Digital-Twin Attention-Augmented Adaptive Model Predictive Control for Degradation-Aware Residential Battery Energy Management
- 1Department of Mathematics, Physics and Electrical Engineering, Northumbria University, Newcastle upon Tyne NE1 8SA, UK
Abstract
Residential battery energy management systems increasingly rely on model predictive control (MPC), yet most schemes assume a fixed horizon, a noise-free state and an uncalibrated forecaster. This paper proposes DT-AAMPC, an adaptive MPC that combines a Kalman-filtered battery digital twin with an attention-BiLSTM forecaster (TAB-Net) whose attention entropy yields a confidence signal that adapts the controller’s horizon, uncertainty penalty, and state-of-charge back-off online. We evaluate it using corrected code executed on the real dataset at a reduced but realistic validation scale (three seeds, ten training epochs; full protocol: five seeds, forty epochs), against a PID/droop controller, a naive-forecast MPC, and a fixed-horizon ablation (Fixed-DT-MPC) across six household stress-test scenarios. Averaged over eighteen matched blocks, DT-AAMPC lowered net cost to 104.83 currency units (against 149.29, 170.79 and 106.49 for PID, Classical MPC and Fixed-DT-MPC) and cut state-of-health loss to 0.69% (against 1.70%, 2.36% and 0.73%), significant against both classical baselines after Holm–Bonferroni correction (p<0.001); DT-AAMPC was not, however, significantly different from Fixed-DT-MPC on any headline metric (p>0.11), so online adaptation itself is not yet shown to drive the benefit, and self-sufficiency/comfort violation were not significant against either classical baseline. A previously unflagged bug, where ambient temperature was in Fahrenheit, silently treated as Celsius, was found and fixed here; the thermal interlock then never activates across 72 runs, and equivalent-cycle counts fall to a physically unremarkable 0.8, −4.1 per six-day window (against 45–90 previously). Calibration analysis shows the confidence signal is uncorrelated with realised forecast error (ρ=0.047, p=0.568), moderating the confidence-adaptive claims.
Related
- Green synthesis of Ag/MgO nanocomposites from red pepper seed waste for efficient sunlight-driven diclofenac degradation: experimental investigation and EEFO-optimized DT_LSBoost predictive modeling
- Predictive-feedforward energy management for a hybrid agricultural machine operating in mountainous orchards
- Model-Based Assessment of Geogenic Radon Exposure in Chaco Canyon and Adjacent Ancestral Puebloan Residential Settings
- Asthma Intervention with Residential Ventilation and Air Cleaner (AIRVAC) Study: A 4-arm parallel-group randomized controlled trial protocol
- Molecular augmented dynamics: Generating experimentally consistent atomistic structures by design
BibTeX
@article{cavus2026digitaltwin,
title = {{Digital-Twin Attention-Augmented Adaptive Model Predictive Control for Degradation-Aware Residential Battery Energy Management}},
author = {Muhammed Cavus},
journal = {Batteries},
year = {2026},
month = oct,
doi = {10.3390/batteries12100410},
url = {https://www.mdpi.com/2313-0105/12/10/410},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Digital-Twin Attention-Augmented Adaptive Model Predictive Control for Degradation-Aware Residential Battery Energy Management
AU - Muhammed Cavus
PY - 2026
DA - 2026/10/09
JO - Batteries
DO - 10.3390/batteries12100410
UR - https://www.mdpi.com/2313-0105/12/10/410
AB - Residential battery energy management systems increasingly rely on model predictive control (MPC), yet most schemes assume a fixed horizon, a noise-free state and an uncalibrated forecaster. This paper proposes DT-AAMPC, an adaptive MPC that combines a Kalman-filtered battery digital twin with an attention-BiLSTM forecaster (TAB-Net) whose attention entropy yields a confidence signal that adapts the controller’s horizon, uncertainty penalty, and state-of-charge back-off online. We evaluate it using corrected code executed on the real dataset at a reduced but realistic validation scale (three seeds, ten training epochs; full protocol: five seeds, forty epochs), against a PID/droop controller, a naive-forecast MPC, and a fixed-horizon ablation (Fixed-DT-MPC) across six household stress-test scenarios. Averaged over eighteen matched blocks, DT-AAMPC lowered net cost to 104.83 currency units (against 149.29, 170.79 and 106.49 for PID, Classical MPC and Fixed-DT-MPC) and cut state-of-health loss to 0.69% (against 1.70%, 2.36% and 0.73%), significant against both classical baselines after Holm–Bonferroni correction (p<0.001); DT-AAMPC was not, however, significantly different from Fixed-DT-MPC on any headline metric (p>0.11), so online adaptation itself is not yet shown to drive the benefit, and self-sufficiency/comfort violation were not significant against either classical baseline. A previously unflagged bug, where ambient temperature was in Fahrenheit, silently treated as Celsius, was found and fixed here; the thermal interlock then never activates across 72 runs, and equivalent-cycle counts fall to a physically unremarkable 0.8, −4.1 per six-day window (against 45–90 previously). Calibration analysis shows the confidence signal is uncorrelated with realised forecast error (ρ=0.047, p=0.568), moderating the confidence-adaptive claims.
SN - 2313-0105
ER -