High-performance, high-capacity AC motor speed control and energy-saving technology – current status and future prospects
2026-04-06 04:51:18··#1
Abstract: This paper reviews the current status and dynamics of high-performance, high-capacity AC motor speed control systems at home and abroad, and introduces several popular circuit topologies in current research and application fields. Finally, it looks forward to the application of PWM control technology and high-performance, high-capacity AC motor speed control systems in energy, environment, and transportation. Keywords: High voltage, high capacity, multilevel converter, PWM control I. Introduction Energy shortage and environmental pollution are common century-old problems facing mankind. The two global energy crises since the 1970s and the severity of current environmental problems have aroused widespread attention to energy-saving technologies worldwide. China's energy production and consumption rank second in the world, but it is still far from meeting the needs of industrial production and people's lives. Under the circumstances of severe energy shortage, the huge gap in energy saving results in excessive energy consumption per unit of output, and the annual energy waste is staggering. For example, a considerable number of fan and pump loads waste a lot of electrical energy due to constant speed drive. This type of drive system accounts for about half of the total industrial electric drive volume. If speed control energy-saving technology is adopted, at least 20% of electrical energy can be saved. my country's "15th Five-Year Plan" set energy-saving goals to continuously improve energy efficiency and effectiveness, with a focus on promoting widespread energy-saving technologies. One important measure is to gradually achieve economical operation of electric motors, fans, pumps, and related equipment and systems, and to develop motor speed regulation and power electronics energy-saving technologies. Only in this way can we support the rapid, healthy, and sustainable development of the national economy in the long term with a lower energy consumption elasticity coefficient and greater energy savings. Furthermore, large quantities of coal and oil are burned without proper processing, resulting in low thermal utilization and serious environmental pollution. Currently, excessive vehicle emissions have contributed to the global greenhouse effect and are one of the main causes of air pollution in Beijing. A crucial way to solve urban environmental pollution and traffic congestion is to develop high-speed public transportation (subways, urban light rail) and electric vehicles. High-speed electrified trains are the preferred option for achieving rapid intercity transportation, and their core technologies are modern power electronics and AC motor drive technology—a new technology that has developed rapidly alongside microelectronics since the 1980s. Furthermore, large and medium-capacity AC motor speed control systems should be widely used in industrial and civil fields such as steel rolling, papermaking, cement manufacturing, mine hoisting, and ship propulsion. In these applications, AC speed control systems not only achieve energy conservation but also optimize overall system performance, improve process conditions, and significantly enhance production efficiency and product quality. Current data and market availability of large-capacity speed control products show that the total annual sales of AC motor speed control system hardware, software, and peripheral equipment worldwide is $4.85 billion. Europe, the Middle East, and Africa account for 39% of this, Japan 27%, North America 21%, Asia 12%, and Latin America 1%. In terms of system power distribution, low-power speed control systems still dominate the market, with 1-4 kW systems accounting for 21% of total sales and 5-40 kW systems accounting for 26%. However, with the rapid improvement in the voltage withstand capability, current rating, and switching performance of new composite devices such as IGBTs and IGCTs, high-capacity AC motor speed control technology is bound to experience rapid development and significant progress, with a very promising market prospect. Foreign countries are far ahead of us in the research and application of high-performance, high-capacity AC motor drive technology. MVA-level high-voltage inverters are already widely available on the market and applied in systems such as electric locomotives, ship electric propulsion, steel rolling, papermaking, and water supply. AC motor variable frequency speed control technology and its products have become leading industries in some industrialized countries. Currently, my country's research and development of large and medium-capacity AC speed control systems started relatively late, and many essential applications are dominated by foreign products. Therefore, developing reliable, inexpensive, high-performance, large and medium-capacity AC motor variable frequency speed control systems and putting them into mass production as soon as possible will have significant strategic and practical