ABSTRACT Demand response programmes (DRPs) increasingly rely on accurate load forecasting, realistic modelling of consumer participation, and strong resilience against data manipulation. This paper presents an integrated cyber‐secure DRP design framework that jointly addresses these requirements. A hybrid convolutional–bidirectional long short‐term memory (CB‐LSTM) model is first employed to predict consumption behaviour with improved accuracy, supporting consumer categorisation into low‐, medium‐, and high‐usage groups. Building on these predictions, a comprehensive DRP portfolio incorporating TOU and DLC‐based schemes is dynamically assigned to each category. To capture behavioural uncertainty, consumer participation is modelled using a Z‐number possibilistic–probabilistic formulation, enabling more reliable estimation of engagement levels and enhancing the fairness of DRP allocation. A key contribution of this study is the introduction of a CB‐LSTM‐assisted deviation bound‐based detection/correction (CBLADBDC) mechanism to mitigate false data injection (FDI) attacks. The proposed method exhibits strong detection capability, achieving a 98.0% recall and maintaining a low false positive rate of 1.1%, thereby preserving the integrity of load profiles and preventing unnecessary incentive payments. Simulation results demonstrate that the integrated framework reduces peak demand by 24%–27% while maintaining cost‐efficient DRP performance.
Zarandi et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: