In recent years, the global energy crisis and environmental pollution have intensified the urgent need for clean and renewable energy alternatives. The electric car has emerged as a green transportation solution, gaining significant attention due to its zero tailpipe emissions and becoming a focal point for the transformation of the automotive industry worldwide. Within this context, the study of control strategies for charging circuits has become an indispensable aspect. Research indicates that the management of voltage and current during charging not only affects charging efficiency but also directly impacts the lifespan and stability of power batteries. Therefore, optimizing intelligent control strategies during charging to minimize energy loss and ensure battery charging within safe limits is crucial for ensuring the overall performance of electric cars.
This paper focuses on the control strategy for the charging circuit of an electric car, aiming to meet the rapid charging requirements of new energy vehicles and achieve seamless switching between different voltage levels. Based on the charging principles of electric cars and the technical characteristics of current charging systems, we propose an improved control strategy that incorporates fuzzy control and intelligent optimization techniques. Through mathematical modeling and simulation analysis, the feasibility of this control strategy is validated. The simulation results demonstrate advantages in power factor correction, with a total harmonic distortion (THD) content of only 3.37%, which well meets harmonic requirements. The widespread adoption of electric cars hinges on efficient and reliable charging infrastructure, making this research pivotal for advancing electric car technology.
The rapid development of electric cars has concurrently driven the advancement of related equipment. Whether electric cars can be widely used depends heavily on charging devices. Charging equipment for electric cars includes onboard chargers for AC slow charging and off-board DC charging piles for high-power DC fast charging. The high-power charging systems for electric cars typically employ DC fast-charging technology, enabling rapid energy replenishment and are widely used in commercial applications such as passenger cars, electric buses, and logistics vehicles. However, such charging facilities require dedicated sites and professional maintenance teams, and due to high construction costs and insufficient layout density, they often face issues like long charging queues and limited regional coverage. In contrast, AC slow-charging systems, with advantages such as flexible installation and strong adaptability, better meet the daily needs of private electric car users and can be widely deployed in residential areas, parking lots, and other scenarios. Thus, optimizing charging circuits is essential for enhancing the practicality of electric cars.
Developing and optimizing charging systems is a critical component in the evolution of electric cars, as their performance directly influences charging efficiency, battery life, and operational stability. The system must maintain stable voltage and current waveforms while achieving high power factor and quality power output. By optimizing harmonic suppression capabilities to reduce grid interference, energy conversion efficiency is improved, ensuring safe and reliable charging processes for electric cars. In this study, we address these challenges by designing a robust control strategy for the charging circuit of an electric car.
Charging Circuit Structure Design
The charging circuit for an electric car typically involves a two-stage structure: an AC-DC power factor correction (PFC) stage followed by an isolated DC-DC conversion stage. The main circuit topology is designed to ensure efficient energy transfer and minimal harmonic distortion. For the DC-DC stage, a full-bridge LLC resonant circuit is employed due to its soft-switching characteristics, which reduce switching losses and maximize utilization. The primary side consists of switch tubes Q1–Q4, diodes, and capacitor elements C1–C4 forming an inverter bridge, while the secondary side comprises diodes VD1–VD4 forming a rectifier bridge. The LLC converter operates in three states based on the relationship between resonant frequency and switching frequency: over-resonance, resonance point, and under-resonance. This paper primarily analyzes phase-shift control, where voltage regulation is achieved by adjusting the phase-shift angle, enabling step-down operation with fixed frequency and variable duty cycle.
The resonant control criteria are summarized in Table 1, which outlines the control methods and frequency ranges for different operating states. For phase-shift control, we focus on the resonant and over-resonance states. In the resonant state, the converter operates with zero-voltage switching (ZVS), reducing stress on components. In the over-resonance state, similar benefits are achieved with careful design. The working principles involve multiple modes where switch tubes and diodes conduct alternately to transfer energy to the output, ensuring stable charging for the electric car battery.
