Unity power factor control of permanent magnet synchronous motor by using open winding configuration

Abstract

This paper presents the performance of open-winding permanent magnet synchronous motor (OW-PMSM). It mainly includes vector modulation technology considering the unity power factor control.And a topology structure is proposed to optimize the fault tolerance of inverter. Matlab software and Maxwell software collaborative simulation are supplied to obtain the reactive power, speed, terminal voltage, electromagnetic torque etc. under normal and fault status. Finally, the simulation results of an open-winding permanent magnet synchronous motor are verified by the experimental results.

Keywords

Open-winding permanent magnet synchronous motor fault tolerance unity power factor control

1. Introduction

OW-PMSM system opens the neutral point of the motor stator and connects a DC-to-AC converter in series, making the stator winding to supply power at both ends by two inverters. The first inverter provides the active power required for the normal operation of the motor. The second inverter provides reactive power for the motor by capacitor [1]. The advantage of the open-winding system is that the operating performance of the motor is significantly improved and the unit power factor operation of the main inverter is realized on the premise of the motor electromagnetic design and structure unbroken [2–4].

The composition of the OW-PMSM system is shown in Fig. 1. Inverter 1 provides active power required for the motor and Inverter 2 compensates the reactive power required by motor [5,6]. The compensating inverter 2 uses capacitor instead of power supply to supply the reactive power needed by the motor, and the capacitor coupled circuit suppresses zero sequence current, which also saves cost and the system is stable. Therefore, this paper used capacitor to supply power to the compensated inverter. Through the strategy proposed in this paper the DC link voltage increases and the maximum speed of motor is extended.

Fig. 1.

OW-PMSM drive system.

This paper focuses on the normal operation of the motor. The normal operation of motor can be divided into two states: constant torque operation and constant power operation. In constant torque region, maximum torque per ampere (MTPA) is taken to control system. In the flux weakening region, the traditional feedback flux weakening control is adopted. By optimizing the phase relation of voltage and current, a new open winding machine control scheme is proposed in this paper. The operation efficiency of the open winding is improved through perfection. And the fault-switch fault-tolerant control is used to ensure the steady operation of the motor. Finally, simulation and experiment verifies the feasibility of the strategy.

2. Compensation theory and system structure

Fig. 2.

Stator equivalent circuit and steady state vector diagram.

In Fig. 2(a), a PMSM equivalent circuit and a phasor diagram for the motor equivalent circuit are shown. The PMSM equivalent circuit includes stator resistance R, stator leakage reactance jX_sσ, excitation reactance jX_m and the induced electromotive force E₁. In order to better analyze the running state of the motor, the resistance voltage drop caused by the stator resistance R is ignored. The power supply V_s is composed of reactive voltage jI₁X_s and induced electromotive force E₁. The phase difference between V_s and I₁ means that the power supply provides both the reactive power and active power required by the motor, resulting in low utilization of the power supply.

Figure 2(b) shows the equivalent circuit of the motor after adding reactive power compensation. The reactive power provided by jI₁X₁ and the reactive components of E₁ is equivalent compensated by a series of equivalent capacitors X_c to make the power supply V_s unit power factor run. The jI₁X_c provided by the capacitor compensates the reactive power of the motor without affecting the motor itself. As can be seen from the equivalent figure, V_s is less than V₁, which means that the motor of the compensation strategy uses a smaller power supply to achieve normal operation.

3. Control strategy of the OW-PMSM

In order to make the inverter 1 work in the state of maximum power factor, proposed the block diagram of control strategy as shown in Fig. 3.

Fig. 3.

Block diagram of the proposed control strategy.

In Fig. 3, the d-axis and q-axis current references are generated by maximum torque per ampere (MTPA) strategy. The error between the given speed ω and the feed-back speed ω∗ is adjusted by PI to get the current Is. The motor voltage vector V_d, V_q is obtained by using conventional position feedback with an encoder. The connection between the constant torque and constant power region is realized by comparing the voltage V_d, V_q with the bus voltage V_dc. After PI adjustment, the voltage $V_{d}^*, V_q^*$ is added with the compensation voltage to obtain the active voltage required by the motor. The active voltage required by the main inverter 1 can be obtained by subtracting the reactive voltage V_dQ, V_qQ obtained by the reactive power distribution module in inverter 2. The capacitance provides the required reactive power and partial active power of the motor through inverter 2. The voltage vector V_cd, V_cq required by inverter 2 can be obtained through the reactive power distribution module.

The reactive power of the motor can be obtained according to the running state of the motor, as follows: $\begin{eqnarray}\displaystyle Q=3/2\ast \left(V_{\text{q}}\ast I_{\text{d}}-V_{\text{d}}\ast I_{\text{q}}\right). & & \displaystyle\end{eqnarray}$ (1) In the actual operation of the motor, in order to obtain relatively stable motor operation performance, it is necessary to maintain the stability of the DC side voltage of the secondary inverter. If the capacitor terminal voltage is changed at any time according to the running state of the motor, it will lead to a large active power loss in the charge and discharge of the capacitor, and the stability of the motor operation cannot be guaranteed. Therefore, the capacitor terminal voltage should be kept at a constant value.

