Utilities routinely conduct system monitoring activities. However, these efforts may be a more passive than active activity if they are not leveraged into actionable insights. So, system performance monitoring, simulations, and information about problem-areas may be collected, but how often are all the dots connected? Unexpected system anomalies are non-optimal at best and may be even more consequential under certain circumstances. Significant precipitation events can put collection system infrastructure to the test, sometimes resulting in even more devastating consequences. But utilities can diagnose potential problems before they become catastrophic using five steps.
Five steps to diagnose hidden infrastructure problems
Step 1: Create a digital twin
A digital twin provides a virtual and constantly updating representation of your collection system, spanning its lifecycle and the changes, issues, and affects that is included in that. Including near real-time data combined with simulation, machine learning and artificial intelligence anomaly detection and prediction, a digital twin can help with on-the-fly, and even predictive, decision making.
But it’s not enough to just create a digital twin. You’ll also need to iterate, build, and maintain the digital twin so that it remains as accurate as possible. Several data inputs may feed into the digital twin and with a few recommendations, this connection of data sources and data feeds can help maintain the integrity of the digital twin.
Hydraulic Model(s): An agnostic model importer provides the most flexibility and speed. Current, calibrated models provide accuracy, speed, and forecasting capabilities.
GIS and Asset Data: Not all GIS and asset data is one-size-fits-all and your data collection system should be flexible enough to handle your needs. Your GIS and/or asset inventory data and information is never done. It will bever be complete. But anything is better than nothing and provides a great starting point.
Time-Series and Real-Time Data: Telemetry data helps tell the current situation of your system. Data can be collected through several methods, including:
- Permanent and/or Temporary Flow Meters
- Level-Depth Sensors
- Velocity meters
- Overflow alarms
- And others, whatever meets your needs and budget can be a great start
Historical Work Order Data: Most utilities leverage a computerized maintenance management system (CMMS) for work order data, but when that data is a constant part of your digital twin you have insights into the current health of your system, and better yet, insight into issue before they occur.
Precipitation Events Data: Integrate rainfall data, including depth, duration, and frequency, for better predictive accuracy. Some tools also provide a flexible tabular/graphics interface to make it easier to visually recognize patterns.
Step 2: Compare/review simulated versus observed data
Now that your digital twin is taking shape and data is flowing in, the next step involves comparing and reviewing the simulated data over observed data. Take a closer look at:
- The accuracy of modeled values
- Modeled vs Observed Flows in tandem with Precipitation events that may cause expected (and unexpected) events in your system
- Operational events (cleaning and inspections) vs blockages and overflows
- Statistical qualifications of precipitation events
Step 3: Identify anomalies in your system
Digital twins enable utilities to gain an advantage in finding anomalies before they feed predicted outcomes, and possible errors can become troublesome. Some of these anomalies include:
- Metered locations with faulty data
- Large discrepancies between the modeled vs observed results
- Blockages or strange results
For example, most sewer collapses are detected after they fail or worse become catastrophic and the road collapses. But by using digital twin sewer system software which derives insights from simulated and observed real-time conditions, users are alerted to performance deviations for investigation.
Step 4: Evaluate the effectiveness of your interventions
Look at differences in flows and system behavior before and after operational interventions. For example, do you notice a significant change after cleanings? You should also consider system design and performance as it relates to factors such as size and redundancy and how it withstands precipitation events. If your system was designed to handle large storms, see how it performs during larger than expected storms.
Step 5: Take advantage of the connected data platform you have created
Once you have a connected data platform, share it across your organization. Some tools make that amazingly easy by combining that connected data into user-friendly dashboards with output capabilities or exportable reports. And set up alerts to notify you of any system anomalies to stay on top of potential issues. When searching for a system to help you diagnose hidden problems in your collection system infrastructure, also look for a system that can:
- Help identify capacity issues
- Predict the performance with hypothetical storms
- Flag dry weather overflow events
- Track maintenance hole flooding
Become a more resilient utility by tapping into your data
Imagine the possibility of elevating system anomalies to your most experienced staff before problems occur. Utilities today have the power to become digital-first and resilient enough to manage the predictable and unpredictable. Schedule a free consultation with the Trinnex team to learn more about our digital services and products.