Hybrid excitation synchronous machine adaptive speed region control and experimental verification

Abstract

With the merits of permanent magnet (PM) machine combining the possibility of controllable flux by field windings, hybrid excitation synchronous machine (HESM) is a novel candidate for wide speed range regulation application. The HESM with claw pole topology incorporates the advantages of the hybrid excited synchronous machine and claw pole machine. In this paper, the topology, operating principle and basic mathematical model of HESM are analyzed and established, respectively. Then, the control model of the HESM is deduced on the basis of the space vector control method, and the operating performance of the HESM across its entire operating range is also investigated. A new adaptive control strategy for the HESM drive system is proposed and designed on the basis of analyzing the influences of the flux-weakening coefficient, respectively, which can operate in both flux-enhancing and flux-weakening conditions. The correctness and effectiveness of the proposed adaptive control strategy and drive system design are verified by simulation and experimental results, which demonstrate that the adaptive control strategy maximizes the range of speed regulation while exhibits the high efficiency.

Keywords

Hybrid excitation speed region control wide speed range adjusting vector control

1. Introduction

Hybrid excitation synchronous machine (HESM), as a novel embranchment of permanent magnet synchronous machine (PMSM), which has been drawn extensive attentions in recent years since it incorporates the merits of PMSM with the possibility of controllable flux by auxiliary excitation windings. With the feature of wide range speed regulation operation by adjusting the excitation current and d axis current [1–3], HESM is one promising candidate in many industrial application fields, such as electric vehicles (EVs), hybrid EVs and servo systems [4–7]. However, for those traditional HESMs, there were still several disadvantages, such as, longer field coil per turn, large resistance, and high field excitation loss [8]. Moreover, in order to obtain satisfied field excitation proportion, the machine axial lengths were usually enlarged due to the limited field winding space, which resulted in bulk volume [9]. To overcome these drawbacks, one novel hybrid excitation synchronous machine with claw-pole topology is developed in this paper. The proposed machine owns the merits of HESM and claw-pole machine, namely, which combines the fixed excitation of PM as well as the variable flux produced by a toroidal field winding located on the inner stator.

Due to the similarities of structure and performance between HESM and PMSM, many traditional PMSM control methods were developed to realize the wide range operation for HESM. A hybrid excitation brushless DC motor fuzzy control scheme was presented in [10], which regulate armature and excitation current by using the fuzzy controller. A common coordinate system for HESM based dynamic vector control model was proposed in [11]. A copper loss minimization vector control method for non-salient pole HESM was proposed in [12], the idea of which was based on i _d = 0. A simplified control method for HESM was proposed in [13], the coil current and direction of which are controlled by stator winding current and the back EMF. In addition, a maximum torque/copper loss control method for HESM was presented in [14], and a simplified subsection control strategy and no excitation current control mode were carried out and compared in [15], respectively. A fuzzy control and particle swarm optimization algorithm was proposed to improve the efficiency and stability of the HESM in [16], and the similar ideas of which could be found in [17], respectively. All these research works have positive reference to reveal the electromagnetic property of HESM. However, HESM has already been proved to be a multi-variable nonlinear high order system with strong coupling [18–20], the electromagnetic torque of which was related to d axis current, q-axis current and excitation current together. The influence of power supply voltage is not considered in the process of flux-weakening control, which means that the flux-weakening of d-axis current is not utilized. When the excitation current provided inadequate flux-weakening effect, the machine speed regulation range could not be obtained fully expanded without d-axis current. The disadvantages of above mentioned control methods were that it cannot be maintain comprehensive performance optimal through the flux-weakening control operation due to the static flux-weakening coefficient. Once the machine works in different operation state, it cannot be taken into account both the efficiency and the speed range at the same time. Hence, a reasonable allocation of excitation and armature current should be integrated consideration to achieve optimal static and dynamic characteristics with the existent of an additional controllable excitation current.