The real power flow to the inverter 2 is controlled to regulate the bridge capacitor voltage using an error signal and a PI controller: $\begin{eqnarray}\displaystyle P=\left(K_{\text{p}}+K_{\text{i}}{\displaystyle \frac{1}{\text{p}}}\right)(V_{\text{c}}\ast -V_{\text{c}}). & & \displaystyle\end{eqnarray}$ (2)

Capacitors provide a small amount of active and most reactive power in an open winding system. V_cd and V_cq are components of capacitor voltage on d-q axis. Then decompose V_cd and V_cq into the form of active and reactive voltage provided on dq axis: $\begin{eqnarray}\displaystyle \text{} & \text{} & \displaystyle V_{\text{cd}}=-(V_{\text{dP}}+V_{\text{dQ}})\end{eqnarray}$ (3) $\begin{eqnarray}\displaystyle \text{} & \text{} & \displaystyle V_{\text{cq}}=V_{\text{qP}}+V_{\text{qQ}}.\end{eqnarray}$ (4)

When calculating the active power provided by the capacitor, consider the capacitor as a resistance, and the active voltage on the dq axis is the same as the stator current phase, it can be deduced that: $\begin{eqnarray}\displaystyle V_{\text{dP}}=I_{\text{d}}/I_{\text{q}}\ast V_{\text{qP}}. & & \displaystyle\end{eqnarray}$ (5)

When calculating the reactive power provided by the capacitor, take the capacitor as an ideal capacitor, and the capacitor voltage lags the motor current by 90 degrees. $\begin{eqnarray}\displaystyle V_{\text{qQ}}=-I_{\text{d}}/I_{\text{q}}\ast V_{\text{dQ}}. & & \displaystyle\end{eqnarray}$ (6)

Substitute (7), (8) into (3), (4), then get the following: $\begin{eqnarray}\displaystyle \text{} & \text{} & \displaystyle V_{\text{qP}}=2/3\ast P/(I_{\text{q}}+I_{\text{d}}\ast I_{\text{d}}/I_{\text{q}})\end{eqnarray}$ (7) $\begin{eqnarray}\displaystyle \text{} & \text{} & \displaystyle V_{\text{dQ}}=-2/3\ast Q/(I_{\text{q}}+I_{\text{d}}\ast I_{\text{d}}/I_{\text{q}}).\end{eqnarray}$ (8)

4. Study of vector control system in the OW-PMSM

Table 1 shows the basic parameters of OW-PMSM. Figure 4 shows the comparison of traditional PMSM and OW-PMSM. As shown in the Fig. 4, the motor terminal voltage increased by 43.3% and the maximum speed extended by 26.3%.

Table 1
The basic parameters of OW-PMSM

Parameter Symbol Value

DC bus voltage V _dc 311 V

Rated power P _rated 2 kW

Number of pole pairs P 8

Rated torque Te 20 Nm

Rated angular speed ω 104.7 rad/s

Stator d-axis inductance Ld 4.97e-3H

Stator q-axis inductance Lq 4.97e-3H

Stator resistance R 0.77 Ω

Parameter	Symbol	Value
DC bus voltage	V _dc	311 V
Rated power	P _rated	2 kW
Number of pole pairs	P	8
Rated torque	Te	20 Nm
Rated angular speed	ω	104.7 rad/s
Stator d-axis inductance	Ld	4.97e-3H
Stator q-axis inductance	Lq	4.97e-3H
Stator resistance	R	0.77 Ω

Fig. 4.

Comparison of traditional PMSM and OW-PMSM.

Since the motor terminal voltage is synthesized by the output voltage of inverters on both sides, the motor terminal voltage can break the maximum voltage limit reached by the single inverter when the output of the main inverter is constant. The degree of voltage increase is related to the power factor of the motor. The lower the power factor, the greater the voltage amplitude increase and the greater the motor’s operating range expansion.

As the PMSM drive system in this paper is different from the traditional PMSM drive system of three-phase six-switch inverter, the stator current flows into the DC capacitance midpoint, which leads to periodic fluctuation of capacitor voltage and imbalance of stator current, and also increases torque pulsation. As shown in the Fig. 4(c), torque pulsation of the open-winding motor in the low-speed zone is larger than in the high-speed zone.

Figure 5 is the joint simulation of OW-PMSM based on control strategy. And the magnetic field of stator yoke increases due to the increase of voltage. Figure 6 shows the reactive power Q of OW-PMSM under different speeds.

Fig. 5.

Magnetic flux density distribution of OW-PMSM.

Fig. 6.

Reactive power Q at different speeds.

As shown in Fig. 6, the higher the motor speed under this strategy, the more stable the compensation intermediate variable Q is, and the better the compensation effect is. Under the support of the capacitor of one side inverter, reactive power can be output for compensation. Through side converter control capacitor compensating reactive power needed for the motor speed, make the output voltage of the DC power supply for the most part or all of the used for output active power, thus greatly improve the power density, increase output torque at low speed, high output power, so as to solve the single insufficient voltage converter motor speed limited. But this topology inevitably increases the hardware cost and control system complexity. The problem of power electronic device loss and electromagnetic compatibility caused by multi-inverter can’t be ignored.

