Impacts of Industrial Baseline Errors on Costs and Social Welfare in the Demand Side Management of Day-Ahead Wholesale Markets

Abstract

Demand Side Management (DSM) has been recognized for its potential to counteract the intermittent nature of renewable energy, increase system efficiency, and reduce system costs. While the popular approach among academia adopts a social welfare maximization formulation, the industrial practice in the United States electricity market compensates customers according to their load reduction from a predefined electricity consumption baseline that would have occurred without DSM. This paper is an extension of a previous paper studying the differences between the industrial & academic approach to dispatching demands. In the previous paper, the comparison of the two models showed that while the social welfare model uses a stochastic net load composed of two terms, the industrial DSM model uses a stochastic net load composed of three terms including the additional baseline term. That work showed that the academic and industrial optimization method have the same dispatch result in the absence of baseline errors given the proper reconciliation of their respective cost functions. DSM participants, however, and very much unfortunately, are likely to manipulate the baseline in order to receive greater financial compensation. This paper now seeks to study the impacts of erroneous industrial baselines in a day-ahead wholesale market context. Using the same system configuration and mathematical formalism, the industrial model is compared to the social welfare model. The erroneous baseline is shown to result in a different and more importantly costlier dispatch. It is also likely to require more control activity in subsequent layers of enterprise control. Thus an erroneous baseline is likely to increase system costs and overestimate the potential for social welfare improvements. NOMENCLATURE

Demand side management in a day-ahead wholesale market: A comparison of industrial & social welfare approaches

“…In Hour 5 & 18, the industrial model shows that the virtual generation participate in maintaining a relatively constant dispatchable generation level. This is because the model assumes higher DSM participation than actually exists [85]. However, the dispatched level may not always be achievable, thus requiring more subsequent control [85].…”

“…This is because the model assumes higher DSM participation than actually exists [85]. However, the dispatched level may not always be achievable, thus requiring more subsequent control [85].…”

“…For example, in the case that the virtual generators are dispatched between 0% and 20% of the baseline, then they are being dispatched to demand more than the original load forecast or correct baseline value. This yields a higher social welfare value but does not have a basis in reality [85]. Fig.…”

“…Furthermore, this work assumes this error-free baseline is equal to the maximum capacity of the dispatchable demand unit in the social welfare model.P DCj ¼ P DCj . The erroneous baseline was set to 120% of its true value to emphasize its impact [85]. This has the implicit effect of allowing demand units to have a maximum load reduction (capacity) of 120% as that found in the social welfare model.…”

“…For a day-ahead scheduling scenario, a unit commitment problem incorporating DSM has been studied simply from the suppliers side to minimize generation costs [77] or to maximize electric power utility profit [78,79], and from a markets perspective using either the social welfare [80] and industrial method [81][82][83]. This paper aims to rigorously compare the social welfare and the industrial load reduction approaches and study the effects of an erroneous industrial baseline in a day-ahead wholesale market context using the same system configuration and mathematical formalism [84,85].…”