implications for promoting national economic development, transforming the economic growth model, reducing energy consumption per unit of output, and breaking the monopoly of Western countries in this field. II. Current Status of High-Capacity AC Motor Speed Control Technology Since the 1980s, modern power electronics technology has rapidly developed towards higher frequencies, higher efficiency (low switching losses), higher power factors, higher power density (combined integration), and higher voltage and power. Self-turn-off devices, represented by GTOs, BJTs, and MOSFETs, have seen significant development, especially the remarkable advancement of bipolar composite devices represented by IGBTs. This has propelled power electronic devices towards higher capacity, higher frequencies, easier driving, lower losses, and intelligent modularity. Along with the rapid development of power electronic devices, high-power inverters and AC speed control technology are also becoming increasingly high-performance. 1. Traditional high-power inverter circuits The converters used in traditional high-power AC motor speed control systems are mainly: (1) Ordinary AC-DC-AC three-phase inverter (2) Step-down - ordinary frequency converter - step-up (3) AC-AC frequency converter (4) Transformer-coupled multi-pulse inverter The research on the above high-power conversion circuits is relatively mature, but while realizing high-power AC drive, there is no breakthrough in performance, and the device is complex, the manufacturing cost is high, the control method has low reliability, and it seriously pollutes the power grid, with low power factor and large reactive power loss. It is necessary to add harmonic control devices, and the equipment cost increases several times. Therefore, in the past ten years, some new high-voltage high-power inverters, especially voltage-type multilevel converter topologies, have attracted the attention of many scholars. 2. New multilevel voltage-type inverters A. Nabae et al. of Nagaoka University of Technology in Japan first proposed the three-level inverter, also known as the neutral-point clamped (NPC) inverter, at the IAS annual meeting in 1980. Its emergence opened up a new idea for the development of high-voltage, high-capacity voltage-type inverters. Based on this, after years of research, several main multilevel converter topologies have been developed, mainly divided into two types [1][2][3]: one is a clamped converter topology with a single DC power supply, including diode clamped, capacitor clamped, and general-purpose topologies developed on this basis, as well as stacked multi-cell topologies; the second is a cascaded inverter with a separate DC source. Figure 1 classifies the existing multilevel converters as follows: Figure 1 Classification diagram of existing multilevel converters According to the nature of the DC voltage source and the different series connection methods, the above two topologies can be represented by two circuit models: a single DC power supply direct series voltage divider model and a series model of multiple electrically independent DC power supplies, as shown in Figures 2 and 3. In Figure 2, the multilevel converter circuit can be equivalent to the multiple switches in the dashed line, which in reality are composed of a network of power switching devices, and different switching states represent being connected to different nodes. In Figure 3, Vdc1…Vdcn, acting as DC power supplies, can be combined into various output voltage levels through different switching states of the conversion circuit. Figure 2 shows a single DC source multilevel circuit model, and Figure 3 shows a discrete DC source multilevel circuit model. Compared with ordinary two-level inverters, multilevel converter topologies have the following advantages: • More suitable for high-capacity, high-voltage applications. • Can generate M-level stepped output voltages; theoretically, increasing the number of levels can approach a pure sine wave with very low harmonic content. • Electromagnetic interference (EMI) problems are greatly reduced because the dv/dt of a single operation of the switching element is typically only 1/(M-1) of that of a traditional two-level inverter. • High efficiency. Eliminating the same harmonics, two-level inverters use PWM control, resulting in high switching frequencies and large losses, while multilevel inverters can use lower switching frequencies, resulting in lower losses and higher efficiency. In addition to the common features mentioned above, each of the topologies has its own advantages and disadvantages, which are compared as follows: (1) Diode-clamped multilevel inverter The diode-clamped multilevel structure is one of the earliest and most widely used structures. The characteristic of this structure is that multiple diodes are used to clamp the corresponding switching elements and output the corresponding M-level phase voltage. The diode-clamped topology has the common advantages of multilevel inverters, but it has its own shortcomings: a) The clamping diodes are not uniformly subjected to voltage. b) The