5. Study of system fault tolerant control in the OW-PMSM

In addition, power electronic devices are prone to failure due to high loss, high temperature rise and other problems. Fault tolerant switch is used to switch and disconnect the fault bridge arm during operation, and the motor continues to run in the original state.

Fig. 7.

Fault tolerant topology of motor after inverter failure.

Figure 7(a) is the fault tolerant topology of inverter 1. When four switches are used to replace the main inverter, capacitors replace T1 and T4 bridge arms, and the main inverter 1 forms a three-phase four-switch topology. At the same time, the main inverter control strategy is switched to three-phase four-switch SVPWM control strategy.

Since the capacitor needs to be charged and discharged to form the bridge arm instead of the power supply, and the amplitude of the capacitor is easily changed by the change of the motor state, the time required for the initial start of the motor is relatively prolonged, and the overshoot of the speed is large. When the topology reaches the maximum speed of expansion, it is limited by the capacity of capacitor power supply, the switching time increases, the overshoot increases, and the motor runs stably after the capacitor charge and discharge is stable.

Figure 7(b) is the fault tolerant topology of inverter 2. When replacing secondary inverter 2 with four switches, capacitance replaces T5’, T2’. The secondary inverter is mainly responsible for compensation function, and the capacitance is prone to fluctuation due to motor state fluctuation. The compensation effect of the open-winding system is poor at low speed, and the periodic fluctuation of the capacitor compensation is large, which leads to the voltage pulsation of the capacitor and the increase of torque pulsation at low speed. The compensation effect is effective in high-speed zone, the inverter compensation capacity is stable, the torque ripple is basically eliminated.

Fig. 8.

Speed and torque comparison of OW-PMSM fault tolerant system.

Fig. 9.

Phase current comparison of OW-PMSM fault tolerant system.

In Figs 8 and 9, the fault-tolerant system proposed in this paper can realize the normal operation of the open winding motor, has good robustness, and can make the open winding system low-cost.

6. Experiment about open winding control system

The OW-PMSM experimental system is conducted to verify the effectiveness of simulation, and experimental system is shown in Fig. 10, it includes IGBT drive, DSP board, oscilloscope, DC power supply, and OW-PMSM, load etc. Figure 11 shows the experimental results of the line voltage comparation about traditional PMSM and OW-PMSM.

Fig. 10.

Experimental system.

Fig. 11.

Experimental results of the terminal voltage and speed.

Figure 11(a) shows the terminal voltage of OW-PMSM and traditional PMSM. Through the experiment, the terminal voltage of the open winding is increased by 33.6% compared with the general control system. Figure 11(b) shows the speed curve of OW-PMSM. The stability of the system is verified by adding rated load suddenly when the motor is running without load.

7. Conclusion

In this paper, an unity power factor control system and fault tolerant control system of permanent magnet synchronous motor based on the open winding configuration and three-phase four-switch topology. The block diagram of the proposed control strategy of OW-PMSM were first analyzed in detail. Then the simulation and experiment platform for the proposed inverter were built and the simulation and experiment ware carried out. As result, the simulation and experimental results have validated the performance and effectiveness of the inverter. Therefore, the proposed open winding inverter with low cost has great competitiveness in applications of motor control and the fault tolerance.

Footnotes

Acknowledgements

This research was jointly supported by the Project of Liaoning Education Department (No. 201634090), and project Support of Liaoning Science and Technology Department (No. 20180550037).

References

Welchko

B.A.

and Nagashima

J.M.

, The influence of topology selection on the design of EV/HEV propulsion systems, IEEE Power Electronics Letters1 (2003), 36–40.

Pan

Liang

Wang

and Lipo

T.A.

, Extension of the operating region of an IPM motor utilizing series compensation, IEEE Transactions on Industry Applications50 (2014), 539–548.

Lin

Sun

and Chen

, Research on high-speed operation of hybrid-inverter fed open winding permanent magnet synchronous motor, in: International Conference on Electrical Machines and Systems, 2013, pp. 1179–1183.

Ewanchuk

Salmon

and Chapelsky

, A method for supply voltage boosting in an open-ended induction machine using a dual inverter system with a floating capacitor bridge, IEEE Transactions on Power Electronics28 (2013), 1348–1357.

Nami

Zare

Ghosh

and Blaabjerg

, A hybrid cascade converter topology with series-connected symmetrical and asymmetrical diode-clamped H-bridge cells, IEEE Transactions on Power Electronics26 (2011), 51–65.

Ozpineci

Tolbert

L.M.

and Chiasson

J.N.

, DC–AC cascaded H-bridge multilevel boost inverter with no inductors for electric/hybrid electric vehicle applications, IEEE Transactions on Industry Applications45 (2009), 963–970.