Prediction models for dynamic demand response: Requirements, challenges, and insights

Abstract

As Smart Grids move closer to dynamic curtailment programs, Demand Response (DR) events will become necessary not only on fixed time intervals and weekdays predetermined by static policies, but also during changing decision periods and weekends to react to real-time demand signals. Unique challenges arise in this context vis-a-vis demand prediction and curtailment estimation and the transformation of such tasks into an automated, efficient dynamic demand response (D 2 R) process. While existing work has concentrated on increasing the accuracy of prediction models for DR, there is a lack of studies for prediction models for D 2 R, which we address in this paper. Our first contribution is the formal definition of D 2 R, and the description of its challenges and requirements. Our second contribution is a feasibility analysis of very-short-term prediction of electricity consumption for D 2 R over a diverse, large-scale dataset that includes both small residential customers and large buildings. Our third, and major contribution is a set of insights into the predictability of electricity consumption in the context of D 2 R. Specifically, we focus on prediction models that can operate at a very small data granularity (here 15-min intervals), for both weekdays and weekends -all conditions that characterize scenarios for D 2 R. We find that short-term time series and simple averaging models used by Independent Service Operators and utilities achieve superior prediction accuracy. We also observe that workdays are more predictable than weekends and holiday. Also, smaller customers have large variation in consumption and are less predictable than larger buildings. Key implications of our findings are that better models are required for small customers and for non-workdays, both of which are critical for D 2 R. Also, prediction models require just few days' worth of data indicating that small amounts of historical training data can be used to make reliable predictions, simplifying the complexity of big data challenge associated with D 2 R.

Demystifying Fog Computing: Characterizing Architectures, Applications and Abstractions

“…Demand-response (DR) optimization refers to shaping or shifting power demand to match supply capacity [2]. Here, liability and privacy concerns mean that demand-response and load control decisions happen at multiple-levels.…”

Knowledge-infused and consistent Complex Event Processing over real-time and persistent streams

“…Curtailment strategies similarly use static means such as using timeof-use pricing to encourage energy curtailment during historically high-demand periods [1]. But the ability to collect real-time power consumption data from smart meters at the consumer is allowing for dynamic DR decisions, where predictions are done using real-time energy consumption data and curtailment strategies target individual customers with specific usage profiles [6]. However, even such time-series forecasting models need to be adjusted for outlier events that may occur, and curtailment strategies need to be responsive to real-time opportunities.…”

Characterizing application scheduling on edge, fog, and cloud computing resources

“…Smart power grids are another key domain in smart cities, with net‐connected smart meters reporting power demand at households and industries every few minutes to the utility . Smart grid applications, such as demand‐response optimization, help shape or shift power demand using forecasting models on the cloud that trigger curtailment strategies on the edge when a load mismatch is detected . State estimation to determine the health of the distribution network is even more time sensitive, , and needs computing at the edge …”

“…43 Smart grid applications, such as demand-response optimization, help shape or shift power demand using forecasting models on the cloud that trigger curtailment strategies on the edge when a load mismatch is detected. 44,45 State estimation to determine the health of the distribution network is even more time sensitive, (ms), and needs computing at the edge. 46 While edge and fog computing are still emerging technologies, this taxonomy throws more light on these resource abstractions and their effective use from an application and scheduling model, based on current literature and technology.…”