Work

A smart building directrix load formulation method for smart building clusters based on directional aggregation mechanism

Year: 2026

Type: article

Abstract: Accurate formulation of customer directrix loads for smart buildings is an important foundation for guiding smart buildings to exert their demand response potential. However, due to the diversity of types and time coupling characteristics of adjustable resources, existing customer directrix load(CDL) formulation methods do not consider the regulatory characteristics of adjustable loads themselves, resulting in a low response rate. To this end, this paper proposes a customer directrix load formulation method for smart building clusters based on a directional aggregation mechanism, describing the problem of customer directrix load formulation for smart building clusters as a mixed integer linear programming problem under the constraints of cluster response capacity, and realizing the directional aggregation of response resources in the cluster by embedding specific aggregation objectives. First, to accurately construct the directional aggregation target curve of the smart building cluster, we obtain typical cluster load curves through multi-feature extraction and clustering algorithms, predict the response demand curve using the LSTM neural network, and then superimpose the two to obtain the ideal customer directrix load. Then, taking the ideal customer directrix load as the aggregation direction, we use the feasible region projection method to directionally project the high-dimensional time-series operation space inside the cluster into the external feasible region of the cluster. Finally, taking the boundary of the cluster's external feasible region as the constraint, the cluster directrix load generation and decomposition optimization model that considers the cluster response capability is constructed to achieve cluster directrix load formulation and refined decomposition into differentiated smart building load types. Case studies show that compared with traditional methods, the proposed method can effectively improve the aggregation effect of the cluster's response capacity and ensure the accessibility of the load cluster directrix load.

Source: Energy Reports

DOI: 10.1016/j.egyr.2026.109655

Authors: Xudong Lei, Jianjun He, Zhantao Fan, Yuxin Lin

Language: English

Topic: Smart Grid Energy Management

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

SDGs: Sustainable cities and communities, Industry, innovation and infrastructure, Affordable and clean energy

Open Access status: gold

Estimated APC paid: $3,040

Funder: China Southern Power Grid