devices require different rated currents. Designing according to the maximum rated current will result in a waste of (M-1)(M-2)/2 of the switching element capacity, resulting in low utilization efficiency. c) The DC side capacitor may have an imbalance in voltage when transferring active power because the current flowing in and out of the DC side capacitor in one cycle may not be equal. When active power is transferred, if a constant voltage device is not added, the M level will gradually become a three-level (M is odd) or a two-level (M is even). The solution is usually to use a PWM voltage regulator or a battery to replace the capacitor, but this will make the system more complex and increase the cost. To solve the above problems, several improved structures have emerged on the traditional diode clamped multilevel structure. The improved topology of adding clamping capacitors to the two adjacent clamping diodes not only solves the diode series problem, but also clamps the overvoltage when the switching device is turned off. Due to the charging and discharging effect of the added capacitor, the imbalance of the DC side capacitor voltage is reduced, and bidirectional current flow can be realized. Another improved structure uses two identical converters back to back, with the left side as the rectifier and the right side as the inverter. The corresponding nodes of the DC side capacitors are connected, which can better balance the capacitor voltage. (2) Capacitor clamped multilevel inverter The capacitor clamped multilevel inverter was first proposed by TAMeynard and H.Foch at the PESC conference in 1992. The initial purpose was to reduce the excessive clamping diodes in the NPC multilevel inverter, that is, to use floating capacitors to replace the clamping diodes, while the DC side capacitor remains unchanged. Its working principle is similar to that of a diode clamping circuit. Compared to a diode-clamped multilevel inverter, this topology eliminates a large number of diodes but introduces a significant number of capacitors. For high-voltage, high-capacity systems, capacitors are bulky, occupy a large area, are costly, and difficult to package. The introduction of capacitors increases the options for voltage synthesis and provides greater flexibility in selecting switching states. By combining appropriate different switching states at the same level, the capacitor voltage can be kept balanced, making it suitable for active power regulation and variable frequency speed control systems. However, the control method becomes more complex, and the switching frequency increases, leading to higher switching losses and reduced efficiency. To maintain capacitor voltage balance, Meynard proposed a back-to-back converter structure to adjust the capacitor charging and discharging balance, and used a certain proportion of switching modes to simultaneously control the rectifier bridge and inverter bridge, ensuring that the power flowing to and from the capacitor is the same. By detecting the capacitor voltage, if an imbalance occurs, the control of the rectifier bridge can be appropriately adjusted. Its disadvantages are: the introduction of a large number of floating capacitors and the problem of capacitor voltage balance. Currently, the French company ALSTOM has developed a product to address this issue. (3) Voltage Self-Balancing Multilevel Inverter Topology In 2000, Dr. Peng Fangzheng of the University of Michigan proposed a voltage self-balancing multilevel topology. It does not require additional circuitry to suppress the voltage deviation of the DC-side capacitor, theoretically realizing a truly practical multilevel structure. Traditional diode-clamped and capacitor-clamped circuit topologies can also be simplified and developed from it. One of the technical challenges of high-voltage, high-capacity multilevel circuits is the control of the midpoint voltage. For topologies with three or more levels, if the midpoint voltage is not well controlled, it cannot be effectively applied to high-capacity power conversion applications. For the above topologies, when the voltage is higher than three levels, either an isolated DC power supply is required, or a complex circuit structure needs to be added to help maintain the balance of the midpoint voltage. This new topology has a voltage self-balancing function, which can effectively control the midpoint voltage for various inverter control strategies and load conditions. Figure 4 shows a two-level unit. Figure 5 shows the single-phase topology of this novel self-balancing multilevel structure. As can be seen from the figure, it is composed of the basic unit shown in Figure 4. Because the basic unit is a two-level single-phase circuit, the multilevel structure composed of it is also called a P2 multilevel inverter. The characteristics of this voltage-self-balancing P2 multilevel topology are: • The power loss of the system is inversely proportional to the capacitance and the switching frequency. Increasing the switching frequency and adding some specific switching states can greatly reduce the loss and improve the system