An Adaptive Demand Response Framework using Price Elasticity Model in Distribution Networks: A Case Study

TL;DR

This paper introduces an adaptive demand response framework using a dynamic price elasticity model, incorporating stochastic and deterministic methods to better capture customer load flexibility and improve demand management in distribution networks.

Contribution

It presents a novel adaptive economic DR framework with dynamic elasticity modeling, including load recovery and intertemporal constraints, validated through a case study on IEEE distribution data.

Findings

The proposed model effectively captures load flexibility and customer sensitivity.

Stochastic approach with geometric Brownian motion enhances load recovery modeling.

The framework improves demand response performance compared to existing models.

Abstract

Price elasticity model (PEM) is an appealing and modest model for assessing the potential of flexible demand in DR. It measures the customers demand sensitivity through elasticity in relation to price variation. However, application of PEM in DR is partially apprehensible on attributing the adaptability and adjustability with intertemporal constraints in DR. Thus, this article presents an adaptive economic DR framework with attributes of DR via a dynamic elasticity approach to model customers sensitivity. This dynamic elasticity is modeled through the deterministic and stochastic approaches. Both approaches envision the notion of load recovery for shiftable/flexible loads to make the proposed DR framework adaptive and adjustable relative to price variation. In stochastic approach, a geometric Brownian motion is employed to emulate load recovery with inclusion of intertemporal constraint of load flexibility. The proposed mathematical model shows what should be the customers elasticity value to achieve the factual DR. The case study is carried out on standard IEEE 33 distribution system bus load data to assess technical and socio-economic impact of DR on customers and is also compared with the exiting model.