Abstract
As an electric vehicle (EV) researcher, I have focused on addressing the critical challenge of thermal management, where energy efficiency and component durability are paramount. This study presents an integrated thermal management system (ITMS) for pure electric vehicles, leveraging heat pump technology to utilize motor waste heat effectively. By connecting independent circuits through heat exchangers, the system achieves optimized energy utilization. Additionally, I propose two advanced control strategies—anti-saturation integral fuzzy control and multi-level fuzzy control—to tackle the nonlinear complexities of thermal management. Joint simulations via AMESim and Simulink validate that the ITMS reduces cabin heating time by 27.8%, improves the coefficient of performance (COP) by 31.3%, and enhances winter driving range by 9.57% compared to independent thermal systems. The optimized fuzzy controls further shorten winter heating time by 18.4% and reduce battery cooling time by 3.6% in summer.
1. Introduction
The thermal management of electric vehicles faces significant challenges: excessive heating energy consumption in low temperatures severely impacts driving range, while inadequate cooling in high temperatures threatens component safety [1]. Traditional EVs often employ three independent thermal subsystems for batteries, air conditioning, and motor-electronics, which lack coordinated energy management. My research aims to integrate these subsystems to harness waste heat, particularly from the motor, which is a substantial energy source in EVs.
Heat pump systems have emerged as an efficient solution for EV thermal management. For example, Tian et al. [3] demonstrated that integrating battery, motor, and heat pump loops improved COP by 25.55% and range by 31.71% in winter. Ahn et al. [4] showed that dual-source heat pumps (air and waste heat) outperform single-source systems in cold environments. Zou et al. reported energy savings of 3–18% by coupling battery cooling with heat pump systems.
However, controlling integrated systems is complex due to their nonlinearity and multi-parameter interactions. Conventional PID control often fails to optimize performance across diverse conditions, leading to oscillations or instability. Fuzzy control, requiring no precise system model, offers robust anti-disturbance and fast response capabilities. This study combines fuzzy control with anti-saturation integration and multi-level architecture to enhance control accuracy and reduce rule complexity.
2. Integrated Thermal Management System Design
2.1 System Architecture
I designed an ITMS for pure electric heavy-duty vehicles to simplify layout, utilize waste heat, and extend range. The system couples four loops: heat pump air conditioning, battery thermal management, motor-electronics thermal management, and a warm air loop (Table 1).
Table 1. ITMS Loop Components and Functions
| Loop | Key Components | Primary Function |
|---|---|---|
| Heat pump air conditioning | Compressor, chillers, expansion valves | Refrigerant circulation for heating/cooling |
| Battery thermal management | PTC heater, heat exchanger, pump | Temperature regulation via cooling/heating |
| Motor-electronics thermal management | Heat exchanger, 四通阀 (four-way valve), radiator | Waste heat recovery and overheat protection |
| Warm air loop | Ethylene glycol medium, PTC heater, heat exchanger | Cabin heating via heat pump, waste heat, or PTC |
The ITMS architecture replaces traditional four-way valves with a cost-effective combination of check valves and stop valves. Two chillers couple the heat pump loop with other circuits: Chiller 1 enables heat exchange with the warm air loop for winter cabin heating, while Chiller 2 transfers motor/battery waste heat to the heat pump, improving COP.
2.2 System Modeling
2.2.1 Battery Heat Generation Model
The battery heat generation power q is calculated using Bernardi’s equation [9]:\(q = I^2 R_{\text{lsat}} + IT \frac{dU_{\text{OCV}}}{dT}\) where I is discharge current, \(R_{\text{lsat}}\) is equivalent resistance, \(U_{\text{OCV}}\) is open-circuit voltage, T is temperature, and \(dU_{\text{OCV}}/dT\) is the entropic coefficient. For the lithium iron phosphate battery used (Table 2), this model accounts for ohmic and polarization losses, as well as reversible reaction heat.
