Sub-Network Water Flow Estimation Using Hydraulic Model Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating input and output water flows in water distribution systems are either expensive and difficult to implement, as they require physical closure of valves and installation of flow meters, or are limited to specific network topologies, such as those organized around a main pipe.
Innovation Solution
A method that uses sensors to acquire observations of state variables and applies an optimization procedure to determine control variables, allowing for the estimation of water flow rates at boundaries of any sub-network without physically closing valves or installing flow meters, using a hydraulic model and iterative adjustment of control variables.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If valves are physically closed to limit the number of pipes between a sub-network and the rest of the water distribution network, then the measurement of water flow becomes simpler and more direct, but the device complexity increases due to the number of valves and flow meters to install and operate
Solution Approach 1:
The patent uses a virtual copy of the physical network (hydraulic model) to estimate flows instead of physically measuring every boundary flow. The model replicates network behavior using sensor data from within the sub-network, eliminating the need for physical flow meters at boundaries while maintaining measurement capability through simulation.
Solution Approach 2:
The patent replaces the mechanical measurement system (physical flow meters at boundaries) with a computational system (optimization algorithm + hydraulic model). Instead of mechanically measuring flow at each boundary, the system uses mathematical optimization to infer boundary flows from internal sensor data, substituting mechanical measurement with computational estimation.
2Measurement precision
If valves are physically closed to create District Metered Areas, then water flow estimation for sub-networks becomes more accurate, but the ease of operation deteriorates due to the difficulty of modifying sub-networks and operating valves physically
Solution Approach 1:
The patent implements dynamic sub-network boundaries that can be reconfigured through software without physical valve operations. The optimization framework allows sub-network definitions to be dynamically adjusted by changing parameter settings in the hydraulic model, enabling flexible reconfiguration of monitoring zones without mechanical intervention.
Solution Approach 2:
The patent changes the operational parameters of the system from physical valve positions to software-defined network parameters. By modifying sub-network boundaries through parameter changes in the optimization model rather than physical valve operations, the system achieves flexible reconfiguration while maintaining measurement accuracy.
3Reliability
If physical flow measurement devices are installed at sub-network boundaries, then the reliability of water flow data improves, but the loss of energy increases due to water stagnation in closed pipes and reduced hydraulic efficiency
Solution Approach 1:
The patent introduces an intermediary computational system (optimization algorithm) that mediates between internal sensor measurements and boundary flow estimation. Instead of directly measuring boundary flows (which would require closing valves and reducing hydraulic efficiency), the algorithm acts as an intermediary to infer boundary flows from internal pressure and flow sensor data, maintaining system openness while achieving reliable measurements.
4Difficulty of detecting and measuring
If a specific network topology organized around a main pipe is used, then the detection of leaks becomes simpler with remote sensors, but the adaptability deteriorates because the solution is limited to specific topologies and cannot be applied to general network configurations
Solution Approach 1:
The patent creates a universal optimization framework that can detect leaks and estimate flows in any network topology, not just main-pipe configurations. The hydraulic model and optimization algorithm are topology-agnostic, working with any network structure by incorporating its specific geometry and characteristics into the model parameters, thus achieving multi-functionality across different network types.
Data Source
AI summary
A system for measuring water flows in a sub-network of a water distribution network is provided. The system includes a plurality of sensors, for example pressure sensors, on the network. The system further comprises communication links between the sensor and one a computing device, and a measurement acquisition system. The computing device is configured to retrieve values of measurements, directly or through the measurement acquisition system; use values of measurements to determine values of control variables of a model of the water distribution network which minimize residue values between measurements values and predicted physical values on the network; then use the model parameter with the values of control variables to calculate water flows at the boundaries of the sub-network.