The solution was to go with OSIsofts PI Enterprise Corporate server, which is able to collect data from new control systems. All of the control system data resides in the OSIsoft PI Enterprise server and is available for existing and future modeling and forecasting.
By using this product, the current systems are fully scalable and future-proof. Thanks to the common interface between CHRIS, RtWebParts, DataLink and ProcessBook, we have enabled an agile environment.
Needing Real-Time Access and Self-Service
Finally, we required real-time access to information and self-service for all users. The goal was to increase ROI while understanding the reality that training and support only increase as user demand goes up.
What OSIsoft provided was on-site training sessions in the train the trainers model and computer-based training. All users went through hands-on training to facilitate easy adoption. The RtWebParts solution was also in production with a scalable vector graphics viewer available and was implemented quickly on all 1,200+ desktops. Separately, DataLink and ProcessBook were available on request.
The OSIsoft System at Work
Once we had these pieces in place, we were able to implement some impressive real-time monitoring solutions, such as our storm sewer overflow predictor (SSOP). Before the PI System, data was not easily accessible. What was available could only be viewed on one SCADA console, which severely limited our ability to act. Data and possible trends could not be analyzed effectively, as the program was not supported by Excel. This lack of information led to local problems. For example, on the north shore of Vancouver, overflowing storm sewers caused flooding in residential basements. The GVRD received calls after the overflow event after it was too late to open the valves.
With the PI System, we are able to get data very quickly and in real time and to be creative and efficient with its use, which is how the SSOP was born. Through DataLink and using a linear regression model, current flow data in Excel is compared to historical data to predict the probability of an overflow. The spreadsheet is then stored as an HTML file and made available to all necessary employees on the intranet. This has eliminated overflows and has greatly increased customer satisfaction. Marilyn Towill, GVRD senior project engineer, says, Ive been monitoring the SSO Predictor, and it gave me some comfort through the rainy holiday season knowing that no overflow was expected.
Similarly, another predictor forecasts wastewater overflow, which prevents raw sewage overflows, damage to the environment and fines to the GVRD, which can cost up to $1 million per violation.
Real-time data also drives the dam monitoring system, an advanced warning system to increase local residents safety in the event of a storm or unforeseen structural problem (See Figure 1 below).
FIGURE 1: FLOOD ALERT!
|Real-time data in the dam monitoring system provides advanced warning in the event of a heavy storm or unforeseen structural problem.
Once we had the OSIsoft PI system in place, we could calculate tags for alarm limits. This allowed alarm-limit history to be stored and trended along side monitored tags. These tags are now monitored by an email program that sends alerts when alarm limits are breached.
We plan to incorporate PI and related products in other arenas, including the ozonization plant, air quality monitoring and water filtration plant information. With the help of OSIsoft, the GVRD has been able to build and maintain a sustainable region, improving the quality of life in the greater Vancouver area.
|About the Author|
Gary L. S. Wong is the manager, corporate applications, at the Greater Vancouver Regional District.