An improved intelligent model predictive controller for cooling system of electric vehicle

Xie, Yi; Liu, Zhaoming; Li, Kuining; Liu, Jiangyan; Zhang, Yangjun; Dan, Dan; Wu, Cunxue; Wang, Pingzhong; Wang, Xiaobo

doi:10.1016/j.applthermaleng.2020.116084

Cited by 56 publications

(17 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the HVAC system reacts gradually rather than instantly, information of the previous time steps is required for the agents to decide the next action. This nature of the HVAC system can be easily inferred from a mathematical modeling of each system component, where most of the dynamics are differential equations with time [28], [36]. In the above equation, the current error of temperature and SC are denoted by T t − T target and Γ t − Γ target .…”

Section: B State Representationmentioning

confidence: 99%

Multi-Agent Reinforcement Learning Based Actuator Control for EV HVAC Systems

Joo¹,

Lee²,

Kim³

et al. 2023

IEEE Access

View full text Add to dashboard Cite

While electric vehicles (EVs) continue to draw more attention as an alternative to traditional fossil fuel vehicles, the relatively short driving range of EVs is often pointed out as their biggest drawback. In terms of energy consumption, one of the most energy-intensive systems in EVs is the heating, ventilation, and air conditioning (HVAC) system. Most HVAC systems use On/Off or PID control for the actuators, but these control methods have low efficiency and are difficult to apply in multiple-input multiple-output systems. In this paper, we propose a novel multi-agent deep reinforcement learning (MADRL) method to directly control actuators such as the compressor and electric expansion valve to reach the setpoint temperature while minimizing the work. The proposed method is evaluated with an industrial vehicle simulator and compared with conventional PID control. The experimental results show that the MADRLbased method improves the efficiency of the HVAC system compared to PID control while achieving comparable target convergence performance.

show abstract

Section: B State Representationmentioning

confidence: 99%

Multi-Agent Reinforcement Learning Based Actuator Control for EV HVAC Systems

Joo¹,

Lee²,

Kim³

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The switching operation of thyristors is replaced with new power devices such as bipolar junction transistors (BJT), metal oxide semiconductor field effect transistors (MOSFET), gate turn-off thyristors (GTO), and insulated gate bipolar transistors (IGBT). For electric propulsion vehicles, there are three main types of semiconductor devices often employed [54]- [57]. After examining the properties of all power devices, IGBTs are gaining attention for use in electric vehicles.…”

Section: E Power Electronic Drivermentioning

confidence: 99%

Emerging Intelligent Bidirectional Charging Strategy Based on Recurrent Neural Network Accosting EMI and Temperature Effects for Electric Vehicle

et al. 2022

View full text Add to dashboard Cite

Transportation is currently advancing towards Electric Vehicle (EV) technology. This paper presents a brief and systematic analysis of the real-time issues obtained in Electric Vehicles (EVs) due to the various ranges of energy storage devices. In general, EV energy management system is integrated with power electronic circuits for effective power conversion and reliable operation. Some issues are addressed while using the batteries in EV systems such as charging time, efficiency of battery, and raw materials. Not only battery issues but also real-time non-technical issues and operational issues are also discussed in this paper. During energy conversion with power electronic circuits, the system attains extreme temperature levels which in turn reduces the performance of the system. To maintain an optimum temperature level, it is important to study the temperature effect of the system at the most prior levels. Due to the adaption of power electronic components, some extent of noise is generated, technically treated as Electromagnetic Interference (EMI), as system capacity increases the EMI content also improved proportionately. Therefore, to mitigate the EMI effect, the low pass filter-based EMI filter is introduced in the system such that the noise level is suppressed. Bidirectional Charging System (BCS) is one of the emerging technologies in EV to obtain autonomous power supply systems in the form of Vehicle to Grid (V2G), Grid to Vehicle (G2V), and Vehicle to Load (V2L). To know the behaviour of BCS the proposed RNN controller is employed and is compared with ANN bidirectional charging model. BCS charging system with RNN controller has better dynamic response to exchange the power via DC/DC converter and AC/DC converter as compared to ANN controller.

show abstract

“…To quantify the comfort level inside the cabin two indices from Fanger's model [8], namely Predicted Mean Vote (PMV) and Predicted Percentage of Dissatisfied (PPD), were used. These two indices were initially proposed for steadystate and uniform conditions but have been widely used to evaluate the thermal comfort level in passenger vehicles and public transport systems [9,10,11,12]. PMV predicts the mean of a large group of people's votes on a seven-point thermal sensation scale, as ±3 (hot/cold), ±2 (warm/cool), ±1 (slightly warm/ slightly cool), and 0 (thermally neutral sensation).…”

Section: Comfort Levelmentioning

confidence: 99%

Efficient Heating Strategies Based On Variable Set-Point Temperature of a Bus Cabin with No Air Recirculation

Afrasiabian¹,

Douglas²,

Cunningham³

et al. 2021

International Conference on Fluid Flow, Heat and Mass Transfer

View full text Add to dashboard Cite

A coupled model of a generic bus and an all-fresh air Heat Pump (HP) system was developed to investigate the impact of the cabin set-point temperature on the thermal comfort level and the energy consumption of heating. Predicted Mean Vote (PMV) and Predicted Percentage of Dissatisfied (PPD) from Fanger's model were used to quantify the comfort level inside the cabin. Different heating strategies based on fixed and variable set-point temperature were examined for an all-fresh air HP under different outdoor temperatures. The results indicate that for the studied all-fresh air HP, adopting a proper heating strategy based on variable set-point temperature could reduce the power demand up to 61% when the outdoor temperature is -5 °.

show abstract

An improved intelligent model predictive controller for cooling system of electric vehicle

Cited by 56 publications

References 22 publications

Multi-Agent Reinforcement Learning Based Actuator Control for EV HVAC Systems

Multi-Agent Reinforcement Learning Based Actuator Control for EV HVAC Systems

Emerging Intelligent Bidirectional Charging Strategy Based on Recurrent Neural Network Accosting EMI and Temperature Effects for Electric Vehicle

Efficient Heating Strategies Based On Variable Set-Point Temperature of a Bus Cabin with No Air Recirculation

Contact Info

Product

Resources

About