What Is Direct Torque Control?

Direct torque control (DTC for short) is a way for an inverter to control the torque of a three-phase motor. The method is to calculate the estimated motor flux and torque based on the measured motor voltage and current. After controlling the torque, the speed of the motor can also be controlled. Direct torque control is a patent of European ABB company. .

Direct Torque Control (DTC), some of the original foreign texts are also called Direct self-control-DSC, translated directly into direct self-control, this "direct self-control" idea is centered on torque Comprehensive control, not only for controlling torque, but also for
In direct torque control, the amplitude of the stator flux linkage of the motor is maintained at the rated value through the vector control of the above voltage. To change the torque, it can be achieved by controlling the angle between the stator and the rotor flux linkage. The included angle can be adjusted through the control of the voltage space vector. Because the rotation speed of the rotor flux is kept constant, the adjustment of the included angle can be achieved by adjusting the instantaneous rotation speed of the stator flux.
Assume that the motor rotor rotates counterclockwise. If the actual torque is less than a given value, a voltage vector that rotates the stator flux in a counterclockwise direction is selected, so that the angle increases and the actual torque increases. Once the actual torque is higher than the given value, Then select the voltage vector to rotate the stator flux in the opposite direction. As a result, the angle is reduced. By selecting the voltage vector in this way, the stator flux is always rotating, and its rotation direction is determined by the torque hysteresis controller.
The control of torque and flux by direct torque control is realized by a hysteresis comparator. The operating principle of a hysteresis comparator is: When the error between the current value and a given value is within the tolerance range of the hysteresis comparator, the output of the comparator remains unchanged. Once it exceeds this range, the hysteresis comparator gives the corresponding Value.
The principle block diagram of direct torque control is shown below. The given speed is compared with the estimated speed to get the given torque; the torque difference is processed by the torque regulator as a hysteresis to obtain the torque control signal; Compared with a given flux linkage, a flux linkage control signal is obtained via a hysteresis comparator; a segment is divided according to the calculated rotor displacement; a space is obtained based on the segment, the torque and flux linkage control signals, and a lookup table Vector, generate PWM wave; output to inverter, power motor [1]
Like the VC system, it also separately controls the speed and flux of the asynchronous motor. However, in terms of specific control methods, the DTC system and the VC system have different characteristics:
1) Torque and flux control uses a two-position bang-bang controller, and the two control signals are used directly in the PWM inverter to generate voltage SVPWM waveforms, thereby avoiding the breakdown of stator current into torque And magnetic flux components, eliminating the need for rotation conversion and current control, and simplifying the structure of the controller.
2) The stator flux is selected as the controlled quantity instead of the rotor flux as in the VC system. In this way, the model for calculating the flux is not affected by changes in rotor parameters, which improves the robustness of the control system. If the law of controlling the stator flux is derived from the mathematical model, it is obviously more complicated than when the rotor flux is oriented. However, because of the bang-bang control, this complexity does not affect the controller.
3) Due to the use of direct torque control, fast torque response can be obtained during the dynamic process of acceleration or deceleration or load changes, but care must be taken to limit the excessive inrush current so as not to damage the power switching device, so the actual torque The speed of response is also limited.
4) The mathematical model of the analysis motor in the stator coordinate system directly controls the magnetic flux and torque, and it does not need to be compared, equivalent, and converted to the DC machine, eliminating complicated calculations [2]
In view of its shortcomings, the current direct torque control technology has been greatly improved compared to the earlier direct torque control technology, which is mainly reflected in the following aspects:
(1) Research on speed sensorless direct torque control system
In practical applications, installing a speed sensor will increase the cost of the system, increase the complexity of the system, and reduce the stability and reliability of the system. In addition, the speed sensor is not used in harsh environments such as humidity and dust. Therefore, the research of speed sensorless has become an important research direction in AC drive system, and certain results have been obtained. There are many methods for rotor speed estimation. Kalman filter position estimation method, model reference adaptive method, flux linkage position estimation method, state observer position estimation method, and detection method of motor phase inductance change are commonly used. Some scholars start from the model reference adaptive theory and use the rotor flux equation to construct a speed sensorless direct torque control system. As long as the appropriate parameter adaptive law is selected, the speed identifier can accurately identify the motor speed.
