The Not Too Distant Future: How OSIsoft Is Helping Water Utilities With Near-Term Analytics To Save Energy, Time And Water
United Utilities: predicting day-ahead water demand with 98% accuracy
San Leandro, CA - It is difficult to predict what will occur a year from now. It can even be harder to predict what will happen tomorrow.
OSIsoft, a leader in operational intelligence, is collaborating with water utilities worldwide on improving and implementing systems for near-term predictions, i.e. precise forecasts concerning water demand, equipment usage or other operational metrics to help them reduce maintenance costs, energy and improve service.
OSIsoft, for instance, is a technology partner in Stream, an R&D initiative created by United Utilities, one of the largest water utilities in the United Kingdom, and a host of universities to develop analytics for the water industry. In one project, for instance, United and the University of Exeter created 24 Evolutionary Artificial Neural Networks (EANN), a type of predictive model, to precisely estimate day-ahead water demand. Real-time data from OSIsoft’s PI System is used to continually update the EANNs to fine-tune the predictions.
In a pilot project involving five different district usage zones ranging in size from 335 customers, 5.1 kilometers of water mains and 5.4 cubic meters of water per hour to nearly 9,000 residential and customers consuming 216 cubic meters per hour over 119 kilometers of water mains, United was able to accurately predict water consumption 24 hours in advance with 98% accuracy, paving the way for United to save energy and equipment wear by forgoing pumping or other activities.
United is also employing PI System data on another project to predict the storm drains most at risk for flooding and debris clogs before large storms strike.
“The challenge with near term analytics is that you have to be right. Water utilities can’t rely on analytics that are ‘close enough’ or ‘in the general ballpark’. Otherwise, they run the risk of compromising water quality, breaking assets or hurting customer service,” said Gary Wong, Industry Principal at OSIsoft. “United is showing how analytics will help utilities run better while running leaner.”
Other utilities engaged near-future analytics include Municipal Water and Sewage Company S.A. in Wroclaw (MPWiK), one of the largest water utilities in Poland, which is employing predictive analytics to better synchronize water flow and pressure with city activities. Yorkshire Water, meanwhile, uses PI System data and meter data to more quickly identify and fix water leaks to cut water losses and energy as well as meet its regulatory requirements. (Yorkshire will further discuss its results at the OSIsoft EMEA Users Conference taking place in London October 16-19.)
“Energy can be 30% percent of the cost of producing water and water utilities are often the largest consumers of power in their regions. By simply monitoring energy better we see utilities regularly reducing energy consumption by 6 to 12% and reducing labor by hundreds of thousands of dollars per year,” said Wong. “We’re excited about what these utilities are doing with analytics because it promises to accelerate what already is being achieved. Water utilities are conservative by nature, but they could become one of the most important test beds for analytics.”
About OSIsoft
OSIsoft is dedicated to helping people transform their world through data. Our PI System captures operational data from sensors, manufacturing equipment and other devices and turns it into rich, real-time insights to reduce costs, make critical business decisions and enhance products. Over 1,000 leading utilities, 95 percent of the largest oil and gas companies and more than 65 percent of the Fortune 500 industrial companies, rely on the PI System to manage their businesses. Worldwide, PI System manage over 1.5 billion sensor-based data streams. To learn more, please visit www.osisoft.com.
Media Contact
Michael Kanellos
(510) 877-9331
Source: OSIsoft