Pretty bad article. Its conclusions are broadly correct, but the way it gets there is not. There hasn't been any major change in motors- the major change has come in Variable Frequency Drives (VFDs). Older motors are hooked directly into the grid and deal poorly with changing speed and torque, and can drop drastically in efficiency. A motor can be 60% efficient at low load, and 98% efficient at its rated load.
VFDs are very complex pieces of circuitry- far more than you'd expect. Efficient drivers require a great deal of computation, and 16 or 32 bit processors are not uncommon. That also requires high-power, cheap silicon transistors which are only gradually taking over from simpler control schemes. They make a huge difference in a lot of cases.
The author is also very wrong on Synchronous Reluctance-assisted Permanent Magnet motors, but its hard to fault them on that; it's complicated even for many engineers. The purpose is not to increase the power density, it's to increase efficiency. The magnets act like a "cruise" motor. At low torque, they provide all of the rotor magnetization at a very high efficiency. At higher torque, the stator induces a stronger field into the rotor, causing it to act like a reluctance motor. That allows you to turn on extra power as demanded at the price of lower efficiency (the same as a reluctance motor).
If, instead, you just used a larger PM motor, it would be more expensive and it would also have an efficiency drop at low torque (where the motor spends most of its time operating). The magnets are highly efficient but they "set" the operating torque of the motor somewhat, so there is a loss at low power/high speed to hysteresis. A reluctance motor meanwhile never reaches the peak efficiency of a PM motor.
Anyway the article doesn't really say much convincing and feels mostly like fluff.
Just wanted to mention that Visedo also has its own line of VFDs utilizing IGBTs (insulated-gate bipolar transistors). The software is definitely a large part of the drive.
They aren't used for their cost, rather it's the fact that they are synchronous that is advantageous. It's possible to use an induction motor but it's much more complex and because of that less accurate. A reluctance motor and PM motor with the same pole count can be driven at the same speed, whereas the speed of an induction motor depends heavily on the voltage driving it, the current speed, and the torque on it. You can only solve the drive equations accurately at zero load.
Commutated DC motors could also be used but they are low efficiency.
VFDs are very complex pieces of circuitry- far more than you'd expect. Efficient drivers require a great deal of computation, and 16 or 32 bit processors are not uncommon. That also requires high-power, cheap silicon transistors which are only gradually taking over from simpler control schemes. They make a huge difference in a lot of cases.
The author is also very wrong on Synchronous Reluctance-assisted Permanent Magnet motors, but its hard to fault them on that; it's complicated even for many engineers. The purpose is not to increase the power density, it's to increase efficiency. The magnets act like a "cruise" motor. At low torque, they provide all of the rotor magnetization at a very high efficiency. At higher torque, the stator induces a stronger field into the rotor, causing it to act like a reluctance motor. That allows you to turn on extra power as demanded at the price of lower efficiency (the same as a reluctance motor).
If, instead, you just used a larger PM motor, it would be more expensive and it would also have an efficiency drop at low torque (where the motor spends most of its time operating). The magnets are highly efficient but they "set" the operating torque of the motor somewhat, so there is a loss at low power/high speed to hysteresis. A reluctance motor meanwhile never reaches the peak efficiency of a PM motor.
Anyway the article doesn't really say much convincing and feels mostly like fluff.