The transition to battery electric vehicles (BEVs) represents a paradigm shift in automotive engineering, demanding innovative solutions for challenges never faced by their internal combustion engine counterparts. One of the most critical subsystems is the thermal management system. Unlike conventional vehicles that can scavenge waste heat from the engine, a battery electric vehicle must create its own heating and cooling entirely from electrical energy. This places a significant burden on the vehicle’s precious battery capacity. Effective thermal management is not merely a matter of passenger comfort; it is intrinsically linked to safety, performance, and the all-important driving range. The lithium-ion battery packs that power these vehicles operate optimally within a narrow temperature band, typically between 15°C and 35°C. Operating outside this range accelerates degradation, reduces usable capacity, and in extreme cases, can lead to thermal runaway events. Simultaneously, the cabin heating in cold climates, if addressed by inefficient Positive Temperature Coefficient (PTC) heaters, can reduce the driving range by over 50%. Therefore, developing a highly efficient, integrated thermal management system is paramount for the success of any battery electric vehicle.
Compounding this technical challenge is a pressing environmental mandate. Global regulations, such as the Kigali Amendment to the Montreal Protocol, are driving the phasedown of high-Global Warming Potential (GWP) hydrofluorocarbon (HFC) refrigerants like R134a, which has been the standard in mobile air-conditioning for decades. The search for sustainable, low-GWP alternatives has led researchers to natural refrigerants. Among them, R290 (propane) stands out due to its excellent thermodynamic properties, zero Ozone Depletion Potential (ODP), and a negligible GWP of 3. Its high latent heat of vaporization and favorable transport properties suggest it could offer superior performance. However, its flammability requires careful system design and engineering controls for safe application in a battery electric vehicle. This study presents the design, experimental validation, and simulation-based analysis of a novel R290-based integrated thermal management system for battery electric vehicles, with a specific focus on its heating performance under challenging low-ambient conditions.
The core of the proposed system is an air-source heat pump cycle. To mitigate the safety concerns associated with the direct use of flammable R290 in the cabin HVAC unit, an indirect heat exchange architecture was adopted. The primary refrigerant (R290) circuit exchanges heat with a secondary coolant loop—a 50:50 ethylene glycol-water solution—via a brazed plate heat exchanger (condenser in heating mode). This cooled or heated coolant is then circulated to various vehicle subsystems. The system’s versatility is achieved through a network of multi-port valves that reconfigure the coolant flow paths. The key operational modes for heating are summarized in the table below:
| Operational Mode | 8-way Valve State | 9-way Valve State | Description |
|---|---|---|---|
| Cabin Heating Only | Mode 1 (Bypass Battery Loop) | Mode 1 or 2 | Coolant flows exclusively through the cabin HVAC heater core. Heat source can be ambient air (Mode 1 on Valve 9) or the powertrain’s waste heat (Mode 2). |
| Battery Heating Only | Mode 2 (Bypass HVAC Loop) | Mode 1 or 2 | Coolant flows exclusively through the battery chiller/ heater plate. Heat source is selected via Valve 9. |
| Simultaneous Cabin & Battery Heating | Mode 3 (Proportional Split) | Mode 1 or 2 | Coolant flow is proportionally split between the cabin HVAC and battery loops to service both demands concurrently. |
The primary thermodynamic cycle for heating follows the reverse Rankine cycle. The performance of any heat pump system, especially for a battery electric vehicle where energy efficiency translates directly to range, is governed by the First Law of Thermodynamics. The heating capacity \(Q_h\) provided to the coolant is derived from the energy balance across the condenser/ gas cooler:
$$ Q_h = \dot{m}_{coolant} \cdot (h_{coolant,out} – h_{coolant,in}) $$
where \(\dot{m}_{coolant}\) is the mass flow rate of the secondary coolant, and \(h_{coolant,in}\) and \(h_{coolant,out}\) are the specific enthalpies at the inlet and outlet of the condenser, respectively. The Coefficient of Performance for heating (\(COP_h\)) is the primary metric of efficiency, defined as the ratio of useful heating effect to the work input required to drive the cycle:
$$ COP_h = \frac{Q_h}{W_{comp}} $$
Here, \(W_{comp}\) is the electrical power input to the compressor. A higher \(COP_h\) means more heat is delivered per unit of battery energy consumed, which is the ultimate goal for extending the winter range of a battery electric vehicle.
