Design of Switched Reluctance Motor Monitoring System Based on Virtual Instrument

The structure and working principle of the switched reluctance motor (SRM) are relatively simple, with good fault tolerance. At low speed, only a small current is needed to obtain a large torque. At high speed, the constant power range is wide, and it can be used in places with harsh working environments such as coal mines, textiles, chemicals, and electric vehicles. However, due to the double salient pole structure of the SRM stator and rotor, the non-sinusoidal characteristics of the winding current, and the deep saturation of the core flux density, it is difficult to achieve smooth control of the SRM, especially in reducing the noise during its operation. At present, the common SRM control system often focuses on a single aspect of motor performance and cannot achieve comprehensive adjustment of multiple parameters. If the control system can reflect the important parameters of the SRM in real time during operation and conduct a comprehensive analysis, the work efficiency will be greatly improved. The virtual instrument has powerful data processing capabilities and a friendly operation interface. At the same time, its development cycle is short and its size is small, and it has received more and more attention.

The data in this paper is obtained by using the PCI-6143 data acquisition card launched by NI Corporation of the United States and programming in LabVIEW8.6 as the development environment. Considering that the system may be used in a relatively harsh working environment, in order to achieve real-time control more safely and effectively, DSP is used as the backup processor of the PC. The experimental prototype is an 8/6-pole SRM with a power of 150 W.

1 Composition of the switched reluctance motor monitoring system

The SRM operation monitoring system is mainly composed of the SRM drive system and various sensors, data acquisition cards, PCs and DSPs. The system structure diagram is shown in Figure 1.

1.1 Phase voltage signal detection

Phase voltage is an important parameter that reflects the start-up, stable operation, speed regulation or braking status of SRM. The phase voltage is measured by using Hall voltage sensor as the main measurement circuit. The primary coil of the Hall voltage sensor is connected in parallel to the two ends of a phase winding of the motor. In order to make it work in the best state, a resistor of appropriate size should be connected in series with the primary coil, preferably an adjustable resistor. The secondary coil of the Hall voltage sensor is connected in series with a precision resistor of appropriate resistance value, which is connected to the data acquisition card after being processed by the operational amplifier.

Considering that the Hall sensor is easy to burn out when overvoltage occurs, a resistor voltage divider circuit is used as a backup voltage measurement circuit. By measuring the voltage across the voltage divider resistor, the SRM phase voltage can be calculated. This method is simple and easy, but the accuracy is relatively low. It should be noted that the measurement circuit needs to be isolated from the main circuit by a photocoupler to prevent the two circuits from affecting each other.

1.2 Phase current signal detection

According to the law of electromagnetic induction, there is a magnetic field around the current-carrying wire, and its size is proportional to the current in the wire. Therefore, the Hall effect can be used to measure the magnetic field, and the size of the current in the wire can be determined. The Hall current sensor can accurately measure the SRM phase current, and the measurement circuit is isolated from the main circuit, without electrical contact, which is a safe measurement method. When overcurrent occurs, the Hall current sensor is also more easily damaged. Therefore, a resistor with a relatively small resistance but high accuracy and power is connected in series in each phase winding. By measuring the voltage at the resistor terminal, the winding current can be detected. Of course, its accuracy is also relatively low, but it can also be used as a backup circuit for current measurement.

The phase current waveform of SRM varies greatly depending on the operating mode and operating conditions, and pulsation may occur. In order to reflect the current changes as realistically as possible, the sampling frequency of the data acquisition card needs to be set relatively high, preferably above 10 kHz. In addition, the detection circuit should also have the characteristics of fast performance, wide detection frequency band, and good isolation between the main circuit and the control circuit. The current detection circuit mainly realizes two functions: current observation and overcurrent protection.

1.3 Vibration signal detection

The switched reluctance motor is a double-pole structure, and the noise is relatively large during operation. To reduce the noise, it is necessary to avoid the motor operating at a frequency when its vibration is relatively severe. In addition to changing the base size and other methods, this can also be achieved by changing the motor running speed. In addition, although the SRM has a strong fault tolerance performance and can still operate when the phase is missing, the noise at this time is relatively large. Long-term phase-missing operation is more damaging to the SRM. Therefore, the vibration signal is also the most commonly used characteristic signal in motor fault diagnosis. In this system, the core part of the vibration information acquisition device is the piezoelectric crystal accelerometer, which is firmly fixed in the middle of the SRM casing. When the motor is running, it will generate an electric charge corresponding to the vibration, which will be converted into a voltage signal by the charge amplifier and input into the data acquisition card, and then the spectrum analysis will be performed to determine the vibration frequency of the motor. Finally, based on historical data, determine whether the operating speed needs to be changed at present.

