Introduction with the development of power electronics technology, power electronic devices have developed from SCR (thyristor) in the 1960s to hvigbt (high voltage resistant insulated gate bipolar transistor). After VVVF frequency conversion, vector control frequency conversion and direct torque control frequency conversion appear. Their common shortcomings are low input power factor, DC circuit needs high-voltage and large capacity energy storage capacitor, and regenerative energy cannot be fed back to electricity. Matrix AC-AC converter can overcome the above shortcomings, and has attracted more and more attention in recent years
compared with the traditional AC-DC-AC converter and AC-AC converter, the matrix converter has the following remarkable characteristics:
(1) the output voltage amplitude and frequency can be controlled independently, and the output frequency can be higher than or lower than the input frequency, which can reach any value theoretically
(2) under some control laws, the input power factor angle can be flexibly adjusted to more than 0.99, and can be freely adjusted, leading, lagging or close to the unit power factor angle
(3) adopting four quadrant switch can realize two-way flow of energy
(4) there is no intermediate energy storage link, with compact structure and high efficiency
(5) the input current waveform is good without low-order harmonics
(6) it has strong controllability
The control strategy of matrix converter includes the determination, realization and safe commutation of switching function s. The determination methods of switching function include direct transformation method, space vector modulation method [1] and hysteresis current tracking method. At present, the research of space vector modulation method is relatively mature. There are four steps, three steps, two steps and soft switching commutation in the research of commutation methods2. The development of topology
the circuit topology of matrix converter was proposed by llglli in 1976. Until 1979, nturini and esina[7] first proposed a matrix AC-AC converter structure composed of nine power switches, and pointed out that the input power factor angle of the matrix converter can be adjusted arbitrarily, but later found that this converter has an inherent limit, the maximum voltage gain is 0.866, and is independent of the control algorithm. Since the main circuit of the matrix converter adopts nine bidirectional switches, there are also problems in the realization and protection of bidirectional switches. The difficulty lies in the fact that when the switches are commutating, there can be neither dead zone nor overlap, otherwise, any situation will lead to the damage of the switches. In order to realize safe commutation, Rany proposed a four step commutation strategy, which can realize semi soft switching commutation
2.1 topology
when matrix converter was first proposed, it refers to the general structure of converting M-phase input to n-phase output, so it was once called general-purpose converter. According to the different values of M and N and the different properties of the power supply at the input and output terminals, people have proposed many topologies
(1) from three-phase AC to two groups of DC, or a group of DC with changeable polarity
(2) change from three-phase AC to single-phase AC
(3) from single DC to three-phase AC, which is commonly referred to as inverter
(4) from AC three-phase to AC three-phase, its input and output terminals are connected with each other by bidirectional switches, that is, 9-switch matrix converter, which is a topology most studied
(5) from AC three-phase to AC three-phase, but the input and output terminals are connected by three fully controlled bridges, which is called voltage source matrix converter. Its structure is more complex than 9-switch matrix converter, but its performance is better
the circuit topology of AC-AC Matrix Converter with three-phase input and three-phase output is shown in Figure 1
Figure 1 AC-AC matrix converter topology
it contains 9 bidirectional switches, which can realize the transformation of power supply voltage and frequency through its logic control, (2) provide voltage and current with adjustable amplitude and frequency to the load in case of poor forming pressure, mold design, temperature and other conditions
2.2 development history of components
the development of matrix converter components fully reflects the progress and development trend of power electronics technology. In general, it mainly experienced the following processes
(1) research on bidirectional power devices [8]
because the bidirectional power devices required by matrix converter do not exist at present, we studied the synthesis of bidirectional switches using other power electronic devices. Known synthesis methods include: embedding a full control switch in the rectifier bridge; Parallel current switch; Series voltage switch; Common collector anti parallel full control switch; Common emitter anti parallel full control switch
the common emitter structure in the manufacture of bidirectional switches of matrix converters is shown in Figure 2
Figure 2 common emitter bidirectional switching circuit
(2) research on power module
similar to IGBT module, multiple bidirectional switching devices are integrated on a silicon chip, and some even integrate protection circuit and trigger circuit, which reduces the volume and weight of the converter
(3) research on device integration
integrate power devices or modules, drive circuits, protection circuits, and power supplies to form the so-called pebbpower electronics building blocks. It further reduces the volume of the whole converter device. More importantly, it greatly improves the reliability of the converter and reduces the loss
3. Modulation strategy of matrix converter
at present, the modulation strategy of matrix converter is often described by switching function matrix. The determination of switching function, that is, the modulation strategy of matrix converter, mainly has the following three methods:
(1) direct transformation method
is to synthesize the output voltage by continuously chopping the input voltage. It can be divided into coordinate transformation method, harmonic injection method and dual voltage instantaneous value control method. Although these methods have certain advantages, they also have their shortcomings, such as the low output voltage of matrix converter based on coordinate transformation method; The harmonic injection method has a large amount of calculation and complex switching state, so it has high requirements for the control system