This paper is organized as follows. The topology and basic mathematical model of the proposed HESM are presented in Section 2. Then, the proposed HESM adaptive stage control strategy is described and deduced in Section 3, which includes two parts, namely, load torque control and flux weakening control in the low speed and the high speed regions, respectively. The adaptive control system simulation model for the HESM is established, simulated and analyzed in Section 4. Experimental test platform and several performances under different operations for the HESM are developed and tested in Section 5, respectively.

2. Topology and basic mathematic model of HESM

2.1. Machine topology

Figure 1 shows the structure of the proposed HESM, which is composed of dual-stator and a cup-shape rotor. The armature winding and excitation winding are emplaced in outer stator and inner stator, respectively. The surface of claw pole on rotor is divided into two sections: one is a PM pole, and the other is a solid iron pole, both of which are interlocked with each other, as shown in Fig. 1(b). PMs create a nearly constant flux, while the field winding currents generate a variable flux; both form the resultant flux in the air gap. The magnetic field generated by the excitation current is mostly entered into the air gap through the core because the magnetic resistance of PM is much larger than that of the core. Hence, when the motor works in flux-weakening or flux-enhancing, there is no risk of permanent demagnetization for the rotor PM. Meanwhile, a satisfied magnetic flux adjusting capacity can be obtained by regulating excitation current and the d-axis current due to the small reluctance of the whole magnetic circuit.

It can be seen from Fig. 1(b) that the magnetic field of the HESM produced by the excitation current has a slight degree of coupling on the magnetic field produced by the armature current and PMs, which means the air-gap magnetic field of HESM can be regards as approximated linear superimposition. The air-gap magnetic field can be adjusted effectively by changing the size and direction of excitation current and the d-axis current.

2.2. Mathematic model

Hence, neglecting the influence of the temperature, magnetic saturation, and hysteresis loss [21], several basic equations of the HESM, including voltage, flux linkage and electromagnetic torque can be expressed as

Voltage equation: $\begin{eqnarray}\displaystyle \left[\begin{array}{@{}c@{}}u_{d}\\ u_{q}\\ u_{f}\end{array}\right]=\left[\begin{array}{@{}ccc@{}}R_{s}+sL_{d} & -{\omega}_{e}L_{q} & sM_{sf}\\ {\omega}_{e}L_{d} & R_{s}+sL_{q} & {\omega}_{e}M_{sf}\\ sM_{sf} & 0 & R_{f}+sL_{f}\end{array}\right]∼\left[\begin{array}{@{}c@{}}i_{d}\\ i_{q}\\ i_{f}\end{array}\right]+\left[\begin{array}{@{}c@{}}0\\ {\omega}_{e}{\psi}_{\text{pm}}\\ 0\end{array}\right] & & \displaystyle\end{eqnarray}$ (1) where u _d, u _q are the d- and q-axis armature voltage components, respectively; u _f is the excitation voltage, L _d, L _q are the d- and q-axis inductance, respectively; R _s, R _f are the armature resistance and excitation winding resistance, respectively; M _sf is the mutual inductance between armature windings and excitation windings; ω_e is the electrical angular velocity; i _d, i _q are the d- and q-axis current components, respectively; i _f is the excitation current; 𝜓_pm is the permanent magnet flux linkage.

Flux linkage equation: $\begin{eqnarray}\displaystyle \left[\begin{array}{@{}c@{}}{\psi}_{d}\\ {\psi}_{q}\end{array}\right]=\left[\begin{array}{@{}ccc@{}}L_{d} & 0 & M_{sf}\\ 0 & L_{q} & 0\end{array}\right]\left[\begin{array}{@{}c@{}}i_{d}\\ i_{q}\\ i_{f}\end{array}\right]+\left[\begin{array}{@{}c@{}}{\psi}_{\text{pm}}\\[3.60004pt] 0\end{array}\right] & & \displaystyle\end{eqnarray}$ (2) where 𝜓_d, 𝜓_q are the d- and q-axis flux linkages.

