In the name of of Allah the Merciful

دانلود کتاب مدیریت انرژی مبتنی بر یادگیری تقویتی عمیق برای وسایل نقلیه الکتریکی هیبریدی

Deep Reinforcement Learning-Based Energy Management for Hybrid Electric Vehicles, Li Yeuching, Hongwen He, 1636393039, 1636393012, 978-1636393032, 9781636393032, 978-1636393018, 9781636393018, 9781636393025

80,000 Toman
The desired product is not available.

سال انتشار: 2022
زبان کتاب: انگلیسی
تعداد صفحات: 135
فرمت فایل: PDF
حجم فایل: ۶ مگابایت

  • {{value}}
کمی صبر کنید...

(Synthesis Lectures on Advances in Automotive Technology)

The urgent need for vehicle electrification and improvement in fuel  efficiency has gained increasing attention worldwide. Regarding this  concern, the solution of hybrid vehicle systems has proven its value  from academic research and industry applications, where energy  management plays a key role in taking full advantage of hybrid electric  vehicles (HEVs). There are many well-established energy management  approaches, ranging from rules-based strategies to optimization-based  methods, that can provide diverse options to achieve higher fuel economy  performance. However, the research scope for energy management is still  expanding with the development of intelligent transportation systems  and the improvement in onboard sensing and computing resources. Owing to  the boom in machine learning, especially deep learning and deep  reinforcement learning (DRL), research on learning-based energy  management strategies (EMSs) is gradually gaining more momentum. They  have shown great promise in not only being capable of dealing with big  data, but also in generalizing previously learned rules to new scenarios  without complex manually tunning
Focusing on learning-based  energy management with DRL as the core, this book begins with an  introduction to the background of DRL in HEV energy management. The  strengths and limitations of typical DRL-based EMSs are identified  according to the types of state space and action space in energy  management. Accordingly, value-based, policy gradient-based, and hybrid  action space-oriented energy management methods via DRL are discussed,  respectively. Finally, a general online integration scheme for DRL-based  EMS is described to bridge the gap between strategy learning in the  simulator and strategy deployment on the vehicle controller