Table 2. Battery Parameters
| Parameter | Value |
|---|---|
| Type | Lithium iron phosphate |
| Nominal voltage per cell (V) | 3.2 |
| Nominal capacity per cell (A·h) | 268 |
| Dimensions (mm) | 218×200×47.7 |
| Thermal conductivity (W·m⁻¹·K⁻¹) | 11.91 (height), 10.21 (width), 2.5 (thickness) |
| Mass per cell (kg) | 4.65 |
| Charge/discharge cutoff voltage (V) | 3.65/2.5 |
2.2.2 Motor Heat Generation Model
Motor heat generation \(P_{\text{mot}}\) arises from energy loss during electrical-to-mechanical conversion:\(P_{\text{mot}} = P_m (1 – \eta_m)\) where \(P_m\) is motor power and \(\eta_m\) is efficiency (Table 3). The recoverable waste heat power Q is:\(Q = P_{\text{mot}} \eta = c q_m (t_{\text{out}} – t_{\text{in}}) \eta\) with \(\eta\) as heat exchanger efficiency, c as coolant specific heat, \(q_m\) as coolant mass flow, and \(t_{\text{in}}/t_{\text{out}}\) as motor inlet/outlet temperatures.
Table 3. Motor Parameters
| Parameter | Value |
|---|---|
| Type | Permanent magnet synchronous motor |
| Maximum torque (N·m) | 500 |
| Maximum power (kW) | 160 |
| Maximum speed (r·min⁻¹) | 10,000 |
| Maximum efficiency (%) | 94 |
2.2.3 Heat Pump System Modeling
The compressor mass flow \(q_m\) is:\(q_m = \eta_v \rho_{\text{suc}} N D\) where \(\eta_v\) is volumetric efficiency, \(\rho_{\text{suc}}\) is suction density, N is speed, and D is displacement. Enthalpy increase \(h_{\text{inc}}\) is:\(h_{\text{inc}} = h_d – h_s = \frac{h_{\text{dis}} – h_s}{\eta_{\text{is}}}\) with \(h_d/h_s/h_{\text{dis}}\) as discharge, suction, and isentropic discharge enthalpies, and \(\eta_{\text{is}}\) as isentropic efficiency. The compressor torque \(\tau_{\text{is}}\) is:\(\tau_{\text{is}} = \frac{q_m h_{\text{inc}} \eta_{\text{mech}}}{N}\) where \(\eta_{\text{mech}}\) is mechanical efficiency.
For heat exchangers, the internal convective heat transfer \(\Phi_{\text{int}}\) is:\(\Phi_{\text{int}} = h_{\text{ci}} S (T_{\text{ref}} – T_{\text{wall}})\) with \(h_{\text{ci}}\) as convective heat transfer coefficient, S as heat transfer area, and \(T_{\text{ref}}/T_{\text{wall}}\) as refrigerant and wall temperatures. External heat transfer \(\Phi_{\text{ext}}\) is:\(\Phi_{\text{ext}} = h_{\text{ce}} S (T_{\text{air}} – T_{\text{wall}})\) where \(h_{\text{ce}}\) and \(T_{\text{air}}\) are external convective coefficient and air temperature.
3. Control Strategies for Thermal Management
3.1 Working Mode Switching
The ITMS switches modes based on component temperatures using logic threshold control (Table 4). For example, when \(T_{\text{cab}} \geq 25^\circ\text{C}\) and \(T_{\text{bat}} > 35^\circ\text{C}\), the system enters parallel cooling mode for the cabin and battery. Winter modes prioritize waste heat utilization: if \(T_m > T_{\text{amb}} + 5^\circ\text{C}\), motor waste heat heats the cabin via the heat pump or directly.
Table 4. ITMS Working Modes and Trigger Conditions
| Mode | Temperature Conditions | Active Components |
|---|---|---|
| 1. Cabin cooling | \(15^\circ\text{C} < T_{\text{bat}} < 35^\circ\text{C}, T_{\text{cab}} \geq 25^\circ\text{C}\) | Compressor |
| 2. Battery cooling | \(T_{\text{amb}} \geq 25^\circ\text{C}, T_{\text{cab}} < 25^\circ\text{C}, T_{\text{bat}} > 35^\circ\text{C}\) | Compressor |
| 3. Parallel cooling | \(T_{\text{amb}} \geq 25^\circ\text{C}, T_{\text{cab}} \geq 25^\circ\text{C}, T_{\text{bat}} > 35^\circ\text{C}\) | Compressor |
| 4. Heat pump heating (air source) | \(T_{\text{amb}} \geq -10^\circ\text{C}, T_{\text{cab}} < 25^\circ\text{C}, 15^\circ\text{C} < T_{\text{bat}} < 35^\circ\text{C}\) | Compressor |
| 5. PTC heating | \(T_{\text{amb}} < -10^\circ\text{C}, T_{\text{cab}} < 25^\circ\text{C}, 15^\circ\text{C} < T_{\text{bat}} < 35^\circ\text{C}\) | PTC1 |
3.2 Optimized Fuzzy Control Strategies
3.2.1 Anti-Saturation Integral Fuzzy Control
Traditional fuzzy control may exhibit steady-state errors due to empirical rules. I integrated anti-saturation integration to eliminate these errors while preventing integral saturation. The control input u is:\(u = u_{\text{fuzzy}} + u_{\text{int}}\) where \(u_{\text{fuzzy}}\) is the fuzzy output and \(u_{\text{int}}\) is the integral term with anti-saturation limiting.