1.4 Speed ​​signal detection

The photosensitive rotor position sensor is used to measure the speed, including a photoelectric pulse generator and an aluminum turntable. The number of teeth and slots of the turntable is equal to the number of salient poles and grooves of the rotor and is evenly distributed. The experimental object of this system is an 8/6-pole four-phase SRM, so the number of teeth and slots of the turntable is 6, and they are spaced 30° apart. The angle between the two photoelectric pulse generators is 75°, and they are fixed at 37.5° on the left and right sides of the stator pole centerline. The turntable is concentrically fixed on the rotor shaft and rotates synchronously with the rotor shaft. When the motor is running, the working states of the two photoelectric sensors are: 00-01-11-10-00, and they are constantly circulating. The signal is a TTL level signal, which is connected to the counter port of the data acquisition card, and then the signal frequency is obtained using the LabVIEW measurement frequency module to calculate the speed.

1.5 Position signal detection

When working under very harsh conditions, the rotor position sensor may fail. In this case, a safer working mode can be achieved without position sensors. At present, the more common method is to deduce the angle of the motor by measuring the flux and current. However, this requires obtaining the flux and current values ​​of the SRM at different positions in advance as the basis for subsequent judgment during operation. If the system needs to be transformed into a position sensorless technology in the future, in order to accurately obtain the rotor position, an absolute position encoder needs to be fixed on the rotor shaft. The output of the absolute position encoder is Gray code, which needs to be processed to obtain ordinary binary code. Of course, in steady-state operation, the absolute position encoder can also completely replace the photoelectric position sensor to provide more accurate speed information. However, compared with the photoelectric position sensor, the absolute position encoder is expensive and easy to damage, and is not suitable for occasions with severe vibration. The circuit structure of the experimental motor is shown in Figure 2. Among them, R1~R4 are small resistors connected in series with the four phases of the motor, and R5 is a voltage divider resistor for measuring the winding voltage.

1.6 Things to note

(1) Since the SRM is 8/6 poles, phases A and C are not turned on at the same time, so these two phases can share a current sensor. Similarly, phases B and D can also share a current sensor.

(2) The signal obtained by the sensor will inevitably contain some noise. If necessary filtering is not performed, the result may be quite different from the actual value. The Filter module of LabVIEW provides some commonly used filtering methods, which can easily remove noise.

(3) When selecting voltage and current sensors, you should pay attention to their measurement range. At the same time, in order to ensure sufficient accuracy, the voltage sensor and current sensor should work in the best state. 2 Torque and flux value display

Virtual instruments can not only quickly perform data acquisition tasks, but also achieve strong data processing capabilities with the help of PCs. After obtaining the speed, terminal voltage and phase current data, PCs can be used to perform real-time calculations to obtain and display the flux and torque values.

2.1 Magnetic flux calculation

The magnetic flux value can be obtained according to formula (1):

Where: Ts is the interval between two samplings; R is the resistance of the motor winding.

In order to reduce the data calculation time as much as possible and improve the running speed, the flux value can also be obtained by table lookup method. That is, the flux value ψ(i,θ) under different currents i and angles θ is first stored, so that the flux value under different currents and angles can be quickly obtained by linear fitting method during operation. Of course, the result of this processing is that the error increases. However, the error is still within an acceptable range [6].

2.2 Torque calculation

The calculation principle of torque is shown in formula (2):

Where: J is the system moment of inertia; ω is the motor speed; B is the viscous friction coefficient; TL is the load torque.

3 Run analysis

The state monitoring diagram of the motor during light load stable operation is shown in Figure 3. For ease of viewing, only the single-phase monitoring situation is shown in the figure. Assume that the worst case in the work is that the PC cannot work properly and the Hall voltage sensor and Hall current sensor are damaged. At this time, the spare DSP will start to work instead of the PC. When the load is 1.25 NM, the speed is 1 200 rad, the conduction angle is 4°, and the break angle is 20°, the voltage and current waveforms are shown in Figure 4.

In Figure 4, the horizontal axis is 1 ms/grid, the vertical axis is 100 V/grid for the voltage curve, and 3 A/grid for the current curve. The reason for the burrs on the voltage curve in Figure 4 during the forced commutation stage is that although the switches at both ends of the winding are disconnected, the magnetic field energy of the winding is still released through the diodes at both ends. Since a general bridge rectifier circuit is used as a DC power supply, its output voltage cannot be guaranteed to be constant. Therefore, during the forced commutation stage, the voltage at the sampling resistor end is unstable.

4 Conclusion

The monitoring system designed in this solution intuitively and accurately displays the important parameters of the motor operation. Compared with the traditional use of various devices to observe data, it greatly improves work efficiency and saves costs. The system can modify parameters such as the conduction angle, the turn-off angle and the pulse time interval in real time according to the operation of the SRM to control the running speed and output torque of the motor. Experiments have shown that the practicality and reliability of the control system are greatly improved than before. In addition, the backup detection circuit and DSP operation system also increase the reliability of the system.