Electromagnetic torque equation: $\begin{eqnarray}\displaystyle T_{e}=\frac{3}{2}pi_{q}[{\psi}_{\text{pm}}+i_{d}(L_{d}-L_{q})+M_{sf}i_{f}] & & \displaystyle\end{eqnarray}$ (3) where T _e is electromagnetic torque; p is the number of pole pairs of the machine.

3. Two speed region control strategy

For the HESM control system, two typical operations should be fully considering [22], including a large starting torque by the flux-enhancing control below the rated speed and a wide adjustable speed range for high-speed cruise by flux-weakening control above the rated speed, respectively. Hence, the proposed control strategy for whole speed range can be divided into low speed and high speed regions according to the different operations, respectively. For the low speed region I (n _r ≤ n _b, n _r is rotor speed, n _b is base speed), maximum torque per ampere (MTPA) control strategy is used to adjust speed. While for the high speed region II (n _r > n _b), minimum copper loss (MCL) control strategy is adopted to achieve flux weakening, which is based on the idea of adjusting excitation current and d-axial current.

3.1. Maximum torque per ampere (MTPA) control method in low speed region

Available maximum output torque in low speed region is important to the drive system of HESM, which means that a quick response can be realized timely [23]. When the operation speed of the proposed HESM is less than or equal to the base speed n _b, excitation current for speed adjusting are adopted according to its load torque. With the purpose of improving the operation reliability of HESM, a maximum torque current ratio control method based on i _d = 0 is adopted, the torque equation can be obtained by $\begin{eqnarray}\displaystyle T_{e\text{ref}}=\frac{3}{2}p[{\psi}_{\text{pm}}+M_{sf}i_{f\text{ref}}]i_{q\text{ref}} & & \displaystyle\end{eqnarray}$ (4) where, T _eref is the reference value of electromagnetic torque; i _fref and i _qref are the reference q-axis current component and excitation current component, respectively.

When T _eref is small and satisfied $\begin{eqnarray}\displaystyle T_{e\text{ref}}\leq T_{N}=\frac{3}{2}p{\psi}_{\text{pm}}I_{qN} & & \displaystyle\end{eqnarray}$ (5) where T _N is load torque, I _qN is the base value of q-axis current.

In this case, no excitation current is needed for speed regulating, and T _eref can be changed by only adjust the value of i _qref, thus, the reference current of i _qref can be expressed as $\begin{eqnarray}\displaystyle i_{q\text{ref}}=\frac{2T_{e\text{ref}}}{3p{\psi}_{\text{pm}}}. & & \displaystyle\end{eqnarray}$ (6) When T _eref is larger than T _N, namely $\begin{eqnarray}\displaystyle T_{e\text{ref}}>T_{N}=\frac{3}{2}p{\psi}_{\text{pm}}I_{qN}. & & \displaystyle\end{eqnarray}$ (7) The flux-enhancing control is adopted and the reference current i _fref are given by $\begin{eqnarray}\displaystyle i_{f\text{ref}}=\frac{2T_{e\text{ref}}-3p{\psi}_{\text{pm}}I_{qN}}{3pM_{sf}I_{qN}}. & & \displaystyle\end{eqnarray}$ (8) Hence, T _eref can be further rewritten as $\begin{eqnarray}\displaystyle T_{e\text{ref}}=\frac{3}{2}p{\psi}_{\text{pm}}I_{qN}+\frac{3}{2}pM_{sf}i_{f\text{ref}}I_{qN}. & & \displaystyle\end{eqnarray}$ (9) In summary, when T _eref ≤ T _N, the reference currents can be expressed as $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}l@{}}i_{f\text{ref}}=0\\ i_{d\text{ref}}=0\\ i_{q\text{ref}}=\displaystyle \frac{2T_{e\text{ref}}}{3p{\psi}_{\text{pm}}}\end{array}\right. & & \displaystyle\end{eqnarray}$ (10) when T _eref > T _N, the reference currents are given by $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}l@{}}i_{f\text{ref}}=\displaystyle \frac{2T_{e\text{ref}}-3p{\psi}_{\text{pm}}I_{qN}}{3pM_{sf}I_{qN}}\\ i_{d\text{ref}}=0\\ i_{q\text{ref}}=I_{qN}.\end{array}\right. & & \displaystyle\end{eqnarray}$ (11)

