reinforcement learning

A multi-agent deep reinforcement learning framework for coordinated urban traffic signal and parking lot exit control

A hybrid physics-aware multi-objective analog integrated circuit design framework with deep reinforcement learning

Physics-Informed Hybrid Machine Learning Model for Carbonation Depth Prediction in Concrete through Residual Correction and Variogram Analysis of Response Surfaces

A hybrid residual correction framework that integrates a physics-based carbonation model with a stacked ensemble of machine learning algorithms: gradient boosted regression trees (GBRT), support vector regression (SVR), and Gaussian process regression (GPR), combined through an XGBoost metamodel, demonstrating that the residual-based metamodel reproduced observed carbonation depths with higher accuracy.

Reinforcement learning-based parametric generative model for underground road horizontal alignment design considering construction risk

A hierarchical multiscale reinforcement learning framework for energy management in hybrid electric vehicles considering powertrain response coordination

Adaptive optimal control via event-triggered reinforcement learning for switched systems with unknown actuation gains

Stability-aware multi-objective federated learning via deep reinforcement learning: Client selection and resource optimization in dynamic edge networks

Physics-informed reinforcement learning for energy-efficient minute-scale control of split-type air-conditioning systems: Experimental validation on real hardware

Deep reinforcement learning for adaptive rack repositioning and pod sequencing in robotic mobile fulfillment systems via multi-fidelity simulation optimization

Extension-layer fine-tuning for transfer learning in multi-agent reinforcement learning-based HVAC control

An SAC-Based Auto Optimization Model for Dynamic Current Balancing of Multichip Paralleled SiC Power Module

The dynamic current imbalance between the paralleled SiC mosfets in multichip power modules, which is commonly attributed to the asymmetric module layout, severely limits their current capacity and thermal reliability. Adjusting the connection points of bonding wires is an effective method to mitigate imbalanced dynamic current. However, manual trial-and-error is currently the most common method for optimizing connection points, which is both deficient and inefficient. Existing automated solutions usually rely on a prefitting process based on large datasets, which is time-consuming and impractical for high-dimensional parameter applications. Thus, this article proposes an optimization model to mitigate dynamic current imbalance, which can automatically adjust the connection points of bonding wires without any manual intervention. The reinforcement learning (RL) soft actor-critic algorithm was applied to the power module optimization, eliminating the need for prefitting and enabling high-dimensional parameter optimization. After optimization, nearly complete dynamic current balancing in both high-side and low-side switches in a multichip-paralleled half-bridge power module is achieved, as verified by simulations and experiments. This model achieves true dynamic current balancing automation for the first time, providing an important reference for the application of RL to the automated optimization of multichip power modules.

3D reconstruction of spatial transcriptomics data with dual incremental deep reinforcement learning