PFC application case: Overview of PFC direct current control strategy

The control strategy of PFC is divided into discontinuous current mode (DCM) and continuous current mode (CCM), as well as the critical DCM (BCM) between the two, according to whether the input inductor current is continuous. Some circuits also switch the converter between DCM and CCM according to the load power, which is called mixed conduction mode (MCM). CCM can be divided into direct current control and indirect current control according to whether the transient inductor current is directly selected as the feedback quantity. Direct current control detects the input current of the rectifier as feedback and controlled quantity, and has the advantages of fast system dynamic response, easy current limiting, and high current control accuracy. This article summarizes the direct current control strategy of PFC technology, compares and analyzes the advantages and disadvantages of several typical control strategies, and points out the development trend of these control technologies.

Direct current control includes peak current control, hysteresis current control, average current control, predictive current control, zero-beat control, single-cycle control, state feedback control, sliding mode variable structure control, fuzzy control and other methods.

1 Various direct current control strategies

1.1 Peak current control

The input current waveform of peak current control is shown in Figure 1. The switch tube is turned on at a constant clock cycle. When the input current rises to the reference current, the switch tube is turned off. The sampling current comes from the switch current or the inductor current. The advantage of peak current control is that it is easy to implement, but it has many disadvantages:

1) There is an error between the peak current and the average current, which cannot meet the requirement of very small THD;

2) The current peak is sensitive to noise;

3) When the duty cycle is > 0.5, the system generates subharmonic oscillation;

4) A slope compensator needs to be added to the comparator input.

Therefore, in PFC, this control method tends to be eliminated.

1.2 Hysteresis Current Control

The input current waveform of the hysteresis current control is shown in Figure 2. When the switch is turned on, the inductor current rises. When it rises to the upper threshold, the hysteresis comparator outputs a low level, the switch tube is turned off, and the inductor current decreases; when it drops to the lower threshold, the hysteresis comparator outputs a high level, the switch tube is turned on, and the inductor current increases. This cycle works over and over again, and the sampling current comes from the inductor current.

Hysteresis current control is a simple bang-hang control that combines current control with PWM modulation. It has a simple structure, is easy to implement, and has strong robustness and fast dynamic response capabilities. Its disadvantage is that the switching frequency is not fixed and the filter design is difficult.

At present, the research on the improvement of hysteresis current control is still very active, the purpose is to achieve constant frequency control.Combining other control methods with hysteresis current control is one of the development directions of current control strategy of SPWM current converter.

1.3 Average current control

The input current waveform of the average current control is shown in Figure 3. Average current control adds the inductor current signal to the sawtooth signal. When the sum of the two signals exceeds the reference current, the switch tube is turned off, and when the sum is less than the reference current, the switch tube is turned on. The sampling current comes from the actual input current rather than the switch current. Because the current loop has a higher gain bandwidth, small tracking error, and better transient characteristics. THD (<5%) and EMI are small, it is insensitive to noise, the switching frequency is fixed, and it is suitable for high-power applications. It is currently the most widely used control method in PFC. Its disadvantage is that the error between the reference current and the actual current changes with the change of the duty cycle, which can cause low-order current harmonics.

1.4 Predictive Current Control

Predictive current control is to sample the input, output voltage and input current, select the optimized voltage vector (pulse width) to act on the next cycle according to the error between the actual current and the reference current, so that the actual current can track the reference current within one cycle and achieve steady-state error-free. Its advantages are fixed switching frequency, good dynamic performance, small current harmonics, small device switching stress, and simple digital implementation. Its disadvantages are that it requires higher sampling frequency and switching frequency, and at low sampling frequency, it will produce periodic current errors.

1.5 One-cycle control (integral reset control)

One-cycle control is a nonlinear control that has the duality of modulation and control. Its principle is shown in Figure 4. One-cycle control achieves the purpose of tracking the command signal through the reset switch, integrator, trigger circuit , and comparator.

The basic idea of ​​this method is to control the switch duty cycle and force the average value of the switch variable to be equal to or proportional to the control reference in each cycle, so as to automatically eliminate steady-state and transient errors in one cycle, and the error of the previous cycle will not be carried over to the next cycle. Single-cycle control can optimize system response, reduce distortion and suppress power supply interference. It has the advantages of fast response, constant switching frequency, strong robustness, easy implementation, anti-interference, and simple control circuit. It is a very promising control method. Its disadvantage is that it requires an integral circuit that can be quickly reset. Single-cycle control has been fully studied in DC/DC converters. As a modulation method, this technology has also been widely used in PFC.

