METHOD AND SYSTEM FOR CHARGE DISAGGREGATION
A deep neural network-based load disaggregation system with a parallel architecture addresses the limitations of HMM by efficiently detecting and estimating the power consumption of multiple electrical devices, enhancing smart grid management and user awareness.
FR3170074A1Pending Publication Date: 2026-06-19SAGEMCOM ENERGY & TELECOM SAS
Patent Information
- Authority / Receiving Office
- FR · FR
- Patent Type
- Applications
- Current Assignee / Owner
- SAGEMCOM ENERGY & TELECOM SAS
- Filing Date
- 2024-12-18
- Publication Date
- 2026-06-19
Smart Images
Figure 00000000_0000_ABST
Abstract
A load disaggregation method comprising an operational phase which, in each computational branch i (Bi) among M computational branches executed in parallel and each configured for a type i of electrical appliance among M types, with M ≥ 2 and 0 < i ≤ M, performs: a first classification (401-i) to determine whether an electrical appliance of type i is present on the household electrical network; if a type i electrical appliance is found, performs a second classification to generate an estimated value of an operating state of the type i electrical appliance, and then an estimate of the individual power consumption of the type i electrical appliance. Then concatenates (404), in an output vector, M estimated values of individual power consumption associated with the M computational branches. Figure to be published with the abstract: Fig. 4
Need to check novelty before this filing date? Find Prior Art