Richard Goeltz

Residential energy use : Analysis of disaggregate data

Abstract The Energy Information Administration recently published data they collected from the National Interim Energy Consumption Survey (NIECS). NIECS includes detailed information on 4081 individual households: demographic characteristics, energy-related features of the structure, heating equipment and appliances therein, recent conservation actions taken by the household, and fuel consumption and costs for the April 1978–March 1979 one-year period. This data set provides a new and valuable resource for analysis. We summarized and analysed the NIECS data on household energy consumption — total energy use, electricity use, and use of the primary space heating fuel. The regression equations constructed explain roughly half the variation in energy use among households. These equations contain 10 or fewer independent variables, the most important of which are fuel price, year house was built, floor area, and heating degree days.

Indoor temperature changes in retrofit homes

The behavioral response (e.g. changes in indoor temperatures, attention to window and door openings) to residential technical efficiency improvements (e.g. attic insulation, storm windows) is an important and largely unresolved issue. Although there is considerable discussion concerning the extent to which households take back some of the energy savings due to technical efficiency improvements in increased comfort, there is almost no empirical evidence on the subject.

Diagnosis, parsimony, and genetic algorithms

The Communication Alarm Processor Expert System (CAP), developed at Oak Ridge National Laboratory for the Bonneville Power Administration, is a near real-time system that aids microwave communication system operators with interpreting the cause of large communication system problems [Purucker89]. Problems in the communications network are indicated by the real-time arrival of alarms at the central control site. CAP receives and processes these alarms, then presents the operator with a sorted list indicating the most probable cause (and location) generating the alarms. However, to achieve multiple problem diagnosis a diagnostic strategy is needed that: 1) satisfies the previously defined near real-time processing constraints, 2) “scales up” easily to handle large real-world applications (i.e., applications with more than 50 problems/components), and 3) gives the operator highly reliable information on the current status of the communications network. This paper describes recent successful results of our efforts to develop a general multiple problem (fault) diagnostic strategy that meets these requirements. The CAP system is currently being upgraded to incorporate this new strategy.

Estimating energy savings due to conservation programmes : The BPA residential weatherization pilot programme

Abstract The Bonneville Power Administration operated a residential weatherization pilot programme from 1980 to 1982. The programme provided free home energy audits to more than 7000 electrically heated homes in the Pacific Northwest and gave zero-interest loans to weatherize over 4000 of these homes. The total cost of the programme was

Risk Communication and the Cognitive Representation of Uncertainty

11 million. This paper describes several mehtods used to estimate the net energy-saving effect of the BPA programme (ie the electricity saving that could be directly attributed to the programme). The simplest method ivolves estimation of weather adjusted annual electricity consumption for each household. The second approach uses this weather adjusted consumption as the dependent variable in a pooled time-series cross-section regression model of electricity use. The third approach involves estimation of qualitative choice models of the decisions to retrofit and to participate in the BPA programme. Results from these models are used to define Mills ratio terms that are then used as explanatory variables in the regression model of the second approach. The range in estimated programme saving, given the diversity of analytical methods used, is surprisingly small — 3100–3300 kWh/year per average programme participant.

Determinants of electricity use for residential water heating: The hood river conservation project☆

Risk communications often contain uncertainty information associated with technologies, behaviors, medical procedures, etc. To be effective, such messages need to incorporate ideas, images, and logic that will promote comprehension by a lay public. This paper reviews research strategies designed to make explicit the implicit and frequently idiosyncratic elements of lay reasoning about uncertainty. General biases and heuristics are outlined as they may influence the communication process. Empirical data on variables of newly emerging interest are presented as an extension of earlier research paradigms, with special focus on preferences among natural language terms, causal logic, and modalities for the expression of uncertainty. It is suggested that risk communication incorporate multiple ways to communicate the same information.

Testing for Nonresponse Bias Evaluation of Utility Energy Conservation Programs

Abstract Because 85% of the homes in the Pacific Northwest have electric water heaters, water heating is the second most important residential electricity end-use in the region (second to space heating). This paper analyses the determinants of water-heating electricity use, using end-use load data and responses to a detailed home interview. These data are available for 142 homes in Hood River, Oregon. On average, these homes used 5000 kWh/yr for water heating. Almost 60% of the household-to-household variation in electricity use was explained with eight variables in a simple regression model. The number and ages of household members are the strongest determinants of electricity use: use increases by roughly 1000 kWh/yr with each additional household member. Other statistically significant determinants of water-heating electricity use are hot-water temperature, water-heater location, number of showers in the home and house type. Electricity use varies considerably throughout the year (as well as across households). Weekly usage was 50% higher in mid-winter than in summer. About half of this temporal variation is due to changes in outdoor temperatures and half is due to seasonal changes in behavior (i.e. increased use of hot water in winter).

Actual electricity savings and audit predictions for residential retrofit in the pacific northwest

This article discusses two recent evaluations of utility energy conservation programs in which testsfor nonresponse bias were possible. In these evaluations, data were obtained both from the sponsoring utilities (fuel consumption records, energy audit reports) and from households (mail and telephone surveys). Fortunately, data from one source was usually available for those households missing datafrom the other source. This permitted comparisons of respondents and nonrespondents on several dimensions—pre- and post- program energy consumption, energy audit results, and household demographic char acteristics.

Evaluation of utility residential energy conservation programs: A Pacific Northwest example

Abstract An important and unresolved issue concerning home energy audit programs is the accuracy of the engineering calculations. That is, the accuracy of audit predictions of energy savings for recommended retrofit measures is generally unknown. Data available from a recent evaluation of the Bonneville Power Administration Residential Weatherization Pilot Program allow comparison and analysis of the relationships between actual electricity savings and audit estimates of likely savings. The BPA program offered free home energy audits to identify cost-effective conservation measures to reduce space and water heating electricity use. The program also offered zero-interest, deferred-payment loans for installation of measures recommended during the audit. The actual reduction in annual electricity use averaged 4130 kWh across participant households. The cost of the retrofits that yielded this saving averaged

Elicitation of Natural Language Representations of Uncertainty Using Computer Technology

2100. The median ratio of actual-to-estimated saving is 0.66. Thus, on average, two thirds of the expected saving is actually realized. However, there is substantial variation in this ratio. Actual electricity use increases in the second year for more than 10% of these homes. On the other hand, actual savings are more than double the audit estimates in more than 10% of the homes.