AN OVERALL ASSESSMENT OF THE IMPACT EVALUATIONS FOR THE ENERGY SAVINGS PLAN PROGRAM COMPLETED TO DATE (1989 - 1992)
PREPARED BY PACIFIC NORTHWEST LABORATORY
Through its Energy Savings Plan (ESP) Program, Bonneville Power Administration (Bonneville) offers acquisition payments to industrial firms in the Pacific Northwest who install energy conservation measures. Pacific Northwest Laboratory (PNL) has performed impact and process evaluations of the program for Bonneville. This report summarizes the results of the first 10 site-specific impact evaluations of the individual projects in the program of the 52 completed through June 1992.
Briefly, the objectives of the impact evaluations are to determine for each project: 1) annual electrical energy savings in kilowatt-hours (kWh) and kWh/unit of output, 2) changes in output levels at the firm as a result of the ESP and how changes affected energy savings, 3) the net impact of the project to the serving utility, 4) the real levelized cost of the project from the perspectives of Bonneville and the region, and 5) if the project was a free rider (the project would have been installed anyway without an acquisition payment from Bonneville).
From the impact evaluations completed to date, the following conclusions were made:
- Energy savings obtained in the industrial sector by the ESP are significant. For the 10 projects evaluated, savings varied from 286,000 kWh/yr to 24 million kWh/yr, averaging about 2.3 million kWh per project (excluding the 24 million kWh/yr project).
- Six technologies account for 90% of the savings achieved by the 52 ESP projects completed. The six technologies are electrochemical improvements, refrigeration controls, compressor improvements, waste heat recovery, adjustable speed drives, and lighting improvements. Electrochemical improvements at three firms account for nearly half of the program's total reduction in energy use.
- While the output of many plants varies over time, output changes were not a direct result of ESP participation. Rather, output changes are a reflection of a changing market condition. However, a few ESP projects facilitated subsequent increases in plant output which were caused by market changes.
- It is difficult to predict with any certainty the long-term energy savings (persistence) from energy conservation projects in the industrial sector, mostly because of the dynamic environment in which firms operate.
- For most projects evaluated, the unit savings (energy savings per unit of plant output) were not determined. There are two primary reasons for this. Many firms consider output data proprietary, and some ESP projects are installed on auxiliary processes that are not directly affected by plant output.
- Three of the 10 projects evaluated, all installed prior to 1991, were found to be free riders.
- Nevertheless, the ESP is a cost-effective means to obtain energy conservation in the Northwest. The real levelized cost of industrial energy conservation from the ESP projects evaluated is considerably lower than the cost of power from a new generating resource.
Impact evaluations performed to date led to the formulation of the following five recommendations:
- Implement a more effective free-rider screen in the ESP that includes three features: 1) allow only projects that show simple payback of 2 years or more; 2) consider all financial savings in the simple payback calculation, not just the avoided cost of energy saved; and 3) do not permit firms to order, purchase, or install equipment before an ESP contract is in place (current rules prohibit only purchase or installation).
- Perform long-term evaluations of the impact of selected ESP projects to assess persistence of energy savings in the industrial sector. Bonneville already plans to implement this recommendation.
- Develop an estimating procedure to evaluate the impact of energy management and control systems at cold storage plants, even before such systems are installed.
- Develop a methodology for evaluating program impacts in the industrial sector that allows evaluators to predict the impact of entire programs, not just the impact of individual projects after the fact. Such a methodology would be a useful tool for resource planners.