efficiency. • Compared with general diode-clamped and capacitor-clamped topologies, the midpoint voltage of each stage of the system can be well controlled. • For an M-level P2 inverter system, the number of switching devices/diodes required is M*(M-1); the number of capacitors required is M*(M-1)/2. • The calculation is simple, and the device stress can be minimized. By simplifying and modifying the system in Figure 5, we can obtain traditional diode-clamped and capacitor-clamped multilevel topologies, as well as some other improved topologies. Removing all the clamping switches in Figure 5 yields a diode-capacitor-clamped multilevel system, as shown in Figure 6; removing the clamping switches and diodes results in a capacitor-clamped multilevel system, as shown in Figure 7; removing the clamping switches and capacitors yields a diode-clamped topology, as shown in Figure 8; and reversing the diode connections results in an improved back-to-back diode-clamped system, as shown in Figure 9. Applications of this general multilevel topology include switched-capacitor DC-DC converters and voltage multiplier circuits; furthermore, combined with other circuits, it can achieve bidirectional DC-DC conversion. A three-level unit can also be used instead of a two-level unit to implement a multilevel inverter. Figure 6 Diode-capacitor clamping system Figure 7 Capacitor clamping multilevel system Figure 8 Diode clamping system Figure 9 Improved back-to-back diode clamping system (4) Stacked multi-unit structure (SMC) See Figure 10, which can also achieve high voltage and multi-level output [12][13]. This structure has certain advantages over the general capacitor clamping structure. It can use fewer capacitors and smaller volume, reducing the size of the device, especially in applications with high voltage output above three levels. The SMC topology is a hybrid structure based on the basic converter unit composed of bridging capacitors and switches. Figure 11 shows the structure of a two-layer stacked two-unit converter. This structure is equivalent to stacking two capacitor clamping units. In the figure, S21a, S21b and S21 are complementary switches and cannot be turned on at the same time. Similarly, other switches also have similar complementary switch pairs. By using a similar capacitor clamping switching method in the upper and lower layers, multi-level output can be achieved. Figure 10 Schematic diagram of stacked multi-unit structure Figure 11 Two-layer stacked two-unit SMC multilevel converter structure However, this structure also has some disadvantages: In order to meet the withstand voltage requirements when the bottom layer and the top layer are turned on, the power switches on the outside of the topology are two tubes directly connected in series, which brings about the problem of turn-on and turn-off synchronization. Moreover, since it does not always work in the above two states, from another perspective, the withstand voltage capacity of the power devices is wasted. Moreover, when it is necessary to further increase the voltage and the number of stacks exceeds two layers, the number of switches will increase greatly and the capacitance will also increase. At the same time, the control method of this type of topology is also relatively complicated, and its advantages are not obvious. (5) Series multilevel inverter with separate DC power supply Figure 12 Topology structure of H-bridge series five-level converter with separate DC power supply For series multilevel inverter with separate DC power supply, to obtain more levels, it is only necessary to increase the number of unit inverter bridges connected in series in each phase by the same amount. Its characteristics are: The DC side uses DC power supplies with the same voltage but isolated from each other, eliminating voltage balancing issues. It requires no diodes or capacitor clamping, facilitating speed control. Because each H-bridge uses single-phase control, AC current flows through the DC capacitor at all times, requiring a relatively large DC capacitor. The control method is relatively simple. Due to the identical structure of each stage, each stage can be PWM controlled separately, and then the waveform can be reassembled. For the same number of voltage levels, the series structure requires the fewest devices. Generally, diode-clamped and capacitor-floating structures are limited to 7 or 9 voltage levels, while the series structure, without the limitations of diodes and capacitors, can have a larger number of voltage levels, suitable for higher voltages, and with less harmonic content. Because each inverter bridge stage has the same structure, it facilitates modular design and manufacturing, simplifies assembly, and improves system reliability. Furthermore, if one inverter bridge stage fails, it is bypassed, and the remaining modules can be powered uninterrupted, minimizing production losses. Because this structure easily uses low-voltage power switching devices to achieve multi-stage voltage series connection, obtaining high voltage and large capacity, it has significant practicality. Of