| Operation | Control Method | Frequency |
|---|---|---|
| Step-up | Variable frequency control left of resonance point | $$f_s < f_r$$ |
| Step-down | Phase-shift control near resonance point | $$f_s = f_r$$ |
| Step-down | Phase-shift control right of resonance point | $$f_s > f_r$$ |
For the electric car charging circuit, power device selection and parameter matching are critical. The design follows standards such as the “Electric Vehicle Conduction Charging System” series, which specifies requirements for power factor, output voltage ripple, and efficiency. The LLC circuit parameters are designed through a systematic process. First, the transformer turns ratio is determined based on input and output voltages. For an input voltage of 400 V and output reference voltage of 240 V, the turns ratio is calculated as:
$$n = \frac{V_{in}}{V_o} = \frac{400}{240} = 1.67$$
where $$V_{in}$$ is the input voltage and $$V_o$$ is the output voltage. Next, the inductance coefficient is set between 0.1 and 0.5; here, we choose 0.2 to balance gain curve steepness and efficiency:
$$\lambda = \frac{L_r}{L_m} = 0.2$$
where $$L_r$$ is the resonant inductance and $$L_m$$ is the magnetizing inductance. The resonant frequency is selected as 100 kHz to minimize size while considering losses. The magnetizing inductance design ensures ZVS under light load conditions, with the constraint:
$$L_m \leq \frac{n V_o t_d}{8 C_{oss} V_{in} f_r}$$
where $$t_d$$ is the dead time (300 ns), and $$C_{oss}$$ is the output capacitance of the switch tubes. This design ensures reliable operation for electric car charging.
Mathematical modeling and equivalent analysis of the circuit are essential for control strategy development. The equivalent input impedance and input-output voltage gain are derived as:
$$Z_{in} = \frac{A R_e + B}{C R_e + B}, \quad G = \frac{n U_o}{U_{in}} = \frac{R_e}{A R_e + B}$$
where $$A$$, $$B$$, and $$C$$ are coefficients from the equivalent circuit, and $$R_e$$ is the equivalent resistance on the primary side. The frequency characteristics of input and output voltages are expressed as:
$$H(j\omega) = \frac{j\omega L_m R_e}{R_e + j\omega L_m} \cdot \frac{1}{j\omega L_r + \frac{1}{j\omega C_r} + \frac{j\omega L_m R_e}{R_e + j\omega L_m}}$$
The resonant circuit parameters are defined as:
$$\alpha_3 = L_m L_r C_r, \quad \alpha_2 = L_r C_r R_e + L_m C_r R_e, \quad \alpha_1 = L_m, \quad \beta_0 = R_e, \quad \Delta = L_m L_r C_r R_e$$
where $$C_r$$ is the resonant capacitor. This model facilitates stability and efficiency analysis for the electric car charging system.
Control Strategy Design and Optimization
The control strategy for the electric car charging circuit aims to suppress input current harmonics and ensure stable output voltage. We employ a double-loop control scheme for the Boost PFC stage, with an inner current loop for sinusoidal input current and an outer voltage loop for output regulation. The Boost PFC main circuit uses a multiplier-based control strategy to achieve power factor correction. In continuous conduction mode (CCM), the input voltage is $$v_{in} = V_{in} \sin(\omega t)$$, where $$\omega = 2\pi f$$ and $$f$$ is the grid frequency. The rectified voltage is $$v_t = |v_{in}|$$, and the Boost circuit ensures input current sinusoidalization and output voltage stability.
Input current distortion in Boost PFC arises from two main sources: tracking errors in the reference current due to inner loop controller limitations, and distortion in the reference current caused by second-harmonic ripple in the output voltage. To address this, we propose a harmonic suppression strategy that minimizes THD. The state-space average model for the Boost PFC is:
$$\frac{di}{dt} = \frac{v_{in} – k e}{L}$$
where $$i$$ is the inductor current, $$L$$ is the inductance, $$k$$ is the duty cycle, and $$e$$ is the switch voltage. By constructing a virtual axis, we transform the system into a stationary coordinate frame:
$$\frac{di_{\alpha\beta}}{dt} = \frac{v_{\alpha\beta} – e_{\alpha\beta}}{L}$$
where $$v_{\alpha\beta} = v_{in}$$ and $$e_{\alpha\beta}$$ is the switch voltage in stationary coordinates. This transformation enables decoupled control and harmonic reduction for electric car charging.
