Supply Grid Pressure Control Using Predicted Node Pressure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for pressure regulation in supply networks face challenges in achieving rapid and efficient control, particularly in complex networks where simulation models are complex to create or data is insufficient, leading to delayed reconstruction of pressure or flow values, which can be time-consuming and computationally intensive.
Innovation Solution
A method utilizing a self-learning system, such as an artificial neural network, trained to predict pressure at specific locations in the supply network based on measured flows and pressures, allowing for rapid pressure control without the need for sensors at every location, enabling quick and accurate prediction of pressure values for pump and valve activation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If simulation models are used to reconstruct pressure values at multiple locations, then pressure control accuracy is improved, but computational time and complexity increase significantly
Solution Approach 1:
The patent divides the supply network into multiple pressure zones, each with its own control strategy. This segmentation allows independent optimization of each zone without requiring full-network simulation, reducing computational complexity while maintaining control accuracy within each zone.
Solution Approach 2:
The patent pre-calculates and stores pressure control strategies during the planning phase, creating lookup tables of optimal control actions for various demand scenarios. During operation, the system simply retrieves pre-computed solutions rather than performing real-time simulations, dramatically reducing computational time while maintaining accuracy.
2Reliability
If pressure is maintained at high levels to ensure adequate supply to all consumers, then consumer reliability is improved, but network component wear and energy consumption increase
Solution Approach 1:
The patent implements dynamic pressure control that adjusts pressure levels in real-time based on actual demand conditions. Pressure is maintained at high levels only when and where needed, rather than uniformly across the entire network, reducing energy consumption while ensuring consumer reliability through adaptive response to changing conditions.
Solution Approach 2:
The patent applies different pressure levels to different zones and time periods based on local demand characteristics. High pressure is applied only to zones with high demand or during peak periods, while low-pressure zones operate at reduced pressure, optimizing the balance between consumer reliability and energy consumption locally rather than uniformly across the entire network.
3Loss of information
If extensive sensor placement is implemented to monitor pressure at all locations, then measurement coverage is improved, but system complexity and cost increase
Solution Approach 1:
The patent introduces a simulation model as an intermediary that estimates pressure values at unmonitored locations based on measurements from a limited number of sensors. This mediator allows the system to achieve comprehensive pressure monitoring coverage without physically installing sensors at every location, reducing system complexity while maintaining information completeness.
Solution Approach 2:
The patent creates a virtual copy of the supply network through simulation modeling. This digital twin replicates the hydraulic behavior of the physical network, allowing pressure values at all locations to be inferred from limited sensor measurements through the simulation model, effectively extending monitoring coverage without additional physical sensors.
Data Source
Figure 1~2
Figure 3
AI summary
The invention relates to a method for controlling pressure in a supply grid (10). The supply grid (10) is suitable for supplying fluid to loads. The supply grid (10) has first sensors (14) for measuring the flow and/or the pressure of the fluid at first locations (141) in the supply grid (10) and a pump (16) for pumping the fluid or a valve for controlling the flow of the fluid. The method for controlling pressure has the following steps: a) measuring the flow and/or pressure of the fluid at the first locations (141) in the supply grid (10) by means of the first sensors (14); b) predicting the pressure at the second location (151) in the supply grid (10) using a self-learning system (SS) on the basis of the flows or pressures measured in step a), wherein the self-learning system (SS) is trained to predict the pressure at a specified location in the supply grid (10); and c) actuating the pump (16) or the valve at least also on the basis of the pressure predicted by the trained system (SS) at the second location (151). The invention also relates to a device (30) for controlling the pressure in a supply grid (10) and to an assembly comprising a supply grid (10) and the aforementioned device (30).