A Bilevel Programming Framework for Demand Response Incentive Design with Non-Intrusive Load Monitoring-Based Flexibility Estimation.
Saved in:
| Title: | A Bilevel Programming Framework for Demand Response Incentive Design with Non-Intrusive Load Monitoring-Based Flexibility Estimation. |
|---|---|
| Authors: | Ding, Ye1 (AUTHOR), Zhou, Kai1,2 (AUTHOR) jhtechnical@126.com, He, Xiuming3 (AUTHOR), Sun, Yuan1 (AUTHOR) yuansun@suda.edu.cn |
| Source: | Energies (19961073). Jun2026, Vol. 19 Issue 12, p2818. 21p. |
| Subject Terms: | *Bilevel programming, *Energy demand management, *Incentive (Psychology), *Mixed integer linear programming, *Hidden Markov models, *Marginal pricing |
| Abstract: | Demand response (DR) plays a key role in enhancing power system flexibility under increasing renewable penetration, yet most existing approaches rely on aggregate demand models that fail to capture appliance-level heterogeneity. A bilevel programming framework for DR incentive design incorporating non-intrusive load monitoring (NILM)-based flexibility estimation is proposed. A conditional factorial hidden Markov model (CFHMM) is used to disaggregate smart meter data and recover appliance-level consumption patterns, which are then mapped to willingness-to-accept (WTA) values to construct device-informed DR potential functions. These estimates are embedded in a bilevel optimization model, where a retailer determines optimal incentives while accounting for the endogenous impact of demand response on locational marginal prices through market clearing. The model is reformulated as a single-level mixed-integer linear program using Karush–Kuhn–Tucker (KKT) conditions. Case studies using real-world data and the IEEE test system show that the proposed framework produces more effective incentive strategies than aggregate DR modeling, leading to improved DR utilization and higher retailer profitability. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Demand response (DR) plays a key role in enhancing power system flexibility under increasing renewable penetration, yet most existing approaches rely on aggregate demand models that fail to capture appliance-level heterogeneity. A bilevel programming framework for DR incentive design incorporating non-intrusive load monitoring (NILM)-based flexibility estimation is proposed. A conditional factorial hidden Markov model (CFHMM) is used to disaggregate smart meter data and recover appliance-level consumption patterns, which are then mapped to willingness-to-accept (WTA) values to construct device-informed DR potential functions. These estimates are embedded in a bilevel optimization model, where a retailer determines optimal incentives while accounting for the endogenous impact of demand response on locational marginal prices through market clearing. The model is reformulated as a single-level mixed-integer linear program using Karush–Kuhn–Tucker (KKT) conditions. Case studies using real-world data and the IEEE test system show that the proposed framework produces more effective incentive strategies than aggregate DR modeling, leading to improved DR utilization and higher retailer profitability. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 19961073 |
| DOI: | 10.3390/en19122818 |