1.6 Deadbeat Control

The basic idea of ​​deadbeat control is to divide the output parameter into several sampling periods at equal intervals. Based on the starting value of the circuit in each sampling period, the value of a circuit variable at the end of the sampling period under the action of a square wave pulse symmetrical about the sampling period is predicted. By properly controlling the polarity and width of the square wave pulse, the output waveform can be made to coincide with the required parameter waveform. By continuously adjusting the polarity and width of the square wave pulse in each sampling period, an output with small waveform distortion can be obtained.

This method is a fully digital control technology. It uses the command current value and the actual compensation current value of the previous moment to calculate the switching mode that the rectifier should meet at the next moment according to the space vector theory. Its advantages are rigorous mathematical derivation, tracking without overshoot, good dynamic performance, and easy computer execution. The disadvantages are large amount of calculation and high dependence on system parameters. However, with the increasing popularity of digital signal processing microcontroller (DSP) applications, this is a very promising control method.

The current deadbeat control method based on space voltage vector PWM has constant switching frequency and good regulation performance, representing the advanced level of PFC technology in the world.

1.7 Sliding Mode Variable Structure Control

Sliding mode variable structure control, which was developed in the former Soviet Union in the 1950s, is naturally reasonable for controlling power electronic converters. Because of the discontinuity caused by the electronic switches that constitute various converters, various power electronic converters are described as variable structure systems. In variable structure systems, the sliding mode of sliding mode variable structure control is invariant, that is, it is insensitive to system changes and external interference and has strong robustness. In this way, sliding mode variable structure control can be easily applied to rectifiers, inverters and related fields driven by switching converters, and obtain good control effects.

The time-varying parameters of converters are a problem that people have been trying to solve. Considering that the switching action of the switching converter corresponds to the high-frequency switching of the moving points of the variable structure system along the switching surface, it is possible to consider using the sliding mode variable structure method to control the converter.

In the power factor correction system of the rectifier, there is a contradictory relationship between the steady-state characteristics of the input current and the transient characteristics of the output voltage. The sliding mode variable structure control method can be used to coordinate the steady-state characteristics of the input current and the transient characteristics of the output voltage, and the dynamic response of the output voltage can be improved as much as possible under the premise that the input current meets the relevant standards.

1.8 Duty Cycle Control

This control method does not use a current sensor, because it is based on a slope comparison technique. Therefore, the switching frequency is fixed; in addition, previous control methods are mathematical models obtained under an ideal three-phase balanced state. The duty cycle control method has greater advantages than traditional control methods in analyzing three-phase unbalanced systems, such as in modeling, voltage regulator parameter adjustment, etc.

1.9 Lvapunov-based nonlinear large signal control method

The mathematical modeling of traditional control methods is generally based on the small signal linearization of the system. The common disadvantage of this method is that it cannot guarantee the stability of the system under large signal disturbances. Based on this consideration, the literature [16] proposed to use large signal methods to directly analyze this nonlinear system. The simulation and experimental results show that the system has strong robustness to large signal disturbances.

Control methods related to modern control theory, such as state feedback control (pole configuration), quadratic optimal control, nonlinear state feedback, fuzzy control, neural network control, etc., can all be used in PFC circuits. However, these methods are still immature and are under active exploration. Based on the requirements of high-power electronic equipment, multi-level converters and various simple topologies such as series and parallel topologies have been proposed. In addition to adopting existing control strategies, more targeted control technologies are also being developed for the control of these circuits.

2 Summary and Outlook

In CCM control, direct current control is suitable for high-power occasions with high requirements for system performance indicators and rapidity, and should be the mainstream of development. Medium and high-power power electronic equipment accounts for a large proportion in the power grid, so three-phase PFC should be the focus of PFC research. With the increase in the cost of three-phase PFC and the decrease in switching frequency, digital control has become the mainstream of development relying on high-speed digital processors. Since various control strategies have advantages and disadvantages, reasonable combination of various control strategies, taking advantage of their strengths and making up for their weaknesses, can achieve ideal control effects, which is also a direction for the development of control technology.