints including pressure and demand, flow velocities, and the hydraulics defined by the simulation model. Once a cost-optimal design has been determined based upon a 'design demand' it may still appear desirable to verify network operation under different demand patterns. KANET offers both, simulation and optimization. Operation of the package is guided by a user interface, facilitated by graphics, a database providing connectivity. The simulation algorithm, a Newton-Raphson iteration scheme solving a set of simultaneous linear equations for unknown flow correction terms, is time extended including the propagation of water quality. Optimization is by four modules. TREEOPT optimizes a branch network using Linear Programming, TREEALL evaluates the total number of branch networks of a reticulate system. For larger networks TREEGEN, a genetic algorithm, improves a population of optimal branch networks converging towards the global. The result may serve as input to FLOWGEN, a second genetic algorithm, performing an optimal adjusunent of flows restoring branch networks to reticulation. The paper is descriptive and does not reiterate mathematical formulations presented elsewhere.
机译:统计上,供水系统的大部分成本是由于分布。两类模型可用于分配网络分析。首先,仿真,“液压网络平衡”,处理其尺寸和输入的现有或计划的系统。仿真也扩展到包括系统的动态液压行为,还追踪水质的传播。第二类模型利用优化设计新网络。成本最佳直径,泵头,流入速率决定采用运营研究原理,观察包括压力和需求,流速和仿真模型所定义的液压的限制。一旦基于“设计需求”确定了成本最优设计,仍然可能似乎希望根据不同需求模式验证网络运行。 Kanet提供仿真和优化。包的操作由用户界面引导,通过图形促进,数据库提供连接。仿真算法,一种用于未知流校正术语的一组同时线性方程的仿真算法是延伸的时间,包括水质的传播。优化是四个模块。 TreeOpt使用线性编程优化分支网络,TreeAll评估了网状系统的分支网络总数。对于较大的网络树,一种遗传算法,提高了朝向全局融合的最佳分支网络群体。结果可以用作流动的输入,第二遗传算法,执行恢复分支网络的流程的最佳呼吁才能旋转。本文是描述性的,并不重申在其他地方提出的数学制片。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号