For the inner loop, we design a PID controller with feedforward decoupling. The PID transfer function is:
$$G(s) = K \left( k_p + \frac{k_i}{s} \right)$$
where $$K$$ is the PWM gain, and $$k_p$$ and $$k_i$$ are PI controller coefficients. In the dq-frame, the decoupled equations are:
$$v_d^* = \hat{L} \omega i_q + G(s) (i_d^* – i_d), \quad v_q^* = -\hat{L} \omega i_d + G(s) (i_q^* – i_q)$$
where $$\hat{L}$$ is the estimated inductance. When $$\hat{L} = L$$, perfect decoupling is achieved, but in practice, inductance variations may occur due to saturation or measurement errors, affecting dynamic performance. Therefore, we incorporate adaptive techniques to enhance robustness for electric car charging applications.
Charging state recognition and switching logic are vital for managing different operating modes. The grid voltage is rectified, and the switch $$\phi$$ controls energy transfer. In mode 1, when $$\phi$$ is on, the current flows through the inductor and switch; in mode 2, when $$\phi$$ is off, the current flows through the inductor and freewheeling diode. This ensures continuous operation and efficient power delivery to the electric car battery.
To optimize the control strategy, we integrate fuzzy logic and intelligent optimization. Fuzzy control adjusts PID parameters online based on real-time circuit changes, providing adaptability and robustness without relying on precise mathematical models. This is particularly beneficial for electric car charging, where load and environmental conditions can vary widely. The fuzzy rules are designed to minimize THD and improve power factor, ensuring compliance with standards for electric car charging systems.
Simulation Modeling and Experimental Verification
To validate the proposed control strategy for the electric car charging circuit, we built a simulation model using MATLAB/Simulink. The model includes the charging circuit control system hardware structure, onboard power battery model, drive motor model, and charging pile model. The hardware structure comprises a DC-DC converter, onboard inverter, and DC-AC converter, reflecting real-world electric car charging scenarios. The simulation platform allows for comprehensive testing under various conditions.
The simulation scenarios and parameter settings are based on typical electric car charging requirements. For the Boost PFC circuit, key parameters are listed in Table 2. These values are chosen to align with industrial standards for electric car charging infrastructure.
| Parameter | Value |
|---|---|
| Input Voltage Amplitude (V) | 220 |
| Resistance (Ω) | 160 |
| Switching Frequency (Hz) | 10^5 |
| Inductance (mH) | 6 |
| Capacitance (μF) | 320 |
| AC Voltage Frequency (Hz) | 50 |
We conducted simulations to analyze the control strategy’s performance. For the voltage outer loop, two bandwidth settings were compared: high bandwidth and low bandwidth with harmonic compensation. Under high bandwidth, the input current waveform distorted, failing to track voltage phase, resulting in high THD. With harmonic compensation, the THD was significantly reduced to 3.37%, meeting the requirement of below 5%. The input current achieved sinusoidal shape and unity power factor, demonstrating the effectiveness of our approach for electric car charging.
The simulation results show that the proposed control strategy excels in power factor correction. The harmonic spectrum analysis indicates that third-harmonic content is substantially suppressed, and other odd harmonics are reduced, leading to low distortion. This is crucial for electric car charging, as it minimizes grid pollution and improves energy efficiency. Additionally, the dynamic response of the system was tested under load variations, confirming robustness and stability for electric car applications.
Conclusion
In this paper, we have presented a comprehensive study on the control strategy for the charging circuit of an electric car. The proposed strategy employs an improved feedforward decoupling control scheme with fuzzy logic and intelligent optimization techniques. Through mathematical modeling and simulation, we validated the feasibility and effectiveness of this approach. The results demonstrate significant advantages in power factor correction and harmonic suppression, with a THD content of only 3.37%, which well satisfies the stringent requirements for electric car charging systems. This research contributes to the advancement of electric car technology by enhancing charging efficiency and reliability, ultimately supporting the widespread adoption of electric cars as a sustainable transportation solution. Future work may involve hardware implementation and real-world testing to further refine the control strategy for diverse electric car charging scenarios.