Existing facilities offer fertile opportunities
As background, we believe it’s fair to say that water treatment plants in many communities, and also in industry, are not always state-of-the-art. The operations often date back to the 1970s or 80s, and are costly in terms of energy and chemical consumption. Process control is frequently done via manual adjustments of parameters like aeration pumps and chemical dosage, which in turn are based on manual sampling and retrospective tests done on regular intervals. To be on the safe side with respect to contaminant limits in effluent, over-aerating and overdosing of chemicals is common.
Approximately two years ago Kemira started a program to explore the opportunities Big Data might offer to help aging plants both increase operating efficiency and also meet rising demand for water. While Big Data can mean many things and encompass many different subjects such as smart technology, machine learning, artificial intelligence and other new areas, we decided from the start that any Big Data program must be aimed at tangible problem solving. Specifically, the target was to reduce OPEX and CAPEX for water treatment plant operators and owners.
Interviews explore the challenges, and needs
A series of three increasingly-targeted interview programs were carried out with dozens of water treatment operators. Eventually national and regional water regulatory agencies were also involved in the interviews. Not surprisingly, the interviews revealed that around 80% of challenges in WWTP are related to poor plant operation. Quite often a lack of understanding of the chemistry involved is seen as part of the problem.
When we presented our ideas on how we might use data to improve operations, the reaction from interviewees was overwhelmingly positive. In fact, most participants were surprised that we were as far along as we were at that point, and several asked if they could join the program in a deeper manner immediately.
Rapid, accurate prediction of sludge properties
With this encouragement, intensive work was done which resulted in a new tool that can very accurately and rapidly, within seconds in fact, predict the properties of a sludge at a water treatment plant. This is not a future scenario; it is what we can already do today. By combining existing operational data, historical process data, machine data, chemical data and site data, and then applying newly-developed advanced analytics to benchmark the customer against similar sites, we can obtain a very accurate prediction of sludge properties.
The key benefit of this tool is that it gives operators tangible ways to reduce operating costs via better identification and understanding of the key process conditions and chemical properties which impact the sludge dryness. This also allows smoothing out of plant operations. The algorithms are now being further developed to do even more, enabled by artificial intelligence and machine learning.
Pioneering work continues
For many years Kemira has been a pioneer and thought leader in the use of real-time sensors to optimize water treatment. The present work is an important step towards harnessing the combined power of such sensors and Big Data to get cost benefits.
In the long run, we imagine an operation where all water inflow and outflow is measured constantly in real-time by sensors for variables such as pH, oxygen, nutrients, phosphates, nitrates, sludge dryness, pathogens, etc. With the right algorithms, this data can be used to continuously optimize pumping and aerating energy consumption as well as chemical dosing. In this way, operations can be smoothed out and operating costs trimmed, while still remaining safely within legal limits.