course, the disadvantage of this structure is that it requires a lot of isolated DC power supplies, which limits its application. At present, many well-known international electrical companies, including Robicon, Toshiba, ANSLADO, and Mitsubishi, have similar products that can be used in industries such as high-capacity motor speed regulation and reactive power compensation. Domestic products have also been launched and can be used in speed regulation systems for driving fans, water pumps, etc. (6) Three-phase inverter series structure In 1999, E.Cengelci et al. proposed a new type of transformer-coupled unit series high-voltage frequency converter structure. Its main idea is to use a transformer to superimpose the output of three conventional inverter units composed of IGBT or IGCT to achieve a higher voltage output. Furthermore, these three conventional inverters can use the same control method, which greatly simplifies the circuit structure and control method. Its topology is shown below: Figure 13 Three-phase inverter series inverter topology diagram The advantages of this three-phase inverter series inverter structure are: Three conventional inverters form the core of the high-voltage frequency converter, and each inverter can use the common PWM method The three conventional inverters operate in a balanced manner, each sharing 1/3 of the total output power The output of the entire frequency converter can be equivalent to 7-level PWM, with low harmonics and low dv/dt The capacity of the output transformer is only 1/3 of the total capacity 18-pulse input, no harmonics on the grid side and high power factor Figure 14 Motor line voltage PWM waveform Figure 15 Output transformer winding Because the voltage, current and power of the three inverters in the three-phase inverter series structure are completely symmetrical, the three inverters can use the same control law, but it is equivalent to a two-level high-voltage frequency converter, and the dv/dt is too large. Therefore, the PWM signals of the three inverters can be staggered by 1/3 of a cycle. For SPWM, this means each of the three inverters uses a triangular wave with a 120° phase difference, equivalent to a high-frequency transformer with a line voltage of 7 levels. In summary, diode-clamped and capacitor-clamped systems are more suitable for reactive power regulation due to voltage equalization issues, but are more difficult to control in active power transfer, such as motor speed regulation, requiring additional algorithms. The voltage-self-balancing P2 multilevel system does not require a large number of transformers, has a compact structure, high power factor, no electromagnetic interference, and low losses, attracting widespread attention and application in the field of multilevel inverter implementation. Provided the cost of the input transformer is acceptable, a series structure can achieve high voltage and high capacity with lower voltage-rated devices. Since the number of levels can be large, the harmonics on the grid side and output side are very low. If four-quadrant rectification is used and combined with modern motor control theory, high-performance four-quadrant high-capacity AC motor frequency conversion speed regulation will become possible, and its application in the AC drive field will be very promising. Three-phase inverters in series can ensure balanced power utilization and operation under variable torque load conditions, and have low harmonic pollution to the power grid. They can be well used in medium-voltage (2300~4160 V) AC motor speed control drive systems. 3. PWM control technology While the circuit topology of high-power inverters is constantly being updated, the corresponding PWM control technology has also developed rapidly. Scholars from various countries have not only innovated the traditional PWM, but also proposed some new control strategies. 1) Traditional PWM control technology and its development Traditional PWM control technology is mostly used for gate drive control of two-level inverters. Its main method is to rely on the comparison of the carrier wave and the modulation wave to obtain the intersection point, or to use the microcomputer calculation method to obtain the gate trigger pulse control signal. Sinusoidal pulse width modulation (SPWM) uses a sine wave as the modulation wave. Typical methods for implementation include natural sampling PWM, regular sampling PWM, and equal area PWM. In a three-level circuit, if two sine waves are compared with a triangular wave, bidirectional dipolar modulation PWM[14] can be obtained, which can greatly reduce the harmonics of the phase voltage. All of these methods can be used in multilevel circuits. The implementation methods vary depending on the structure. 2) Optimized PWM Technology: In recent years, optimized PWM technology has developed rapidly. It seeks the PWM control waveform based on objective functions such as minimum harmonic content, minimum harmonic distortion rate (THD), and minimum torque ripple. Optimized PWM has special advantages that general PWM methods do not possess, such as high voltage utilization, fewer switching cycles, and the ability to achieve specific optimization goals. Optimized PWM can be used in multilevel inverters, and the control law of each switching element can be optimized using the characteristics of NPC inverters to improve overall performance and reduce motor losses. 