Client: William Kerwin, Jr., Executive Director, Orange Water and Sewer Authority

Student Consultant: Amy Buege

Date: May 1997

Executive Summary

The Orange Water and Sewer Authority (OWASA) treats and supplies water to the Towns of Chapel Hill and Carrboro, North Carolina. A major expense in operating it’s water treatment plant is the electrical cost associated with pumping raw water from the reservoirs to the water treatment plant (WTP), finished water from the plant to its elevated tanks, and finished water in the reverse direction from the clearwell up through the filter media.

Duke Power provides electrical power to OWASA under cost Schedule OPT. This schedule divides the day into two segments, on-peak and off-peak, and charges rates during the on-peak period that are three times greater than those charged during the off-peak period. Thus, OWASA can decrease its electrical bills by taking advantage of the off-peak rates and scheduling the majority of its pumping during the that period. This study addresses this issue by determining optimal pumping strategies that minimize electrical costs while ensuring a sufficient supply of raw and finished water. At OWASA’s request, the study also examines the strategy of Peak Shaving, which involves the use of OWASA’s back-up generators to reduce on-peak electrical consumption.

Optimal Strategies for Finished-Water Pumping

To provide a quantitative means of determining optimal pumping strategies, the project formulated two models of the finished-water pumping system. One applies to the winter months, October through May, and the other to the summer months, June through September. The principal differences between these models are in demand levels and the rates charged for on-peak electricity. The models use an operations research optimization technique called dynamic programming to determine pumping strategies that minimize OWASA’s finished water pumping costs for an entire day. A pumping strategy is a daily schedule that specifies when each OWASA pump is to be turned on and off. The model was executed using the Microsoft Excel spreadsheet software so that it can be up dated easily to reflect any changes in the parameter estimates that occur.

This study shows that:

  • For winter demand ranging from 5 to 11 million gallons a day (MGD), and summer demand between 7 and 10 MGD, the optimal pumping strategies use either Pump 2 or Pump 3 during the on-peak time period. These strategies induce an on-peak kilowatt demand no greater than 115 kW. By contrast, current strategies used at OWASA often cause on-peak demands as high as 400 kW.
  • The optimal strategy for a 7 MG winter demand has an associated daily cost of $172.69. By comparison, a 7 MG strategy employed in February had a daily cost of $284.98, indicating a potential cost savings of 40 percent when using the optimal strategy.
  • If OWASA implements the optimal winter and summer strategies, it would further benefit by negotiating a separate demand contract with Duke Power of around 500 kW for on-peak electrical consumption.
  • The labor and generator-maintenance costs resulting from Peak Shaving exceed the electrical savings realized from reduced electrical consumption, indicating that Peak Shaving itself would not be profitable at this time.
  • A modified form of Peak Shaving, which uses back-up generators in the event of a water emergency, may benefit OWASA by allowing it to maintain low kilowatt demand levels. Further analysis in this area is recommended.
  • The optimal finished-water pumping strategies for all the demand levels generated in this study are found in the operators manual provided with this report and in Appendix E. The manual provides the operators with all of the information, such as the pump combinations to be used during each stage of the day and the corresponding elevated tank volume, required to implement the optimal strategies.

Raw-Water Pumping Strategies

As a means of decreasing on-peak electrical consumption the study also examined raw-water pumping. Because raw-water pumping and demand are determined by the finished-water pumping strategy, it was necessary to determine appropriate raw-water strategies corresponding to each of the finished-water solutions. OWASA’s clearwell holds the finished water before it is pumped to the distribution system. Since the current capacity of the clearwell limits the feasible raw-water strategies, the raw-water pumping analysis has been conducted assuming an additional clearwell was available.

The analysis showed that:

  • OWASA can decrease raw-water pumping to the lowest level at both Cane Creek and University Lake, 4 and 5 MGD respectively, while continuing to meet winter and summer demands of up to 8 MGD.
  • In order to use the optimal finished-water strategies for demands exceeding 8 MGD, an additional clearwell is required.

Filter Backwashing Strategies

Filter backwashing costs can be reduced by requiring that backwashing occur off-peak. The study recommends that OWASA operators and management work together to develop a set of rules that allows the backwashing activity in order to take advantage of off-peak electrical rates. As illustration, the following guidelines provide general rules for scheduling routine backwashes in advance, to avoid unnecessary on-peak backwashing. They dictate that a backwash should be performed before the on-peak period begins if:

  1. Any filter exceeds OWASA’s backwashing standards by the end of the on-peak time period.
  2. Three or more filters exceed one of these tighter standards:
    hours = 50 hours
    turbidity = 0.2, or
    headloss = 4 feet (7 feet for the larger filters).
  3. Any of the filters has a recent history of frequent backwashes and appears to be exceeding the typical clogging rate.

The study was performed using both present day and future demands. The results presented hold for the current parameter values. However, the model should be adjusted accordingly and run again if these values change in the future.

Subject of the article Balance of Power in the Fall 1997 issue of the UNC research magazine Endeavors.