arrow_back Retour aux freelances
F

Offline Q-Learning Controller for Hybrid PV–Battery System in MATLAB/Simulink

Freelancer

Partager:
placeUS home_workTélétravail assignmentCDD publicOffre agrégée · US

eventPubliée le 30 août 2026 · verifiedNous avons confirmé le 30 août 2026 qu'elle est toujours active

US$ 50 – US$ 100 par projet

À propos de l'offre

Design and implement an offline tabular Q-learning (Q-table) controller in MATLAB/Simulink for a hybrid PV–battery system feeding a buck converter. The controller should accurately regulate the converter’s output voltage under varying irradiance, load, and battery SOC. The project will be completed in two phases: 1. Develop a complete Simulink model of the hybrid PV–battery–buck system, generate an offline dataset of transitions, train the Q-table, and implement a MATLAB Function block for closed-loop voltage regulation using the greedy Q-table controller. 2. Enhance the controller with physics-informed reinforcement learning by incorporating buck converter equations into the reward function and state features. Demonstrate voltage regulation under varying conditions and provide brief documentation covering encoding, reward design, and usage. Deliverables: - Complete Simulink model of the hybrid PV–battery–buck system. - Offline dataset of transitions and trained Q-table. - MATLAB Function block implementing the greedy Q-table controller. - Physics-informed reward/state features based on buck equations. - Demonstration of voltage regulation under varying conditions. - Brief documentation of encoding, reward design, and usage.

Continuez à lire gratuitement

Créez un compte gratuit pour voir l'offre complète et postuler.

  • badgePortfolio visible par les entreprises
  • notificationsAlerte de nouvelle offre par e-mail
  • favoriteToujours gratuit, sans piège