To thoroughly analyze and optimize this system, a high-fidelity simulation model was developed. The model’s accuracy is paramount for predictive analysis and control strategy development. Key components were modeled and calibrated against experimental data from a dedicated test bench. The bench replicated the integrated system, with environmental chambers simulating outdoor and cabin conditions.
Compressor Model: The compressor is the heart of the system. Its mass flow rate \(\dot{m}_{ref}\) is modeled as:
$$ \dot{m}_{ref} = \eta_{vol} \cdot \rho_{suction} \cdot N \cdot V_{disp} $$
where \(\eta_{vol}\) is the volumetric efficiency (0.92), \(\rho_{suction}\) is the density of the refrigerant at the compressor inlet, \(N\) is the rotational speed (rev/s), and \(V_{disp}\) is the displacement volume (38 cc). The thermodynamic work is linked through the isentropic efficiency \(\eta_{is}\):
$$ \eta_{is} = \frac{h_{dis, is} – h_{suction}}{h_{discharge} – h_{suction}} $$
The mechanical power is then calculated from the indicated work and the mechanical efficiency \(\eta_{mech}\) (0.90).
Heat Exchanger Model: The plate heat exchangers (condenser and chiller) were critically calibrated. The overall heat transfer is governed by the convection on both refrigerant and coolant sides. The convective heat transfer coefficient \(h\) is calculated from the Nusselt number \(Nu\). For the coolant side in forced convection, different correlations are used for laminar and turbulent flow. A widely used correlation for turbulent flow in plate heat exchangers is of the form:
$$ Nu_{Turb} = C \cdot Re^{m} \cdot Pr^{n} $$
where \(Re\) is the Reynolds number, \(Pr\) is the Prandtl number, and the constants \(C, m, n\) were determined through calibration against manufacturer data. The calibration results showed excellent agreement, with average errors in heat transfer rate and pressure drop below 2.2% and 3.0%, respectively, validating the model’s predictive capability for the battery electric vehicle thermal system.
System Performance Analysis: Compressor Speed Influence
One of the primary control variables in a heat pump system for a battery electric vehicle is the compressor speed. An experimental study was conducted at an ambient temperature of -7°C and a cabin temperature of 5°C, varying the compressor speed from 3000 rpm to its maximum of 8500 rpm. The results, also confirmed by the simulation model, reveal a critical trade-off.
As compressor speed \(N\) increases, the refrigerant mass flow rate \(\dot{m}_{ref}\) rises linearly according to the model equation. This leads to a significant increase in the system’s heating capacity \(Q_h\), as more refrigerant circulates and transfers heat in the condenser. However, the compressor power \(W_{comp}\) rises at a steeper rate. The increase in power consumption is due not only to the higher mass flow but also to a decrease in overall compressor efficiency (volumetric and isentropic) at higher pressure ratios and speeds. Consequently, the \(COP_h\) exhibits a clear declining trend. At 3000 rpm, the \(COP_h\) was approximately 2.74, while at 8500 rpm, it dropped to around 1.39. This finding is crucial for the energy management strategy of a battery electric vehicle: to maximize efficiency and preserve driving range, the compressor should operate at the lowest speed that meets the instantaneous thermal demand of the cabin and/or battery.
Comparative Analysis: R290 vs. Conventional R134a
A pivotal part of this research was to benchmark the R290 system against a baseline using the conventional refrigerant R134a. Tests were performed under three challenging winter conditions for a battery electric vehicle: (0°C, 10°C cabin), (-7°C, 5°C cabin), and (-20°C, 0°C cabin). The results decisively favored R290.
