FOC motor control foundation

FOC Fundamentals of motor control

One 、FOC technology

1. What is? FOC？

FOC（field-oriented control） For field oriented control , Also known as vector control （vectorcontrol）, It is a kind of using frequency converter （VFD） Technology of controlling three-phase AC motor , Use to adjust the output frequency of the frequency converter 、 Size and angle of output voltage , To control the output of the motor . Its characteristic is that it can individually control the magnetic field and torque of the motor , Similar to the characteristics of other excited DC motor . Because the three-phase output current and voltage will be represented by vectors during processing , Therefore, it is called vector control . Is the current brushless DC motor （BLDC） And permanent magnet synchronous motor （PMSM） The best choice for efficient control .FOC Accurately control the size and direction of the magnetic field , Make the motor torque stable 、 Small noise 、 Efficient , And it has high-speed dynamic response . because FOC The advantages are obvious , At present, it has gradually replaced the traditional control methods in many applications , It has attracted much attention in the motion control industry .

FOC According to whether the motor has sensors or not, it can be divided into sensors FOC And sensorless FOC. For sensors FOC, Due to the sensor of the motor （ Generally encoder ） It can feed back the position information of the motor rotor , Therefore, the position estimation algorithm can not be used in the control , Relatively sensorless for control FOC Simple , But for motor applications with sensors , It often requires high control performance . For sensorless FOC, Because the motor does not have any sensors , Therefore, the position information of the motor rotor cannot be obtained by simply reading the measured value of the sensor , Therefore, it is necessary to collect the motor phase current in the control , Use the position estimation algorithm to calculate the rotor position . Although no feeling FOC It is difficult to control , But it can avoid the risk of sensor failure , And it saves the cost of sensors , At the same time, the wiring between the motor and the drive board is simplified . at present , non-inductive FOC It is mostly used in fan occasions .

Two 、SVPWM

1.SVPWM principle

SVPWM It's space vector pulse width modulation (Space Vector Pulse Width Modulation) For short , It is usually composed of six power switches of three-phase inverter , After a specific timing and commutation, the resulting Pulse width modulated wave , The final output waveform may be very close to the ideal sine waveform . The details are shown in the following figure ; On the left is the complex plane , The space vector , On the right is the sinusoidal waveform in the time domain ;

3、 ... and 、SPWM

1.PWM The basic principle of control

When a narrow pulse with equal impulse and different beginning is added to a link with inertia , The effect is basically the same . Impulse refers to the area of narrow pulse ; The effect is the same, and the output response waveform of the ring joint is basically the same .

2.SPWM The basic principle of control

（1） Principle introduction

SPWM The full name is (Sinusoidal PWM), Sinusoidal pulse width modulation is a very mature , A very wide range of technologies ;

The basic principle is the principle of area equivalence , In other words, when narrow pulses of equal impulse but different shapes are added to an inertial link , The effect is basically the same .

In other words, through a series of narrow pulses of different shapes , The integral of the corresponding time is equal ( The area is equal ), The end result is the same ;

therefore SPWM It is to input a pulse sequence of equal amplitude to equivalent sine wave , Therefore, the pulse time width with high output is basically sinusoidal ;

（2） Sampling method

The usual sampling method used here is ： Natural sampling and regular sampling ;

Four 、IQmath library

1. Basic introduction

Texas Instruments TMS320C280X Series of IQmath The library is C/C++ The program collects a highly optimized and accurate mathematical function library , And precisely in TMS320C280X The operation code that converts floating-point algorithm into fixed-point algorithm on the chip . These functions are often used in programs with intensive real-time computation and whose speed and accuracy are crucial . Using these functions, you can have better performance than using the same standard ANSI C Code written in language runs faster . In addition, it provides high-precision functions that can be used TI IQmath Library significantly shortens you DSP Application development time .

2. The difference between floating-point numbers and fixed-point numbers

Floating point numbers should be summed by Taylor series expansion , Therefore, there will be many multiplication and summation of floating-point numbers in the calculation process . Fixed point numbers are based on CORDIC Iterative algorithm of + Look up the table Algorithm （ Table value is placed in IQmath In the library file ）, The essence is to use the rotation transformation of plane coordinates + Iteration to achieve the solution of trigonometric function , And this process only uses left shift and right shift , Therefore, it hardly consumes computing time , So this algorithm is very fast .

3.CORDIC Algorithm principle

Single chip or FPGA And other embedded devices with weak computing power are still easy to add and subtract , But I want to calculate trigonometric functions （sin、cos、tan）, Even hyperbola 、 Index 、 A function as complex as logarithm , It will take some effort . Usually, the calculation of these functions requires a look-up table or approximate calculation （ Such as Taylor series approximation ） And other technologies to convert it into a way that hardware is easy to implement .

CORDIC(Coordinate Rotation Digital Computer, Numerical calculation method of coordinate rotation ) Algorithm is an algorithm that reduces complexity to simplicity , Replace multiplication by basic addition, subtraction and shift operations , Gradually approach the target value , Get the numerical solution of the function .

CORDIC The fundamental reason why the algorithm can speed up the operation is that it basically only uses left shift and right shift , Floating point numbers cannot be moved left or right 2 Power multiplication and division of . So only fixed-point numbers can enjoy this algorithm .