3) Multilevel Inverters and Space Vector PWM: The space vector PWM method uses the ideal magnetic flux of an AC motor powered by a three-phase symmetrical sinusoidal voltage as a reference. It approximates the reference circular magnetic flux using the actual magnetic flux generated by different switching modes of the inverter. The comparison result determines the switching sequence of the inverter, forming the required PWM waveform. The harmonic elimination effect of the voltage vector PWM method is similar to that of multilevel SPWM. For three-level and five-level inverters, the switching modes are easy to calculate and easily implemented digitally. However, as the number of levels increases, the computational load of the switching mode increases dramatically, and the required memory also increases significantly. Due to the high redundancy in switching mode selection, choosing an appropriate vector can eliminate common-mode voltage, and for diode-clamped multilevel inverters, it can eliminate or reduce the imbalance of DC-side capacitor voltage. With the advent of multilevel inverters, space vector pulse-width modulation (SVPWM) has been further developed. For example, for a three-level midpoint clamped inverter, selecting an appropriate combination of space vectors and voltage vector conduction time can yield a magnetic flux that closely approximates a circle. Depending on the vector selection, there are various SVPWM control schemes, each yielding different modulation vector angles and control performance. Compared to dual-level space vector, it has a wider vector selection range, better approximates sinusoidal magnetic flux, and achieves better motor control performance. Simultaneously, its superior topology increases system capacity, improves reliability, and reduces losses. Three-level inverters have high voltage on the DC side, thus posing a potential high-voltage threat to devices and limiting reliability. Furthermore, balancing the DC-side capacitor voltage is a challenging aspect of control. This type of inverter also exhibits grid-side harmonics, which can be effectively addressed using special processing methods, such as dual PWM technology. In certain applications (e.g., UPS), multilevel inverters can also employ current hysteresis control PWM. III. Conclusion and Outlook Due to breakthroughs in power device development and topology, the development of high-capacity AC motor speed control technology is showing a new face and holds immense development opportunities. Traditional high-power inverter circuits are increasingly limited in application due to their large size, poor performance, and high harmonic generation on the grid. New multilevel inverters, on the other hand, are gaining increasing attention due to their superior dynamic performance, lower harmonic generation on the grid and motor, and ability to operate at high voltage. The application of PWM technology in multilevel inverters has led to several improvements, playing a crucial role in the application of high-performance, high-capacity inverters. Currently, China's motor speed control technology primarily focuses on low-voltage, small-capacity speed control devices, while high-voltage, high-efficiency variable frequency speed control devices are mainly imported. Faced with the urgent needs for energy saving and improved processes, and the huge market potential, the production of domestically produced high-voltage, high-power frequency converters is still in its infancy. However, difficulties and hopes coexist, challenges and opportunities are intertwined. Internationally, those capable of producing and developing new high-power variable frequency speed control devices are all world-renowned electrical engineering companies. Because they have followed a step-by-step approach in the development of power electronics technology, forming a complete industrial chain from power semiconductor devices to finished product manufacturing, market inertia and their large organizational structures prevent them from immediately switching to entirely new products. my country, as an emerging developing country, although it has invested in older technologies, the investment is relatively small, and the burden is not heavy. It can quickly shift to the development and utilization of the latest technologies, learning from others' experiences and skipping the paths already trodden. Rapid industrialization based on research results in the latest fields can significantly shorten the gap with advanced countries, and in some aspects, even surpass them. Currently, the time is ripe for the application of high-capacity AC motor speed control technology. As long as China gets on track in terms of institutional reform, production management, and business decision-making, its development prospects are limitless.