3.2. Minimum Copper loss (MCL) control method in high speed region

When the working speed n _r exceeds the base speed n _b, the HESM enters into high speed operation stage and its back-EMF approaches to the DC-bus voltage U _dc. If the speed is to be further increased, the flux weakening control should be carried out. The regulation of the armature current and excitation current are restricted by the voltage limit ring [24], which are similar to PMSM flux weakening control method in constant power region.

When the operation is stable, the voltage vector magnitude meets $\begin{eqnarray}\displaystyle u_{s}^{2}=u_{d}^{2}+u_{q}^{2}\leq U_{\lim }^{2} & & \displaystyle\end{eqnarray}$ (12) where U _lim is maximum phase voltage magnitude and its value depends on U _dc.

It can be seen from the Eq. (12), the value of u _s is depended on u _q, while u _q is determined by the q-axis component of the back-EMF E _q. Hence, a flux weakening control method based on maintaining the back-EMF invariable is presented according to the above analysis, namely $\begin{eqnarray}\displaystyle E_{q}=E_{\text{base}} & & \displaystyle\end{eqnarray}$ (13) where E _base is the back-EMF under no-load operation when n _r reaches n _b.

E _base is expressed as $\begin{eqnarray}\displaystyle E_{\text{base}}=pn_{\text{Bdec}}{\psi}_{\text{pm}}{\pi}/30. & & \displaystyle\end{eqnarray}$ (14) Hence, E _q can be rewritten as $\begin{eqnarray}\displaystyle E_{q}=\frac{p{\pi}n_{r}}{30}({\psi}_{\text{pm}}+i_{d}L_{d}+i_{f}M_{sf}) .& & \displaystyle\end{eqnarray}$ (15) When i _d = i _f =0, it is acknowledged that n _r appears to be linearly proportional of E _q.

Therefore, the maximum speed n _max under no-load and no-excitation current also appears linear relationship with U _dc when traditional vector control method is used. While, it is should be noted that flux weakening base speed n _b is restricted by n _max, and n _max can be obtained by no-load vector control experiment of i _d = i _f = 0. Then, the relationships of n _b, n _max and U _dc can be obtained by linear fitting according to the experimental data of bus voltage and no-load speed and can be expressed as $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}l@{}}n_{\max }=k_{v}U_{\text{dc}}+N_{0}\\ n_{b}=k_{b}n_{\max }\end{array}\right. & & \displaystyle\end{eqnarray}$ (16) where for the experimental prototype, the maximum speed to voltage ratio coefficient takes k _v = 5.69, offset value N ₀ = −13.

To ensure the utilization rate of DC bus voltage and the motor efficiency in flux weakening operating state, the flux weakening base speed coefficient k _b is set a range of 0.7 ∼ 0.9, here, k _b = 0.75. So, when the DC bus voltage value sets U _dc = 300 V, then n _b =1270 rpm.

Therefore, in order to maintain the back-EMF in a higher value and keep the flux weakening state unchanged, several assumptions are expressed as $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}c@{}}i_{d}=i_{d\text{ref}}\\ i_{f}=i_{f\text{ref}}\end{array}\right. & & \displaystyle\end{eqnarray}$ (17) where i _dref, i _fref are the reference values of d-axis currents and excitation currents, respectively.

Thus, combing the expressions from Eqs. (13) to (15), it can be obtained $\begin{eqnarray}\displaystyle (L_{d}i_{d\text{ref}}+M_{sf}i_{f\text{ref}})=\displaystyle \frac{{\psi}_{\text{pm}}(n_{b}-n_{r})}{n_{r}}. & & \displaystyle\end{eqnarray}$ (18) It can be seen from Eq. (18) that voltage limit equation is differ than that of traditional PMSM, and the control accuracy of it depends on the ratio of n _b∕n _r.

