Hybrid Customer Baseline Load Estimator for Small and Medium Enterprises
Ph.D. Student: Gururaghav Raman | Research Fellow: Kong Kaonan | Advisor: Jimmy C.-H. Peng
Project Duration: 2018
The success of demand response (DR) programs, and in particular, incentive-based services, is subject to accurately estimating the customer baseline load (CBL). An unbiased CBL estimate allows for fair compensation of a DR participant for their service. A CBL estimate that has a bias favourable to the utility may deter consumers from participating enthusiastically in the program, while a bias against the utility may nullify the benefit for the utility in creating a DR program in the first place.
While conventional CBL estimation methods employed for industries are primarily based on day-matching and control-groups, the challenge lies in refining these to robustly handle the higher demand variations exhibited by small and medium enterprises (SMEs).
For this, we propose an improved day-matching technique, where, for each DR event, we adaptively select a customer-group using similarity theory. The search space here is two-dimensional, along the temporal dimension and the entire customer population. Subsequently, day-matching is performed on the selected group to minimize any biases.
A comparative study of the proposed technique against conventional CBL estimators is conducted using data from the Irish Commission for Energy Regulation (CER). While all methods show similar performances for low demand variances, the proposed similarity-based method is distinctly superior when the variance becomes higher.
Figure 1. Electricity demand of an SME customer over five working days. The high variability in demand is clearly visible.
Figure 2. Comparative performance analysis of the proposed similarity-based CBL estimator vis-à-vis conventionally used day-matching techniques (BL 1-6). Even when the variance is high, the proposed estimator produces the most accurate baseline estimate.
Finally, an empirical expression is derived in this paper for utilities to approximate CBL estimation errors for a given customer based on their inherent variability in their demands. This expression allows the utility to assess the suitability of the proposed CBL estimation method for new or returning customers.
The proposed CBL estimator would therefore be instrumental for utilities to expand the penetration of DR services among SMEs.
Figure 3. Proposed adequacy checking mechanism showing the expected mean absolute percentage error (MAPE) in the CBL estimate for different variances in the consumptions of a given consumer. The red dashed line represents an estimate using the least squares technique, while the solid black line is derived from the proposed weighted least squares technique.