(2) Influence of change in stator resistance
One of the core problems of direct torque is the stator flux observation, and the stator flux observation uses the stator resistance. Using a simple ui flux model, in the high-speed region, the change of the stator resistance can be ignored, and applying the ui flux model of the flux can achieve satisfactory results;
However, the change in stator resistance at low speed will affect the distortion of the magnetic flux and make the system performance worse. Therefore, if the stator resistance can be identified online, the impact of the change in stator resistance can be eliminated fundamentally. At present, the commonly used methods are reference model adaptive method, Kalman filter method, neural network and fuzzy theory to construct online observer to compensate stator resistance. The research results show that online identification is an effective method.
(3) Improvement of magnetic flux and torque hysteresis
Traditional direct torque control generally adopts single hysteresis control for torque and flux linkage, and determines the voltage vector based on the result of hysteresis output. Because different voltage vectors have different adjustment effects on torque and stator flux linkage, only reasonable selection can be made based on the current torque and flux linkage real-time values.
Only the voltage vector can make the adjustment process of torque and magnetic flux reach a more ideal state. Obviously, the finer the difference between the torque and the flux deviation, the more accurate the voltage vector selection, and the better the control performance.
(4) The solution of the dead zone effect
In order to avoid the DC side short circuit caused by the simultaneous conduction of the upper and lower bridge arms, it is necessary to introduce a sufficiently large interlocking delay, which results in a dead zone effect. The errors accumulated by the dead-band effect distort the inverter output voltage, which in turn produces current distortion, exacerbating problems such as torque ripple and system instability. At low frequencies and low voltages, the problem is more serious and it also causes torque ripple. The correction of the dead zone effect can be detected and recorded by the compensation circuit to compensate. This increases the cost and reduces the reliability of the system. The method that can be implemented by software is to calculate all the distortion voltages, make a compensation voltage command table according to the current direction, and then use the forward feedback to compensate. This new scheme also eliminates the zero voltage clamping phenomenon. In addition to the above major aspects, some scholars have also tried to improve the performance of the system through other channels.
The characteristic of direct torque control is to control the stator flux, which directly calculates and controls the flux and rotation of the motor in the stator coordinate system by using the space vector concept and the detected stator voltage and current directly under the stator stationary coordinate system. Torque to obtain high dynamic performance of torque. It does not need to convert the AC motor into an equivalent DC motor, so it eliminates many complicated calculations in vector transformation, it does not need to imitate the control of the DC motor, and it does not need to simplify the mathematical model of the AC motor for decoupling. Only need to care about the magnitude of the electromagnetic torque, so the control is robust to all motor parameters except the stator resistance. The introduced stator flux observer can easily obtain the flux model and easily estimate the synchronization speed Information, it is also easy to obtain the torque model, the flux linkage model and the torque model constitute a complete motor model, so it can easily achieve speed sensorless control. If you set a speed regulator in the system, you can get further High performance dynamic torque control.
It should be noted that the inverters for direct torque control use different switching devices and control methods are also different. The direct self-control theory originally proposed by Depenbrock is widely used in inverter control with high voltage, high power and low switching frequency. The control method currently applied to general inverters is an improved method suitable for high switching frequency inverters. The first ACS600 series direct torque control universal inverter introduced by ABB in 1995, the dynamic torque response speed has reached <2ms, the static speed accuracy reaches 0.001% when the speed sensor PG is used, and in the case without the speed sensor PG Even if it is affected by the change of input voltage or sudden change of load, it can also achieve the speed control accuracy of ± 0.1%. Other companies also aim at direct torque control. For example, the FRENIC5000VG7S series of high-performance speed sensorless vector control universal inverters from Fuji Company, although it is different from the direct torque control method, it has also achieved speed control accuracy ± 0.005%, speed response 100Hz, current response 800Hz and torque control accuracy ± 3% (with PG). Other companies such as Japan's Mitsubishi, Hitachi, Finland VASON and other latest series products have adopted a design similar to the speed sensorless control, which has further improved the performance [3] .

IN OTHER LANGUAGES

Was this article helpful? Thanks for the feedback Thanks for the feedback

How can we help? How can we help?