3.2.2 Multi-Level Fuzzy Control
For multi-variable systems, conventional fuzzy control requires exponentially increasing rules. My multi-level approach simplifies this by decomposing a multi-variable controller into cascaded two-variable controllers. For example, a two-level controller for compressor duty cycle (DR) first processes cabin and battery cooling demands (AC, BC) as inputs to the first level, then combines their outputs for the second level:\(\text{Level 1: } AC = f_1(\Delta T_{\text{cab}}, \dot{\Delta T}_{\text{cab}}), \quad BC = f_1(\Delta T_{\text{bat}}, \dot{\Delta T}_{\text{bat}})\)\(\text{Level 2: } DR = f_2(AC, BC)\) This reduces rule count from \(7^4 = 2401\) to \(3×5 + 5×7 = 50\), improving computational efficiency.
4. Simulation and Results Analysis
4.1 Winter Performance Evaluation
Joint simulations via AMESim and Simulink under WTVC conditions (0°C ambient) compared the ITMS to independent thermal systems. The ITMS shortened cabin heating time to 22°C by 27.8%, with an average COP increase of 31.3%. Battery heating using motor waste heat raised its temperature to 15°C in 2550 s, reducing PTC usage.
Motor coolant temperature in the ITMS remained lower than in independent systems, indicating effective waste heat utilization. This resulted in a 9.57% increase in winter driving range, as shown by the SOC comparison.
Table 5. Winter Performance Metrics Comparison
| Metric | Independent System | ITMS | Improvement |
|---|---|---|---|
| Cabin heating time (s) | 1800 | 1300 | -27.8% |
| COP | 1.8 | 2.4 | +31.3% |
| Battery heating time to 15°C (s) | – | 2550 | – |
| Winter driving range (km) | 280 | 307 | +9.57% |
For control strategies, the anti-saturation integral fuzzy control eliminated the 1.3°C steady-state error of traditional fuzzy control, reducing heating time by 18.4% compared to optimized PID control.
4.2 Summer Performance Evaluation
In summer (35°C ambient), the ITMS cooled the cabin and battery while maintaining motor temperature below 90°C. The multi-level fuzzy control reduced cabin temperature fluctuations and eliminated overshoot, while shortening battery cooling time by 3.6% versus single-level fuzzy control (Tables 6–7).
Table 6. Summer Cabin Cooling Performance
| Control Strategy | Temperature Overshoot (°C) | Stability Time (s) |
|---|---|---|
| Single-level fuzzy | 1.5 | 600 |
| Multi-level fuzzy | 0 | 450 |
Table 7. Summer Battery Cooling Time
| Control Strategy | Cooling Time to 35°C (s) | Improvement |
|---|---|---|
| Single-level fuzzy | 2400 | – |
| Multi-level fuzzy | 2312 | -3.6% |
5. Conclusion
This study presents an integrated thermal management system for electric vehicles that efficiently utilizes motor waste heat through heat pump integration. The ITMS reduces winter cabin heating time by 27.8%, improves COP by 31.3%, and extends driving range by 9.57% compared to independent systems. The proposed anti-saturation integral and multi-level fuzzy controls enhance response speed and stability, with the former eliminating steady-state errors and the latter reducing rule complexity.
Future work will focus on real-world validation of the ITMS in various climatic conditions and further optimizing control strategies for dynamic driving scenarios. The integration of machine learning for predictive thermal management also shows promise for enhancing energy efficiency in electric vehicles.