Double Deep Reinforcement Learning-Based Energy Management for a Parallel Hybrid Electric Vehicle With Engine Start–Stop Strategy

Chen, Jiaxin; Pu, Huayan; Li, Teng; Khajepour, Amir

doi:10.1109/tte.2021.3101470

Cited by 81 publications

(37 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The design of the reward significantly affects the optimization results. To achieve the optimization objectives illustrated above, the reward at time step k is defined as (41).…”

Section: Rewardmentioning

confidence: 99%

Thermal Constrained Energy Optimization of Railway Cophase Systems With ESS Integration—An FRA-Pruned DQN Approach

Xing

2023

IEEE Trans. Transp. Electrific.

View full text Add to dashboard Cite

“…The design of the reward significantly affects the optimization results. To achieve the optimization objectives illustrated above, the reward at time step k is defined as (41).…”

Section: Rewardmentioning

confidence: 99%

Thermal Constrained Energy Optimization of Railway Cophase Systems With ESS Integration—An FRA-Pruned DQN Approach

Xing

2023

IEEE Trans. Transp. Electrific.

View full text Add to dashboard Cite

“…[ 3 ] Based on the type of stored energy, the technologies of EESS have categorized into mechanical, electrical, and chemical. [ 4 ] Flywheel, pumped hydro, and compressed air are the major mechanical ESS; superconducting magnetic and supercapacitor (SC) are the electrical ESS; and battery and fuel cell (FC) are the chemical ESS. [ 5 ] They have different features for different applications like power rating, charge/discharge time, power/energy density, and lifespan.…”

Section: Introductionmentioning

confidence: 99%

Energy Management in Plugin Hybrid Electric Vehicles with Hybrid Energy Storage System Using Hybrid Approach

Ramasamy

Chandramohan

Devadasu³

2022

Energy Tech

View full text Add to dashboard Cite

Herein, an optimal control approach for the energy management of hybrid energy storage system (HESS) like battery, supercapacitor (SC), and integrated charging unit in plugin hybrid electric vehicle (PHEV) is proposed. The proposed approach is the combination of both the jellyfish search optimizer (JS) and gradient boosting decision tree algorithm (GBDT) and it is termed as JS‐GBDT approach. The high energy density battery and high power density SC are combined for satisfying the demand of vehicle. To balance the charging, an uncontrolled rectifier with DC to DC buck converter, and to guarantee smooth transition of energy, two bidirectional DC–DC buck–boost converters are utilized. To meet the load requirements, GBDT approach predicts and integrates the total power supply and the charging level of the power source. The output voltage regulation, reference generation, and smooth tracking of current are performed by the energy management of PHEV using JS approach. The proposed methodology is executed on MATLAB/Simulink working platform. The performance of the HESS is evaluated by utilizing the comparison analysis with existing systems. From the analysis, the proposed approach provides less stress for primary and secondary sources and improves the charging unit performance and enhances the battery life.

show abstract

“…The accelerometer (ACC) will be disturbed by additional translational or rotational acceleration, and the magnetometer (MAG) is easy to be disturbed by the surrounding magnetic field. To solve the drift problem, a variety of calibration algorithms have been proposed, including the adaptive method, [8][9][10] the machine-learningbased method [11][12][13][14] and the zero velocity update (ZUPT) method. 15 Due to the errors in measurements of GYR, ACC, and MAG inside IMUs, numerous studies have proposed sensor fusion algorithms (SFAs) to estimate the three-dimensional (3D) orientation accurately and robustly.…”

Section: Introductionmentioning

confidence: 99%

Sensor network oriented human motion capture via wearable intelligent system

Qiu

Zhao

Jiang

et al. 2021

Int J of Intelligent Sys

View full text Add to dashboard Cite

Using inertial measurement units mounted on foot is a feasible approach to improve the positioning accuracy for the human motion capture system. This paper presents a lightweight and low cost wireless inertial motion capture system for the simultaneous reconstruction of human body attitude and displacement. First of all, the device is based on human sensor networks and distributes 15 sensor nodes on the key human limbs. Then, after an initial sensor alignment with the reduced error, a zero-speed update algorithm is used to calculate foot displacement. In addition, to constantly update the human posture information, a kind of motion reconstruction method based on the gradient descent method was used to fuse the sensor data. Finally, a new method of three-dimensional human body reconstruction is proposed, which is different from the traditional motion capture system. Through unconstrained traversal of the root, the human posture and foot trajectory are combined to realize the synchronous reconstruction of posture and displacement. It is concluded from the experiment results that the estimation errors are well controlled, and motion patterns are consistent with the actual situation.

show abstract

Double Deep Reinforcement Learning-Based Energy Management for a Parallel Hybrid Electric Vehicle With Engine Start–Stop Strategy

Cited by 81 publications

References 44 publications

Thermal Constrained Energy Optimization of Railway Cophase Systems With ESS Integration—An FRA-Pruned DQN Approach

Thermal Constrained Energy Optimization of Railway Cophase Systems With ESS Integration—An FRA-Pruned DQN Approach

Energy Management in Plugin Hybrid Electric Vehicles with Hybrid Energy Storage System Using Hybrid Approach

Sensor network oriented human motion capture via wearable intelligent system

Contact Info

Product

Resources

About