The copper loss equation in high speed region can be obtained by using copper loss minimization principle and it can be expressed as $\begin{eqnarray}\displaystyle P_{\text{cu}\text{-}\text{ref}}=\frac{3}{2}i_{d\text{ref}}^{2}R_{s}+i_{f\text{ref}}^{2}R_{f} & & \displaystyle\end{eqnarray}$ (19) where P _cu−ref is the reference value of machine copper loss.

In order to simplify the computation, the value of i _qref is directly determined by the reference electromagnetic torque T _eref. Thus, it doesn’t include the copper loss produced by q-axis component current i _qref in the Eq. (19), and i _qref is given by $\begin{eqnarray}\displaystyle i_{q\text{ref}}=k_{i}T_{e\text{ref}} & & \displaystyle\end{eqnarray}$ (20) where, k _i is the q-axis current to torque ratio coefficient.

Combining Eqs (19) and (20), the Lagrange multiplier is adopted to get the reference current values based on copper loss minimization control method, and it is defined as $\begin{eqnarray}\displaystyle L(i_{d\text{ref}},i_{f\text{ref}},{\lambda})=P_{\text{cu}\text{-}\text{ref}}+{\lambda}\left[(L_{d}i_{d\text{ref}}+M_{sf}i_{f\text{ref}})-\displaystyle \frac{{\psi}_{\text{pm}}(n_{b}-n_{r})}{n_{r}}\right]. & & \displaystyle\end{eqnarray}$ (21) To solve partial derivative of i _dref, i _fref and 𝜆 for Eq. (21), respectively $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}l@{}}\displaystyle \frac{\partial L}{\partial i_{d\text{ref}}}=3i_{d\text{ref}}R_{s}+{\lambda}L_{d}\\[8.39996pt] \displaystyle \frac{\partial L}{\partial i_{f\text{ref}}}=2i_{f\text{ref}}R_{f}+{\lambda}M_{sf}\\[8.39996pt] \displaystyle \frac{\partial L}{\partial {\lambda}}=(L_{d}i_{d\text{ref}}+M_{sf}i_{f\text{ref}})-\displaystyle \frac{{\psi}_{\text{pm}}(n_{\text{b}}-n_{\text{r}})}{n_{\text{r}}}.\end{array}\right. & & \displaystyle\end{eqnarray}$ (22) Assuming $\frac{\partial L}{\partial i_{d\text{ref}}}=0$ , $\frac{\partial L}{\partial i_{f\text{ref}}}=0$ , $\frac{\partial L}{\partial {\lambda}}=0$ , then the values of i _dref and i _fref is the following as $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}c@{}}i_{d\text{ref}}=-\displaystyle \frac{L_{d}}{3R_{s}}{\lambda}\\[6.0pt] i_{f\text{ref}}=-\displaystyle \frac{M_{sf}}{2R_{f}}{\lambda}\end{array}\right. & & \displaystyle\end{eqnarray}$ (23) where ${\lambda}=\frac{6R_{s}R_{f}{\psi}_{\text{pm}}(n_{r}-n_{b})}{n_{r}(2R_{f}L_{d}^{2}+3R_{s}M_{sf}^{2})}$ .

In summary, speed region current allocation method can be used to obtain flux weakening control for the HESM. Firstly, the realization of flux-weakening control is based on copper loss minimization idea, which regulates i _qref, i _dref and i _fref together. The corresponding reference values for each current component are given as $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}c@{}}i_{q\text{ref}}=k_{i}T_{e\text{ref}}\\ i_{d\text{ref}}=-\frac{L_{d}}{3R_{s}}{\lambda}\\ i_{f\text{ref}}=-\frac{M_{sf}}{2R_{f}}{\lambda}.\end{array}\right. & & \displaystyle\end{eqnarray}$ (24) Secondly, the flux weakening control is implemented by only using the excitation current, namely i _dref = 0, the corresponding reference values for each current component are given as $\begin{eqnarray}\displaystyle \left\{\begin{array}{@{}l@{}}i_{q\text{ref}}=k_{i}T_{e\text{ref}}\\ i_{d\text{ref}}=0\\ i_{f\text{ref}}=\displaystyle \frac{{\psi}_{\text{pm}}}{M_{sf}}\left(\displaystyle \frac{n_{b}}{n_{r}}-1\right).\end{array}\right. & & \displaystyle\end{eqnarray}$ (25)