(2) indirect transformation method
this method can be called AC-DC-AC equivalent transformation method and space vector modulation method. At present, the research on matrix converter is more and more mature. It virtualizes AC-AC transformation into AC-DC and DC-AC transformation, which is equivalent to rectification and inversion. When it is realized, rectification and inversion are completed in one step, and low-order harmonics are well suppressed. Its control scheme is complex and lacks effective dynamic analysis support. On this basis, Danish scholar Christian klumpner et al. Developed a polygonal flux modulation method, which is also a new modulation method based on the indirect modulation model. During the sampling period, only one effective vector and one zero vector in the inverter phase are used to minimize the stator flux error; In the rectification stage, only a single current vector is selected according to the principle of minimizing the angular error of the input current reference vector. Therefore, during the sampling period, the number of switches can be reduced, especially in the low-frequency modulation stage, and the accuracy of the output voltage can be improved; At the same time, the input current vector can be directly controlled. In this method, the flux linkage is projected according to a polygon, which is very close to a circle, so the magnetic leakage of the motor is minimized. Its main advantages are that it can accurately estimate the input current; Direct control of input current vector angle; Reduce switching times and improve pulse resolution; Increase the input switching frequency
(3) current control method
it takes the output voltage as the control target, and generally requires that the current is a symmetrical sinusoidal quantity, so the output current of the converter should track the sinusoidal change of the given current. It has two basic implementation methods: hysteresis current control method and predictive current control method
● hysteresis current tracking method is to compare the three-phase output current signal with the measured output current signal, and determine the switching action according to the comparison result and the current switching power supply state. It has the advantages of easy understanding, simple implementation, fast response, good robustness and so on, but the switching frequency is not stable, the harmonic is randomly distributed, and the input current waveform is not ideal, and there are large harmonics
● the basic idea of predictive current control method is to use the expected current value of the next switching cycle of the converter and the current actual current value to calculate the output voltage vector of the converter that conforms to the current change, and then use the space vector method to synthesize this output voltage vector in the virtual inverter of the converter to achieve the purpose of tracking the output current, but the complexity and calculation will increase
all the above modulation strategies have their own advantages and problems to varying degrees, which affect the depth and breadth of research and application of these methods. In different occasions, the emphasis is different, and different modulation strategies should be used for research
4. The latest development of matrix converter technology
it has been 30 years since matrix converter was proposed in 1976. Many foreign literatures have proposed the experimental prototype of matrix converter, but it has not really entered the practical report. At present, the maximum output power of the converter can reach 20kW, and the control means mainly use TMS320C30 and C40 digital signal processors. Only in this way can 803 ensure that the measured data is correct; 86 microcomputer and PLD device. The matrix converter project team of Aalborg University has done a better job in this regard
in the late 1980s and early 1990s, the research of Yugoslav scholar l.h-uber, American professor rojevic, Japanese scholars professor iguro and njhashi, and Korean scholars on and a gradually matured the theory and control technology of matrix converter. BER and rojevic proposed a PWM technology based on space vector modulation technology
iguro and ruhashi proposed two-wire voltage instantaneous value method. Korean scholars on and a analyzed the dynamic and static characteristics of practical boost nine switch MC by using DQ circuit transformation technology, assuming that MC is composed of non ideal current source and voltage source, which provides an effective method for MC analysis. At the 1994 Virginia Power Electronics Center annual meeting, a three-phase to three-phase matrix converter with power factor correction (PFC) at the input was exhibited. The converter uses a digital signal processor (DSP) to realize space vector modulation, with a maximum output of 2kW, a switching frequency of 20kHz, MOSFET devices, an induction motor with a load of 2kW, a power factor of 0.99 at the input, an output voltage Plastic processing equipment is needed to conduct experiments on their products, and the input current is sinusoidal. From 1995 to 1996, peter.nilsen made a series of research on the peripheral circuit of matrix converter with Siemens C166 as the controller in his doctoral thesis. From 1998 to 1999 and from 1999 to 2000, Christan also developed a set of devices in the United States as a visiting scholar twice, and studied the control strategy of matrix converter under artificial load when the input voltage was unbalanced
the research on matrix converter in China started relatively late. Basically, since the 1990s, Nanjing University of Aeronautics and Astronautics, Xi'an Jiaotong University, Shanghai University and Harbin Institute of technology have carried out research in this field, and achieved remarkable results, reaching a certain level. In 1992, Professor Zhuang Xinfu of Nanjing University of Aeronautics and Astronautics analyzed the DC-AC and AC-DC converters by using space vector modulation method, and obtained the modulation method of AC-AC converter after synthesis. With a 32-bit digital signal processor tms32014 as the controller, a
Copyright © 2011 JIN SHI