Posterior Estimation for Water Distribution Network Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current prediction models for water distribution networks are inadequate due to reliance on short-term data and limited measurement points, leading to inadequate calibration and reduced accuracy over time, especially with changes in system parameters, infrastructure, and operational conditions.
Innovation Solution
A method and system for posterior estimation of variables in water distribution networks, utilizing a computer program that generates and updates models based on data inputs, including recursive estimation to adapt to real-time changes and improve model accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If short-term hydraulic data samples are used for parameter estimation, then the modeling process is faster and requires less data, but the model calibration is inadequate and accuracy deteriorates over time
Solution Approach 1:
The system performs preliminary actions by collecting and storing hydraulic data over extended periods before modeling is needed. Data from multiple measurement points are pre-processed and stored in a database, so when modeling is required, the system can draw from this pre-collected comprehensive dataset rather than requiring long collection periods each time
Solution Approach 2:
The system creates a virtual copy of the physical water distribution network through digital modeling. This digital twin replicates the network's hydraulic behavior, allowing virtual simulations and predictions without requiring physical measurements during the modeling process, thus reducing time loss while maintaining accuracy
2Device complexity
If data from a few measuring points is used, then the measurement system is simpler and cheaper, but the model cannot accurately represent the full range of system conditions
Solution Approach 1:
The system segments the water distribution network into multiple zones and identifies strategic measurement points within each segment. By placing sensors at representative locations in different network zones, the system captures diverse hydraulic conditions across the entire network using a manageable number of measurement points
Solution Approach 2:
The measurement system is designed with multi-functionality where data from each measurement point serves multiple purposes: calibrating model parameters, validating hydraulic simulations, detecting anomalies, and predicting future conditions. This universal use of measurement data maximizes the value extracted from each sensor, reducing the need for additional measurement points
3Reliability
If the model is updated frequently to reflect system changes, then the model remains accurate, but the investment of time and resources increases substantially
Solution Approach 1:
The system implements continuous feedback mechanisms where actual hydraulic measurements are compared against model predictions. When deviations exceed thresholds indicating system changes, the feedback triggers targeted model updates only in affected areas. This selective updating maintains accuracy while minimizing the time and resources required compared to comprehensive frequent re-calibration
Solution Approach 2:
The modeling system transitions from static periodic updates to dynamic adaptive updating. The system continuously monitors data streams and automatically adjusts model parameters in real-time based on detected changes, allowing the model to remain accurate without requiring scheduled manual intervention and extensive re-calibration efforts
Data Source
AI summary
A system for posterior estimation of variables. Receiving a set of data inputs. Determining a first model of the water distribution network based on the set of data inputs. Determining a second model of the water distribution network based on the set of data inputs, and the first model.