国产bbaaaaa片,成年美女黄网站色视频免费,成年黄大片,а天堂中文最新一区二区三区,成人精品视频一区二区三区尤物

首頁> 外國專利> MULTI-AGENT DEEP REINFORCEMENT LEARNING FOR DYNAMICALLY CONTROLLING ELECTRICAL EQUIPMENT IN BUILDINGS

MULTI-AGENT DEEP REINFORCEMENT LEARNING FOR DYNAMICALLY CONTROLLING ELECTRICAL EQUIPMENT IN BUILDINGS

機譯:多功能深度加固,用于在建筑物中動態(tài)控制電氣設(shè)備的深度鋼筋學(xué)習

摘要

Reinforcement Learning agent interacting with a real-world building to determine optimal policy may not be viable due to comfort constraints. Embodiments of the present disclosure provide multi-deep agent RL for dynamically controlling electrical equipment in buildings, wherein a simulation model is generated using design specification of (i) controllable electrical equipment (or subsystem) and (ii) building. Each RL agent is trained using simulation model and deployed in the subsystem. Reward function for each subsystem includes some portion of reward from other subsystem(s). Based on reward function of each RL agent, each RL agent learns an optimal control parameter during execution of RL agent in subsystem. Further, a global optimal control parameter list is generated using the optimal control parameter. The control parameters in the global optimal control parameters list are fine-tuned to improve subsystem's performance. Information on fine-tuning parameters of the subsystem and reward function are used for training RL agents.
機譯:由于舒適的限制,加固學(xué)習代理與真實建筑物的互動以確定最佳政策可能不會是可行的。本公開的實施例提供了用于在建筑物中動態(tài)控制電氣設(shè)備的多深代理R1,其中使用(i)可控電氣設(shè)備(或子系統(tǒng))和(ii)建筑物的設(shè)計規(guī)范產(chǎn)生仿真模型。每個RL代理使用仿真模型進行培訓(xùn)并在子系統(tǒng)中部署。每個子系統(tǒng)的獎勵函數(shù)包括來自其他子系統(tǒng)的一些獎勵?;诿總€RL代理的獎勵功能,每個RL代理在子系統(tǒng)中執(zhí)行RL代理期間的最佳控制參數(shù)。此外,使用最優(yōu)控制參數(shù)生成全局最佳控制參數(shù)列表。全局最優(yōu)控制參數(shù)列表中的控制參數(shù)是微調(diào)的,以提高子系統(tǒng)的性能。有關(guān)子系統(tǒng)和獎勵功能的微調(diào)參數(shù)的信息用于培訓(xùn)RL代理。

著錄項

相似文獻

  • 專利
  • 外文文獻
  • 中文文獻
獲取專利

客服郵箱:kefu@zhangqiaokeyan.com

京公網(wǎng)安備:11010802029741號 ICP備案號:京ICP備15016152號-6 六維聯(lián)合信息科技 (北京) 有限公司?版權(quán)所有
  • 客服微信

  • 服務(wù)號