The Optimization State

News and views from the AWOP states.  Please use this for your enlightenment, enrichment and maybe even your entertainment!  AND think about what your state wants to share for the next AWOP News.

Iowa

Greenfield Success Story

In fall of 2007, the Iowa DNR field staff offered training in the use of the Optimization Assessment Software (OAS) to all of the state’s 32 surface water systems.  In response to the training, several water system operators began using OAS and sending it in to the central office for review each month.  One of these was Water Plant Foreman Garry Miller of the Greenfield Municipal Utilities (GMU) water plant.  The plant utilizes water from six wells, Greenfield Lake, Nodaway Lake, and the Middle Nodaway River to produce water for a population of approximately 2,300 people.  Potassium permanganate is added at the Greenfield Lake inlet, and then water flows by gravity to the treatment plant, where coagulant is added.  Flocculation/sedimentation is accomplished through a Trident Microfloc clarifier/filtration system followed by disinfection and fluoridation.  Because of the lack of sedimentation in this process, the plant is classified as direct filtration.

Miller began using the OAS software in November of 2007, and he noted with his first electronic submittal that the plant had filter to waste, but that filters were put back in service once turbidity dropped below 0.40 NTU.  He thought he could lower that to 0.20 NTU or lower.  He also mentioned that he was in the process of finding a computer programmer to repair a problem in the software that he thought would reduce the combined filter effluent (CFE) to less than 0.1 NTU 95 percent of the time.  On November 29, 2007, the computer programmer arrived and found that at 12:01 a.m. each morning, the computer program was taking that combined filter effluent turbidity measurement and adding it to the 12:00 a.m. measurement and recording it in the spreadsheets as the reading for 12:01 a.m., effectively doubling the reading.  Many times, this 12:01 a.m. reading was the highest reading of the day.  The OAS spreadsheet showed an immediate effect following the fix.  The 95th percentile for CFE did not change from 0.19 NTU, but GMU went from meeting the CFE optimization goal of 0.10 NTU 17.5 percent of the time, to meeting it 49.5 percent of the time after the computer program fix.  It also provided a more accurate picture of how things were going at the plant.

Program fix to more accurately portray CFE turbidity

After taking a look at the data in the OAS spreadsheets in November and December of 2007, Jennifer Bunton of IDNR contacted Miller about days with very high turbidities and found that Garry was reporting turbidity data even on days when filter maintenance was being performed, because he hadn’t realized these numbers were not considered valid for compliance.  They also discussed the fact that most of the maximum daily values were occurring during backwash and filter to waste.  Miller thought about this and decided that maybe there was a problem in the control panel, because the relays were supposed to block turbidity data from reaching the plant computer and spreadsheets during backwash and filter to waste.  He thought that perhaps the input signal to the relay was coming from the wrong terminal in the plant PLC, so he talked with his manager about it, and they agreed this could be a problem.  In November of 2008, Miller and the Utilities Superintendent, Duane Armstead, were able to negotiate a deal with the control panel technician and he came out to fix the problem.  Results were evident immediately, as the OAS data showed. 

Control panel fixed to eliminate recording during backwashes and filter to waste periods

In the year since the backwash and filter to waste data were blocked from recording, GMU has gone from meeting the individual filter goal of 0.10 NTU zero percent of the time to meeting the goal 68.8 percent of the time—a drastic improvement.  The GMU plant is also now meeting the CFE goal 97.3 percent of the time.  There have not been a lot of operational changes at the plant, and Miller has not been able to participate in the state’s Performance Based Training program because of demands on his time, but GMU now has more representative data to use for optimization purposes.  This shows a very different picture from what IDNR saw during its initial data collection efforts in 2006, and it also shows the benefit of just providing optimization information to systems in a format that is easy to understand.  Miller agrees, saying, “I guess I never paid too much attention to all this before I started using the OAS spreadsheets…Thanks for planting the seed as far as keeping a closer eye on how your plant is truly performing.”

Miller says he is a “behind the scenes kind of guy,” but Bunton disagrees.  “It’s only because Garry took the initiative to start thinking about why his data looked the way it did that he was able to convince his manager to make the changes necessary to portray the true picture of what was going on at the GMU plant.  His attitude and persistence are to be commended and his actions show that he is truly a professional.” ♦

Advertisements

Comments are closed.