| Performance Metric | Condition (Ambient, Cabin) | R290 System | R134a System | Advantage |
|---|---|---|---|---|
| Heating Capacity \(Q_h\) | (0°C, 10°C) | 8152 W | ~3287 W | R290 provides 2.5x to 3x more heat. |
| (-7°C, 5°C) | ~7468 W | ~3100 W | ||
| (-20°C, 0°C) | ~5200 W | ~1700 W | ||
| HVAC Outlet Air Temperature | (0°C, 10°C) | 68.1 °C | 37.4 °C | R290 delivers significantly warmer air, ensuring cabin comfort. |
| (-7°C, 5°C) | ~58 °C | ~32 °C | ||
| (-20°C, 0°C) | ~42 °C | 22.9 °C | ||
| COP_h (at high speed) | All conditions | ~1.5 | ~1.5 | Comparable at high load; R290’s advantage is in capacity. |
The superior performance of R290 is rooted in its fundamental thermophysical properties. R290 has a significantly higher latent heat of vaporization compared to R134a. This means that for the same mass flow rate, the amount of heat absorbed during evaporation and released during condensation is much greater for R290. Furthermore, R290 operates at higher pressures and has better transport properties (higher thermal conductivity, lower viscosity), which enhance heat transfer coefficients in the heat exchangers. The dramatic difference in HVAC outlet temperature is particularly noteworthy. At -20°C, the R134a system could only produce an outlet temperature of 22.9°C, which is barely adequate for cabin heating, while the R290 system maintained a robust 42°C, ensuring passenger comfort even in extreme cold—a common challenge for battery electric vehicles.
The thermal performance advantage can be further understood by analyzing the refrigerant’s state during operation. For instance, in the 0°C ambient test, the R290 system’s evaporating temperature stabilized around -15.1°C, maintaining a healthy temperature lift and a stable pressure ratio across the compressor. The R134a system, with its lower latent heat, struggled to achieve a comparable heat transfer rate even at high mass flow, indicating it was operating near its capacity limit with the given charge and component sizing. This highlights that an R290-based system for a battery electric vehicle can be more compact or provide superior performance with similarly sized components.
Dynamic Control and System Stability
A high-capacity system is only useful if it can be controlled effectively. To demonstrate this, the validated simulation model was used to implement a PID (Proportional-Integral-Derivative) controller. The control objective was to regulate the cabin’s HVAC outlet air temperature to a comfortable setpoint of 25°C by modulating the compressor speed—the key actuator identified earlier. This mimics a real-world scenario where the driver sets a desired cabin temperature.
The dynamic response of the controlled system was evaluated under the same three temperature conditions. The results were highly promising for application in a battery electric vehicle. The system demonstrated stable and responsive control:
- At 0°C ambient, the outlet temperature reached and stabilized at 25°C within 72 seconds.
- At -7°C ambient, stabilization was achieved in 174 seconds.
- Even at the severe -20°C condition, the system brought the outlet temperature to the 25°C setpoint and maintained it stably within 332 seconds.
The initial transient in all cases showed a brief dip in outlet temperature. This is a realistic phenomenon caused by the initial influx of cold ambient air from the cabin blower at startup, while the heat pump cycle was still ramping up to deliver heat. The PID controller effectively compensated for this disturbance. This exercise proves that not only does the R290 system have excellent steady-state heating performance, but it is also amenable to precise thermal management control, a non-negotiable requirement for modern battery electric vehicles where seamless human-machine interaction is expected.
Conclusion and Outlook
This comprehensive study, encompassing detailed design, experimental validation, and simulation analysis, conclusively demonstrates the high potential of R290 as a next-generation refrigerant for battery electric vehicle thermal management systems. The designed indirect heat pump system successfully addresses the flammability concern while unlocking the superior thermodynamic properties of propane.
The experimental findings reveal that the R290-based system significantly outperforms a conventional R134a system in heating capacity and achievable air temperature, particularly under low-ambient conditions that are critical for real-world adoption of battery electric vehicles in all climates. The research also provides key insights for energy-efficient operation: compressor speed should be minimized to meet the thermal load, thereby maximizing the system’s COP and preserving the vehicle’s driving range. The development and validation of an accurate simulation model provide a powerful tool for further system optimization, component sizing, and advanced control strategy development.
Finally, the successful implementation of a PID controller to accurately regulate cabin temperature under various conditions underscores the system’s practicality and readiness for integration. The transition to low-GWP refrigerants is an environmental imperative. This work shows that for battery electric vehicles, this transition can be coupled with a substantial improvement in thermal performance, contributing to extended range, enhanced passenger comfort, and faster cabin warm-up times. Future work will involve integrating this system with high-fidelity models of the battery and powertrain to study its performance over dynamic driving cycles and its interaction with the vehicle’s overall energy management system.