Constraint Audited Receding Horizon Dispatch for Curtailment Recovery in Export Capped Utility Scale Photovoltaic Plants With Battery Energy Storage

Hard point-of-interconnection (POI) export limits can strand otherwise-available photovoltaic (PV) energy unless battery storage is dispatched with sufficient anticipation and under physically consistent constraints.This study evaluates a constraint-audited receding-horizon dispatch framework for curtailment recovery at an anonymized 230 MW utility-scale PV plant with a modeled 195.5 MW export cap and a nominal 30 MW/140 MWh battery energy storage system (BESS).The mixed-integer model predictive controller enforces BESS power and energy limits, charge-discharge exclusivity, no grid charging, forecast-dependent discharge headroom, first-action receding-horizon execution, and a common 50%-to-50% terminal-energy protocol.Twelve rolling monthly folds from September 2024 to August 2025 are evaluated against cap-only operation and a protocol-matched perfect-information Oracle.Cap-only operation curtails 26,279 MWh (4.9872% of evaluated PV energy).A transparent 72 h Diurnal model predictive control (MPC) reduces curtailment to 11,719.454MWh, recovers 14,559.546MWh, and closes 98.450% of the modeled cap-only-to-Oracle recoverable-energy opportunity; the matched Oracle curtails 11,490.175MWh.All 3 h closed-loop trajectories satisfy the modeled export cap, BESS limits, terminal equality, charge-discharge exclusivity, discharge-headroom rule, and no-grid-import condition.A matched information-sensitivity analysis then compares History-CatBoost and Weather-CatBoost under the same controller.Weather-CatBoost lowers daylight mean absolute error (MAE) by 2.225 MW at 24 h and 1.175 MW at 72 h relative to History-CatBoost, while annual curtailment is 288.601MWh higher at 24 h and 99.699 MWh lower at 72 h; the monthly block-bootstrap intervals include zero in both cases.Histogram gradient boosting (HGB) robustness and BESS-sensitivity results further show that the incremental control value of weather information is model-and configuration-dependent.Under the tested plant, cap, BESS, 3 h resolution, horizon, and terminal protocol, physically constrained anticipatory dispatch captures most of the modeled recoverable curtailed energy, leaving comparatively limited residual opportunity for additional forecast sophistication.

Authors

Institutions

Publication Details

Journal
Power Engineering and Engineering Thermophysics
Published
2026-10-29
DOI
https://doi.org/10.56578/peet050404
Primary Topic
Electric Power System Optimization
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Constraint Audited Receding Horizon Dispatch for Curtailment Recovery in Export Capped Utility Scale Photovoltaic Plants With Battery Energy Storage

Ahmed Jendeya, Ikram S. Babazade
2 links
Power Engineering and Engineering Thermophysics
Electric Power System Optimization
article

Constraint Audited Receding Horizon Dispatch for Curtailment Recovery in Export Capped Utility Scale Photovoltaic Plants With Battery Energy Storage

Ahmed Jendeya, Ikram S. Babazade
article en

Abstract

Hard point-of-interconnection (POI) export limits can strand otherwise-available photovoltaic (PV) energy unless battery storage is dispatched with sufficient anticipation and under physically consistent constraints.This study evaluates a constraint-audited receding-horizon dispatch framework for curtailment recovery at an anonymized 230 MW utility-scale PV plant with a modeled 195.5 MW export cap and a nominal 30 MW/140 MWh battery energy storage system (BESS).The mixed-integer model predictive controller enforces BESS power and energy limits, charge-discharge exclusivity, no grid charging, forecast-dependent discharge headroom, first-action receding-horizon execution, and a common 50%-to-50% terminal-energy protocol.Twelve rolling monthly folds from September 2024 to August 2025 are evaluated against cap-only operation and a protocol-matched perfect-information Oracle.Cap-only operation curtails 26,279 MWh (4.9872% of evaluated PV energy).A transparent 72 h Diurnal model predictive control (MPC) reduces curtailment to 11,719.454MWh, recovers 14,559.546MWh, and closes 98.450% of the modeled cap-only-to-Oracle recoverable-energy opportunity; the matched Oracle curtails 11,490.175MWh.All 3 h closed-loop trajectories satisfy the modeled export cap, BESS limits, terminal equality, charge-discharge exclusivity, discharge-headroom rule, and no-grid-import condition.A matched information-sensitivity analysis then compares History-CatBoost and Weather-CatBoost under the same controller.Weather-CatBoost lowers daylight mean absolute error (MAE) by 2.225 MW at 24 h and 1.175 MW at 72 h relative to History-CatBoost, while annual curtailment is 288.601MWh higher at 24 h and 99.699 MWh lower at 72 h; the monthly block-bootstrap intervals include zero in both cases.Histogram gradient boosting (HGB) robustness and BESS-sensitivity results further show that the incremental control value of weather information is model-and configuration-dependent.Under the tested plant, cap, BESS, 3 h resolution, horizon, and terminal protocol, physically constrained anticipatory dispatch captures most of the modeled recoverable curtailed energy, leaving comparatively limited residual opportunity for additional forecast sophistication.

Power Engineering and Engineering ThermophysicsVol. 5(4)
Azerbaijan State Oil and Industry University (AZ), Baku Engineering University (AZ)
Affordable and clean energy, Industry, innovation and infrastructure
Openalex Percentile: Top 24%
Electric Power System Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon