Abstract
Demand Response (DR) schemes are effective tools to maintain a dynamic balance in energy markets with higher integration of fluctuating renewable energy sources. DR schemes can be used to harness residential devices' flexibility and to utilize it to achieve social and financial objectives. However, existing DR schemes suffer from low user participation as they fail at taking into account the users' requirements. First, DR schemes are highly demanding for the users, as users need to provide direct information, e.g. via surveys, on their energy consumption preferences. Second, the user utility models based on these surveys are hard-coded and do not adapt over time. Third, the existing scheduling techniques require the users to input their energy requirements on a daily basis. As an alternative, this paper proposes a DR scheme for user-oriented direct load-control of residential appliances operations. Instead of relying on user surveys to evaluate the user utility, we propose an online data-driven approach for estimating user utility functions, purely based on available load consumption data, that adaptively models the users' preference over time. Our scheme is based on a day-ahead scheduling technique that transparently prescribes the users with optimal device operation schedules that take into account both financial benefits and user-perceived quality of service. To model day-ahead user energy demand and flexibility, we propose a probabilistic approach for generating flexibility models under uncertainty. Results on both real-world and simulated datasets show that our DR scheme can provide significant financial benefits while preserving the user-perceived quality of service.
Introduction
Integrating renewable energy sources into the power grid necessitates robust Demand Response (DR) mechanisms to balance supply and demand. While residential devices offer significant flexibility, current DR programs often fail due to "User Response Fatigue." Traditional approaches require intrusive surveys or daily manual scheduling, leading to high attrition rates. Furthermore, static models fail to adapt to evolving household habits. This research presents an automated, data-driven DR scheme that learns user preferences from consumption data, minimizing user burden while maximizing grid benefits.
Methodology
The proposed system focuses on "wet devices" like dishwashers and washing machines through a multi-stage pipeline:
1. Activity Prediction and Flex-Offers
The system first predicts device usage using Naive Bayesian and Linear Regression models. To handle the inherent uncertainty in these predictions, the authors introduce Probabilistic Flex-Offers. Unlike standard models with fixed windows, these use random variables to define "ready" and "latest start" times.
2. Adaptive Utility Modeling
The core innovation lies in modeling user flexibility as an exponential distribution. The system assumes the probability of a user accepting a delay decreases as the duration increases. Using Stochastic Gradient Descent (SGD), the model updates its flexibility parameter ($\lambda$) based on passive feedback—specifically, whether a user accepts a schedule or manually overrides it.
3. Optimal Scheduling
The scheduler identifies start times that maximize Expected User Utility, balancing financial incentives from Spot and Regulation markets against the predicted likelihood of user acceptance.
Experimental Evaluation
The scheme was tested using the Intrepid Dataset (real-world data from 11 households) and the Genetx Dataset (synthetic data for lifestyle robustness). Market evaluation utilized Danish energy market (DK1) pricing.
A critical aspect of the evaluation was understanding prediction accuracy. As shown in the duration error analysis, while most predictions are accurate, the system must be robust to significant outliers that can occur in real-world settings.
Key Results and Insights
- Financial Performance: The system achieved up to 32% savings in Spot Markets and 38% in Regulation Markets for real-world data. Synthetic tests showed savings up to 42%.
- User Acceptance: By learning from feedback, the adaptive model improved acceptance rates from 36% (uniform model) to 64% on real-world data.
- Passive Interaction: The "passive" feedback loop successfully reduces cognitive load, addressing the primary cause of user fatigue.
- Uncertainty Resilience: The probabilistic approach allows the scheduler to absorb prediction errors within the flexibility window without impacting the user experience.
Conclusion
This research demonstrates that data-driven, adaptive DR schemes can effectively harness residential flexibility without requiring high user effort. Future work will expand this model to HVAC and EV systems, which involve different flexibility constraints like temperature ranges, and explore multi-device coordination to prevent localized grid peaks.
Summary updated