4. Speed region control modeling and simulations

In order to investigate the effectiveness of the proposed speed region control strategy, HESM simulation model was established by using MATLAB/SIMULINK. Figure 2 shows the control system model of HESM, which includes HESM module, reference current calculator, transforms of inverse Clarke, Park and Ipark, Speed PID, Current Distributor, I _d PID, I _q PID, SVPWM, I _f PWM, Armature Driver, I _f Driver. Compared with the traditional PMSM control system, HESM control system has additional three functional modules of I _f PWM, I _f Driver and Current Distributor.

Current Distributor divides the entire operation region of HESM into two regions: low speed and high speed stages, and different control methods are used to investigate the machine performance, as shown in Fig. 3. The key function of Current Distributor is coordinating the allocation of the armature current and excitation current to realize a reasonable distribution of the current reference values for the machine operations between two speed ranges, hence to ensure the stable, reliable, efficient operation for the HESM.

The major parameters of the studied HESM are tabulated in Table 1.

Several simulations under different speed regions were implemented and compared, as shown in Fig. 4, which contains three control strategies of i _dref = i _fref = 0, i _dref = 0, and combination of i _d and i _f for field weakening control, respectively.

It can be seen from Fig. 4(a) that the maximum speed values of studied HESM are 1650 rpm, 2350 rpm and 4610 rpm with the three control strategies above mentioned, respectively. It should be pointed that the operation speed is significantly extended with the combination control strategy of d-axis current and the excitation current. Fig. 4(b) shows the back-EMF waveforms of studied HESM under three different control strategies. The back-EMF values with no- field weakening current regulation and the combination of d-axis current and the excitation current grows fastest and slowest among the three ones. All of the back-EMF values are finally stabilized at 165 V. The electromagnetic torque value under the control strategy of i _d and i _f co-action is larger than that of other two methods, especially when the machine enters into stable operation region, which is shown in Fig. 4(c). Similarly, when HESM works in stable region, the current value of i _q under the control strategy of i _d and i _f co-action is obvious larger than that of other two methods, as shown in Fig. 4(d).

5. Experimental verification

With the purpose of verifying the correctness of simulations results with the proposed speed region control strategy, HESM testing platform was established, the controller of which was based on TMS320F2812+AT89C55WD. It contains five parts, namely HESM prototype, torque testing, magnetic brake, driver board and control board, as shown in Fig. 5. TS-7700 Torque Station with MT-6425 torque detectors was used to test the motor torque characteristics. Rated DC bus voltage for HESM prototype was 300 V.

Several experiments were implemented with speed region control strategy. Figure 6 shows the HESM starting current waveform with 1 Nm load. The given speed of HESM is 2800 rpm, which is larger than that of flux weakening base speed of 1270 rpm. With the purpose of improving the starting torque of the motor, a positive rated excitation current was applied ahead half second to the motor to enhance the magnetic field before starting the armature current due to the large inductance of the excitation winding. With the increase of the excitation current, the motor speed decreases gradually. When the speed reaches the basic speed n _b, the flux weakening control for speed adjusting begins, i _f and i _d continually increase toward to negative part with the speed increasing.

Figure 7 shows the steady state current waveform of the motor under flux weakening operation. The motor speed and load torque are 2800 rpm and 1 N m, respectively, and the excitation current basically keeps at minus 0.8 A. The amplitude of the phase current is 4 A and it exhibits higher sinusoidal degree, which means the harmonic is smaller.

