Hierarchical Q-learning network for online simultaneous optimization of energy efficiency and battery life of the battery/ultracapacitor electric vehicle

Xu, Bin; Zhou, Quan; Shi, Junzhe; Li, Sixu

doi:10.1016/j.est.2021.103925

Cited by 21 publications

(10 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Advantages Disadvantages [158] Q-learning -8% Reduction of battery capacity loss -Longer battery life and improving range of the vehicle -Capable of handling various drive cycles and measuring noises -Adaptable numerous hybrid power systems -Further state variables must be taken into account -Invalidated for some experimental information from models that were introduced from outside sources [159] Q-learning -Enhanced dynamic efficiency -Decreased fuel use and calculation time -The performance of the acquired power response is similar to that of the current DP-based technique [160] Q-learning -Fuel consumption and power fluctuation were reduced by 5.59% and 13%, respectively -Convergence speed has increased by 69% -Real-time applications are required [164] DQN -Effective use of acquired knowledge -Increased computational speed -Boosted fuel economy -Self-reliance of prior information in driving cycles…”

Section: References Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Energy management strategies of hybrid electric vehicles: A comparative review

Azim Mohseni,

Bayati,

Ebel

2023

IET Smart Grid

View full text Add to dashboard Cite

With increasing environmental problems, including air pollution, rising greenhouse gases and global warming, the need to use clean energy resources seems essential. Therefore, the usage of clean and renewable energy sources is suggested as a suitable solution to overcome the mentioned concerns. The transportation section contributes to a huge percentage of energy consumption. Hybrid electric vehicles (HEVs), by combining several energy resources, are considered as a crucial solution to decrease fossil fuel consumption and improve the environmental challenges. The existence of an alternative energy resource and the internal combustion engine together provides optimal power distribution among them to maximise power usage and minimise fuel consumption. Energy management strategy (EMS) is an essential challenge in HEV's design procedure to deal with the power distribution in multiple power source systems to improve the performance of the HEVs. A review of various EMSs for HEVs, followed by an analysis of each type, including its benefits and drawbacks, is presented by the authors. In addition to this, the major challenges in EMSs for HEVs are described and a comprehensive review is presented for strategies addressing these issues.

show abstract

Section: References Methodsmentioning

confidence: 99%

“…Ref. [158] aimed to introduce a Q‐learning approach to optimise the supervisory management system of an EV with a mixed charging system of a battery and UC. To allocate two control levels, a hierarchical Q‐learning network composed of two independent Q tables was developed.…”

Section: Energy Management Strategies For Hevsmentioning

confidence: 99%

Energy management strategies of hybrid electric vehicles: A comparative review

Azim Mohseni,

Bayati,

Ebel

2023

IET Smart Grid

View full text Add to dashboard Cite

show abstract

“…The easiest way to model an analogous circuit for an ultracapacitor is with a series resistor and capacitor (RC) circuit. This model comprises a capacitor that represents the charge capacity during the charging and discharging of the ultracapacitor and a resistor that represents the internal resistance [23]. The output current and voltage from the ultracapacitor are calculated as follows;…”

Section: Modelling Of Ultra-capacitormentioning

confidence: 99%

“…The objective function is to minimize the charging cost is find using equation (23). The charging cost of each EV is calculated using below equation;…”

Section: Initializationmentioning

confidence: 99%

Fuzzy-based pelican optimization for battery management and energy scheduling in solar aided electric vehicle

Sharma,

Mishra

2024

Eng. Res. Express

View full text Add to dashboard Cite

Electric vehicles (EVs) are considered an essential mode of transportation due to their advantages, non-polluting and noise-free characteristics. The batteries are considered the primary power source for electric vehicles, ensuring energy supply and thus improving the user experience. However, designing charging stations and uncontrolled charge scheduling are the barriers evaluated in charging stations. To solve these issues, a fuzzy-based pelican optimization algorithm (POA) is proposed in this work. POA schedules the EVs in the charging station. Demand in the charging station is based on the state of charge (SOC) of the EVs arriving at the charging station. Solar power with battery and ultracapacitor is used as an input source for charging stations. The energy management in the system is based on fuzzy to satisfy the load demand from PV and energy storage devices. This proposed method is implemented in the Simulink tool to evaluate the performance. The proposed method is used to do optimal scheduling of EVs with proper energy management. The proposed algorithm has a high fitness range when compared to the existing algorithm.

show abstract

“…However, the stochastic dynamic programming algorithm approach yields a control that depends on a particular state [12][13][14], while the deterministic dynamic algorithm solves the optimization problem by sequentially calculating each state at each time step in a backward order. This work makes two key advances compared to previous approaches.…”

Section: Literature Reviewmentioning

confidence: 99%

Heuristic Adaptive Dynamic Programming-based Energy Optimization Strategies for Hybrid Electric Vehicles

2024

Teh. vjesn.

View full text Add to dashboard Cite

This study suggests a more accurate series-parallel hybrid electric vehicle (HEV) energy management control technique based on heuristic adaptive dynamic programming (HADP). This article provides a solution to the energy management problem of Electric Vehicle using the Stimulant Dependent Heuristic Adaptive Dynamic Programming (SDHADP) algorithm. The reinforced Q learning model along with the three-layer feed-forward neural network with backpropagation is detailed with the necessary diagrams and equations. The approach to connecting the battery of a BEV to a residential system is novel and innovative. The experimental results highlight the importance and effectiveness of the SDHADP algorithm using an electric vehicle. This is done with the help of a BV feed-forward neural network with back-propagation. In the model, which is in charge of controlling the system based on the training given by the BV network, Markov decision theory was adopted by scholars as a paradigm for posing and solving planning problems in the face of uncertainty and is in charge of the battery management system.The suggested technique uses online learning control methods to improve upon the efficiency of the traditional Q-learning method by more than 4.55% under the specified real-world driving circumstances.

show abstract

Hierarchical Q-learning network for online simultaneous optimization of energy efficiency and battery life of the battery/ultracapacitor electric vehicle

Cited by 21 publications

References 31 publications

Energy management strategies of hybrid electric vehicles: A comparative review

Energy management strategies of hybrid electric vehicles: A comparative review

Fuzzy-based pelican optimization for battery management and energy scheduling in solar aided electric vehicle

Heuristic Adaptive Dynamic Programming-based Energy Optimization Strategies for Hybrid Electric Vehicles

Contact Info

Product

Resources

About