Principle and Simulation of PMSM Position Sensorless Control Based on Pulse Vibration High Frequency Voltage Signal Injection Method
Compared with the rotating high-frequency voltage signal injection method, the pulse-vibration high-frequency voltage injection method only injects a high-frequency sinusoidal voltage signal on the d-axis in the estimated synchronously rotating dq coordinate system, which is a pulse-vibration signal in the stationary coordinate system. voltage signal.
1. Current response of three-phase PMSM excited by pulsed high-frequency voltage
In order to accurately estimate the rotor position of the motor, first establish the relationship between the estimated rotor synchronous rotation coordinate system d -q and the actual rotor synchronous rotation coordinate system dq, as shown in Figure 6-8.
In the figure, a-β is the two-phase stationary coordinate system, θ^e is the estimated rotor position angle, and θe is the actual rotor position angle. The angle θ~e between the estimated rotor synchronous rotation coordinate system d ^ -q ^ and the actual rotor synchronous rotation coordinate system d -q is the estimated error angle of the rotor: rewrite the three-phase excitation under high frequency excitation in the synchronous rotation coordinate system
dq The voltage equation of PMSM:
In the synchronous rotating coordinate system dq, the stator inductance of the motor can be expressed as
In the stationary coordinate system α-β, the above formula is transformed into
the relationship between high-frequency voltage and current in the estimated rotor synchronous rotating coordinate system dq In order to
rewrite the above formula:
the pulse vibration high-frequency voltage injection method only injects a high-frequency sinusoidal voltage signal into the à-axis of the estimated rotor synchronous rotation coordinate system α-q: where: uin is the
amplitude of the high-frequency voltage signal, oin is The frequency of a high-frequency voltage signal.
At this time, the high-frequency current can be simplified as
it can be seen that if there is a difference between the ′-axis and the q-axis inductance (AL≠0), then in the estimated rotor synchronous speed rotating coordinate system, the magnitude of the d-axis and q-axis high-frequency current components The values are related to the rotor position estimate error angle 6. related. When the rotor position estimation error angle is zero, the q-axis high-frequency current is equal to zero, therefore, the q-axis high-frequency current can be properly processed and used as the input signal of the rotor position tracking observer to obtain the rotor position and speed .
2. Rotor position estimation method
2.1. Rotor position estimation method based on tracking observer
In order to obtain the position and speed of the rotor, the amplitude modulation of the q-axis high-frequency current can be performed first, and the input signal of the rotor position tracking observer can be obtained after passing through a low-pass filter (LPF). That is, if the rotor position estimation error is small enough,
then The error signal can be linearized, that is,
where:
From the above formula, it can be seen that if f(θe) is adjusted to zero, the estimated error of the rotor position angle is also zero, that is, the estimated value of the rotor position converges to Actual value
The structural block diagram of the sensorless control system using pulse vibration high-frequency voltage signal injection is shown in the figure below. In the figure, T(θ^e) is the transformation matrix that converts the stationary coordinate system to the rotating coordinate system, and T'(θ^e) is its inverse matrix; a band-pass filter (BPF) is used to extract the rotor position information High-frequency current signal; the above equation is filtered using a low-pass filter (LPF) to obtain rotor error information. In order to obtain the motor speed and rotor position information, the rotor position tracking observer method can also be used. The realization principle of this method has been explained in my other high-frequency injection article, so I won’t go into details here.
2.2. Rotor position estimation method based on PLL
In addition to using the rotor position tracking observer to estimate the rotor position information, another commonly used estimation method is the PLL-based rotor position estimation method, and its control block diagram is shown in the figure below.
In order to obtain the rotor position angle of the motor, a PI regulator is used to form a PLL system, and its control block diagram is shown in the figure below. Among them, the LPF filter adopts the form of a first-order low-pass filter with an expected bandwidth of o (this symbol is not good to refer to the following formula), and its transfer function can be expressed as: The transfer function of the PI regulator adopts the following form: According to the
above
figure The control block diagram shown in the figure, its closed-loop transfer function is:
After the pole configuration:
3. Simulation (based on position tracking observer instead of PLL)
*
3.1. Speed loop
3.2, current loop (both are the same)
3.3. Inverter module
where subsystem2
3.4, PMSM (build by yourself, not the system you use)
where fcn1
where fcn2
where fcn3
where the parameter values are:
3.5. Observer module (observer)
3.6. Filter module
where LPF is:
where BPF is:
3.7. Pulse vibration high frequency voltage injection module
3.8, mod (remainder module)
4. Simulation results
4.1. Actual speed and estimated speed
4.2. Error between actual speed and estimated speed
4.3. Actual rotor and estimated rotor
4.4. The error between the actual rotor and the estimated rotor
V. Conclusion
I don't know why the rotor tracking is very good, but the error suddenly changed from 0 to 6 in 0.3 seconds