Figure 8 shows the experimental results of the measured power-speed curves with three different control strategies of i _d = i _f = 0, i _f control (i _d = 0), i _d and i _f co-action control. For the constant power region of HESM, the machine speed stays 700 ∼1300 rpm without excitation current of i _d = i _f = 0. When the speed exceeds 1300 rpm, the output power decreases sharply and there is almost no output torque and power when the speed reaches 1600 rpm. By adopting i _d = 0 control strategy, the machine constant power region was extend to 1600 rpm and its maximum operation speed reaches 2200 rpm. Compared with the i _d = i _f = 0 as well as i _f control, the loading capacity increases in the flux-enhancing condition and the range of speed regulation is broadened under the flux-weakening condition when the i _d and i _f co-action control strategy is used. With the excitation current regulation effect, the constant power region was extended to 450 ∼ 1800 rpm. And although the constant power cannot be maintained, the output power decreased slowly. When the speed is up to 4000 rpm, it still stays 200 W output power. The experimental results show that the load capacity of the studied HESM in the low speed and high speed operating region is effectively improved through the adjustment of the excitation current.

6. Conclusion

HESM is a multi-variable system with the merits of high order, nonlinear and strong coupling. In this paper, a new type of HESM control system was designed. On the basis of the space vector control, the HESM mathematical model was deduced and its simulation control model was established. According to the magnetic field adjusting property of HESM, one novel two-speed-region control method for HESM was presented. The operating performance of the HESM was investigated in the entire operating region. Both simulation and experimental results verify the validity of the proposed control strategy, and the following conclusions could be drawn:

(1) The proposed drive control system was based on rotor magnetic field orientation adaptive region control. Different operation speed regions were applied to different control strategies according to the electrical characteristics of HESM. Experiments show that the HESM control system switch steadily and smoothly in the whole speed ranges.

(2) The magnetizing current and the d-axis current were adopted in the low speed area and high speed area, respectively. Compared with the conventional HESM control methods with only utilization of the current excitation, it exhibited remarkable low speed large torque and wide speed regulation characteristics.

(3) During the process of flux-weakening speed regulation in high speed region, it was effective to realize the decoupling control of the two parties by adjusting the d-axis current and excitation current with maintaining the constant back-EMF, as well as simplified control algorithm of the HESM.

Footnotes

Acknowledgements

This work was supported by Scientific and Technological Program of Henan Science and Technology Agency (16A470004, 17A470008), the Technological innovation talents projects of Henan universities (No. 17HASTIT020), the Fundamental Scientific and Technological Research Funds of Henan Science and Technology Agency (182102210252), and the Henan province youth backbone teacher project (No. 2016GGJS153).

References

Spooner

Khatab

S.W.

and Nicolaou

N.G.

, Hybrid excitation of AC and DC machine, in: 1989 The Fourth International Conference on Electrical Machines and Drives, pp. 48–52.

Chan

C.C.

Chau

K.T.

and Jiang

J.Z.

, Novel permanent magnet motor drives for electric vehicles, IEEE Transactions on Industrial Electronics 43 (1996), 331–339.

Yang

C.F.

Lin

H.Y.

Guo

and Zhu

Z.Q.

, Design and analysis of a novel hybrid excitation synchronous machine with asymmetrically stagger permanent magnet, IEEE Transaction on Magnetics 44 (2008), 4353–4356.

Zhang

Yan

and Yang

, Principle of operation and feature investigation of a new topology of hybrid excitation synchronous machine, IEEE Transaction on Magnetics 44 (2008), 2174–2180.

Lawler

J.S.

Bailey

and McKeever

, Minimum current magnitude control of surface PM synchronous machines during constant power operation, IEEE Power Electronics Letters 3 (2005), 53–56.

Shinnaka

, New dynamic mathematical model and new dynamic vector simulators of hybrid-field synchronous motors, in: 2005 IEEE International Conference on Electric Machines and Drives, pp. 882–889.

Shinnaka

, New optimal current control methods for energy-efficient and wide speed-range operation of hybrid-field synchronous motor, IEEE Transactions on Industrial Electronics 54 (2007), 2443–2450.

