To control the speed of an induction motor, the frequency of the 3 phase output can be varied to control the motor speed. The drive voltage must also be ramped so that the motor is not over/under excited. The motor picks up torque by the rotor slip increasing.
2) Current Source Inverter
The inverter still has the voltage source power stage but the current is sensed in the motor leads and through a PI control loop, the current in the motor leads is made to match a commanded current. The motor torque (sort of) is proportional to motor current. To make the scheme work, the drive frequency and voltage must be controlled in such a way that the torque producing current and rotor flux producing current are both controlled.
Current mode control allows for faster motor responce if this is required.
The main difference between a Voltage Source Inverter (VSI) and a Current Source Inverter (CSI) is that the former uses a stiff DC link voltage (lots of capacitors to keep voltage ripple down) while the CSI has a stiff direct current link (one or two big inductors to smooth the current instead of the voltage).
Operational charateristics are quite different. The VSI usually is a PWM inverter with IGBTs (sometimes power FETs) while the CSI often has thyristors and a much lower switching frequency. CSI inverters are inherently four quadrant drives. Since current is controlled by the input SCR rectifier and the direction of current never changes, braking is achieved by reducing rectifier voltage to negative levels, which means regeneration.
CSI inverters are quite simple, but those who are used to VSI (i.e. "normal" inverters) may need some time to adjust their thinking about the operation. Hope that this short introduction hasn't confused you too much.
Thank you for your information.
Can we say V/f control is voltage source inverter and sensorless/sensor vector control is current source inverter? And what is the difference between operatinal characteristics?
CSI drives, older technology largely abandoned by industry and now only available from limited suppliers at higher costs.
VSI/PWM drives, up-to-date dominant technology produced and supported by a host of manufacturers, research and service organizations, creating a highly competitive marketplace that favors the end user.
Both have merits, but unless there is something specific that you need from a CSI drive, you are often better off with a PWM.
"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more."
Nikola Tesla
The difference between U/f and vector control is that the U/f simply outputs a frequency and a voltage. The voltage is frequency-dependent in a prescribed manner. Usually U is proportional to frequency i.e. U/f is a constant. The normal European voltage and frequency (400 V at 50 Hz) makes this constant equal to 400/50 = 8.0 V/Hz. There are also so-called "fan curves" where U increases with f squared. This is to avoid overmagnetization at lower frequencies, which would be a waste of current because the fan doesn't need much torque at lower speeds and thus less magnetic field.
U/f inverters need a little more than proportional voltage at lower frequencies (because resistive components in the windings get more dominant at low frequencies). So "boost" is introduced. Boost means that the voltage doesn't go to zero as frequency does. You will usually have between 2 and 5 percent voltage at zero Hz and then increasing linearly from there.
A U/f inverter has poor characteristics at lower frequencies. Do not expect to have full torque available below about 10 percent of base speed.
The vector drive is much better in this respect. There is no fixed relationship between frequency and voltage - it depends on load. It is not uncommon to find motors running at close to half rated voltage when lightly loaded.
The reason for this is that magnetizing current (vector Id) is controlled by one loop and the torque producing current (vector Iq) by another loop, the torque loop.
The two vectors are 90 degrees apart (magnetizing current, Id, lags active current, Iq, by 90 degrees) and the resulting current (vector addition, Phytagoras) is the current you actually measure with a current clamp (with HF filter because of PWM interference).
What makes a vector drive superior is that it controls torque in a very straightforward manner. And since good torque control is a prerequisite for good speed control - you also get a fast and crisp speed control. Which is not possible with a U/f inverter.
Professors W. Leonhard and P. Vas have both written good books on these subjects. I think that I prefer Leonhard over Vas. Both have, by the way, been involved with the NFO algorithm (Natural Field Orientation, which is an efficient algorithm that has proven very useful around zero Hz)
I think that I leave the VSI/CSI discussion for a later occasion - or someone else...
The only thing I want to add to skogsgurra excellent explaination of drives is a visiualization point.
The Vector in Vector Drive refers to the position of the magnetic vector on the rotor. In order to control the torque of an induction motor one needs to know the magnatude and position of the flux vector on the rotor. It was shown matematically that you could determine this through use of a "Motor Model" and a shaft position sensor by putting the coordinate system on the rotating rotor and doing a transform to stator coordinates.
This now called (with variations) "Flux Vector Control" and it works very well but you need an "Expensive?" shaft position sensor.
Later it was shown that the same thing could be done without a shaft position sensor with little loss in performance especially over a speed range that did not include zero speed.
