I. Commonly Used Control Methods in Frequency Converters
1. Non-intelligent control method
Non-intelligent control methods used in AC frequency converters include V/f control, slip frequency control, vector control, and direct torque control.
(1) V/f control
V/f is the ratio of the voltage applied to the motor stator to the power supply frequency.
As shown in the figure below, if V/F conforms to the straight line AB, it is a linear type; if it conforms to the broken line segment ABC, it is a multi-point type; and if it conforms to the curve AB, it is a square type.
V/f control is proposed to achieve ideal torque-speed characteristics. It is based on the idea of adjusting the speed by changing the power supply frequency while keeping the motor's magnetic flux constant. Most general-purpose frequency converters use this control method. V/f control frequency converters have a very simple structure, but they use an open-loop control method, which cannot achieve high control performance. Furthermore, at low frequencies, torque compensation is necessary to change the low-frequency torque characteristics.
(2) Slip frequency control
Slip frequency control is a direct torque control method. Based on V/f control, it adjusts the inverter's output frequency according to the power supply frequency corresponding to the actual speed of the asynchronous motor and the desired torque, so that the motor can have the corresponding output torque.
This control method requires the installation of a speed sensor in the control system, and sometimes current feedback is also added to control the frequency and current. Therefore, it is a closed-loop control method, which can give the frequency converter good stability and good response characteristics to rapid acceleration and deceleration and load changes.
(3) Vector control
Vector control uses vector coordinate circuits to control the magnitude and phase of the motor's stator current, thereby controlling the excitation current and torque current in the d, q, and 0 coordinate systems, and ultimately controlling the motor torque. By controlling the sequence and timing of the vectors' actions, as well as the duration of the zero vector, various PWM waves can be generated to achieve different control objectives. For example, a PWM wave with the fewest switching operations can be generated to reduce switching losses. Currently, the vector control methods practically used in frequency converters are mainly two types: slip frequency-based vector control and sensorless vector control.
Both slip frequency-based vector control and slip frequency control have the same steady-state characteristics. However, slip frequency-based vector control requires coordinate transformation to control the phase of the motor stator current, ensuring it meets certain conditions to eliminate fluctuations during torque current transients. Therefore, slip frequency-based vector control offers significantly improved output characteristics compared to slip frequency control. However, this control method is a closed-loop control, requiring a speed sensor on the motor, thus limiting its application.
Sensorless vector control controls the excitation current and torque current separately through coordinate transformation. It then identifies the speed by controlling the voltage and current on the motor stator windings to control both the excitation and torque currents. This control method offers a wide speed range, high starting torque, reliable operation, and ease of use. However, the calculations are complex and generally require a dedicated processor. Therefore, its real-time performance is not ideal, and the control accuracy is affected by the accuracy of the calculations.
(4) Direct Torque Control
Direct torque control utilizes the concept of spatial vector coordinates to analyze the mathematical model of an AC motor in the stator coordinate system, controlling the motor's flux linkage and torque. The stator flux linkage is measured by detecting the stator resistance.
Therefore, it eliminates the need for complex transformation calculations such as vector control, making the system intuitive and simple, with improved calculation speed and accuracy compared to vector control. Even in open-loop mode, it can output 100% of the rated torque and has load balancing function for multiple drives.
(5) Optimal control
The practical applications of optimal control vary depending on the requirements. Based on optimal control theory, individual parameters can be optimized for a specific control requirement. For example, in the control applications of high-voltage frequency converters, time-segmented control and phase-shift control strategies have been successfully employed to achieve the optimal voltage waveform under certain conditions.
(6) Other non-intelligent control methods
In practical applications, some non-intelligent control methods can also be implemented in the control of frequency converters, such as adaptive control, sliding mode variable structure control, differential frequency control, circulating current control, and frequency control.
2. Intelligent control method
Intelligent control methods mainly include neural network control, fuzzy control, expert systems, and learning control. There are some successful examples of using intelligent control methods in the control of frequency converters.
(1) Neural network control
Neural network control is used in the control of frequency converters, which is generally for relatively complex system control. At this time, there is little understanding of the system model. Therefore, the neural network must not only perform the function of system identification, but also perform control.
Furthermore, neural network control can control multiple frequency converters simultaneously, making it suitable for controlling multiple frequency converters cascaded together. However, too many layers in the neural network or overly complex algorithms can bring many practical difficulties in specific applications.
(2) Fuzzy control
Fuzzy control algorithms are used to control the voltage and frequency of frequency converters, thereby controlling the acceleration time of the motor to avoid the impact of excessive acceleration on motor lifespan and excessive acceleration on work efficiency. The key to fuzzy control lies in the universe of discourse, membership degree, and fuzzy level division. This control method is particularly suitable for multi-input single-output control systems.
(3) Expert System
Expert systems are a control method that utilizes the experience of so-called "experts." Therefore, an expert system typically requires an expert database to store expert information, as well as an inference mechanism to seek the ideal control result based on known information. The design of the expert database and the inference mechanism is particularly important, as they determine the quality of the expert system's control. Expert systems can control both the voltage and current of frequency converters.
(4) Learning control
Learning control is primarily used for repetitive inputs, and regular PWM signals (such as center-modulated PWM) precisely meet this condition. Therefore, learning control can also be used in the control of frequency converters.
Learning control does not require much system information, but it does require 1 to 2 learning cycles, so its speed is relatively poor. In addition, the learning control algorithm sometimes needs to implement a lead-ahead mechanism, which cannot be achieved with analog devices. At the same time, learning control also involves a stability issue, which must be paid special attention to when applying it.
II. Future Prospects for Variable Frequency Drive Control
With the development of high-tech technologies such as power electronics, microelectronics, and computer networks, the control methods of frequency converters will develop in the following aspects in the future.
(1) Implementation of digital control frequency converter
Currently, the control method of frequency converters can achieve relatively complex calculations using digital processors. Frequency converter digitization will be an important development direction. At present, frequency converter digitization mainly uses single-chip microcomputers such as MCS51 or 80C196MC, supplemented by SLE4520 or EPLD LCD displays to achieve more complete control performance.
(2) Combination of multiple control methods
Each control method has its own advantages and disadvantages; there is no "one-size-fits-all" control method. In some control situations, it is necessary to combine several control methods, such as combining learning control with neural network control, adaptive control with fuzzy control, or direct torque control with neural network control, or what is called "hybrid control." This way, the strengths of each method can be combined to compensate for their weaknesses, resulting in better control performance.
(3) Implementation of remote control
The development of computer networks has made "the ends of the earth seem near," and remote control of frequency converters via computer networks is also a future trend. Remote control of frequency converters through RS485 interfaces, RTU modules, and various network protocols allows for easy achievement of control objectives even in situations where on-site human operation is not feasible.
(4) Green frequency converter
With the introduction of sustainable development strategies, environmental protection has received increasing attention. The high-order harmonics generated by frequency converters can pollute the power grid. Addressing issues such as reducing noise during inverter operation and enhancing its reliability and safety are all attempts to solve through appropriate control methods, leading to the design of green frequency converters.
In conclusion, the control method of frequency converters is a problem worthy of study. It is hoped that with the joint efforts of insightful people dedicated to this work, domestically produced frequency converters will soon enter the world market and become first-class products.
Disclaimer: This article is a reprint. If it involves copyright issues, please contact us promptly for deletion (QQ: 2737591964). We apologize for any inconvenience.