Amara

Lucidarme

Gabsi

Lécrivain

Hamid

Ahmed

and Akémakou

A.D.

, A new topology of hybrid synchronous machine, IEEE Transactions on Industry Applications 35 (2001), 1273–1278.

Matsuuchi

Fukami

Naoe

Ryoichi

Shinzo

and Toshio

, Performance prediction of a hybrid-excitation synchronous machine with axially arranged excitation poles and permanent-magnet poles, Electrical Engineering in Japan 150 (2005), 43–49.

10.

Chan

C.C.

Zhang

and Chau

K.T.

, Optimal efficiency control of PM hybrid motor drives for electrical vehicles, in: 1997 IEEE Power Electronics Specialists Conference, pp. 363–368.

11.

Fuchs

E.F.

and Myat

M.H.

, Speed and torque range increases of electric drives through compensation of flux weakening, in: 2010 IEEE International Symposium on Power Electronics, Electrical Drives, Automation and Motion, pp. 1569–1574.

12.

Chen

J.J.

and Chin

K.P.

, Automatic flux-weakening control of permanent magnet synchronous motors using a reduced-order controller, IEEE Transactions on Power Electronics 15 (2000), 881–890.

13.

Chen

J.J.

and Chin

K.P.

, Minimum copper loss flux weakening control of surface mounted permanent magnet synchronous motors, IEEE Transactions on Industrial Electronics 18 (2003), 929–936.

14.

Kaehler

and Henneberger

, Transient 3-D FEM computation of eddy-current losses in the rotor of a claw-pole alternator, IEEE Transaction on Magnetics 40 (2004), 1362–1365.

15.

Amara

Vido

Gabsi

Hoang

and Hamid

, Hybrid excitation synchronous machines: Energy-efficient solution for vehicles propulsion, IEEE Transactions on Vehicular Technology 58 (2009), 2137–2149.

16.

Huang

M.M.

Lin

H.Y.

Huang

Y.K.

Jin

and Guo

Y.J.

, Fuzzy control flux weakening of hybrid excitation synchronous motor based on particle swarm optimization algorithm, IEEE Transaction on Magnetics 48 (2012), 2989–2992.

17.

Wang

and Deng

, Hybrid excitation topologies and control strategies of stator permanent magnet machines for DC power system, IEEE Transactions on Industrial Electronics 59 (2012), 4601–1615.

18.

Vujicic

V.P.

, Mnimization of torque ripple and copper losses in switched reluctance drive, IEEE Transactions on Power Electronics 27 (2011), 388–399.

19.

Mbayed

Salloum

Vido

Monmasson

and Gabsi

, Optimal control of the hybrid excitation synchronous machine for electric propulsion in electric vehicle, in: 2013 Fifteenth European Conference on Power Electronics and Applications, pp. 1–10.

20.

Mohammadi

A.S.

Trovão

J.P.

and Maxime

R.D.

, Hybridisation ratio for hybrid excitation synchronous motors in electric vehicles with enhanced performance, IET Electrical Systems in Transportation 8 (2018), 12–19.

21.

Zhang

Z.R.

Liu

Tian

and Wang

W.J.

, Investigation and implementation of a new hybrid excitation synchronous machine drive system, IET Electric Power Applications 11 (2017), 487–494.

22.

Gabriele

Capponi

F.G.

Donato

G.D.

and Caricchi

, Closed-loop flux-weakening control of hybrid-excitation synchronous machine drives, IEEE Transactions on Industry Applications 53 (2017), 1116–1126.

23.

Michal

, A gain-scheduled multivariable LQR controller for hybrid excitation synchronous machine, in: 2015 The Twentieth International Conference on Methods and Models in Automation and Robotics, pp. 24–27.

24.

Zhao

J.L.

Lin

M.Y.

and Xu

, Minimum-copper-loss control of hybrid excited axial field flux-switching machine, IET Electric Power Applications 10 (2016), 82–90.