Search

The paper proposes a novel battery design for high-performance transport applications that is immersion-cooled and switched by a multi-level inverter. Advantages of the proposed AC battery design in terms of weight, modularity, scalability, performance, reliability and safety are presented. To demonstrate the applicability of the design, an electrically powered glider use case is addressed. The derived battery system is evaluated by means of theoretical analysis, simulation and prototyping. Simulations showed that the used multi-level inverter (MLI) power electronics modules could successfully run the motor without additional power electronics and charge batteries from a 110 V AC source. The prototype implementation with a motor-driven propeller demonstrated power levels of up to 3.3 kW, with a behavior in accordance with simulations. Guidelines to further advance the technology readiness level including control strategies and hardware design were derived to overcome limitations in the prototype realization that could not be addressed within the project budget. Finally, research topics to evaluate additional performance metrics such as efficiency and aging behavior are suggested. [ABSTRACT FROM AUTHOR]

This article presents a maximum torque per ampere strategy for a five-phase synchronous reluctance drive. The approach is developed considering general machine parameters and is formalized as a constrained optimization problem. The optimal solution is found analytically by using Lagrange’s multipliers method and is based on the computation of the eigenvalues and eigenvectors of the inductance derivatives matrix. The proposed approach is evaluated both numerically and experimentally. It is also compared with other current references control strategies, effectively showing a reduction of the machine rms currents for the same developed torque. The same approach can also be extended to machines with a different number of phases. [ABSTRACT FROM AUTHOR]

Inverter's performance and operating mode may be negatively affected by inverter input (dc‐link) current and voltage ripple. It is a common experience that even theoretically balanced loads with perfectly balanced supply voltages, such as multiphase ac motors supplied by pulse‐width modulation (PWM) inverters, in practice show a certain degree of current unbalance, in the range of a few percent, which introduces a low‐frequency instantaneous power oscillation. This reflects in current and voltage low‐frequency ripple on the dc‐link inverter side (i.e. at the double‐fundamental frequency). A possible method to analyse this matter is through the symmetric sequence components. In particular, based on the first negative current sequence component and by considering the equivalent dc‐link impedance calculated at the dominant double‐fundamental frequency, the amplitude of the corresponding dc‐link voltage ripple component is calculated in this work for a general multiphase load. Finally, the design of the dc‐link capacitor in multiphase inverters is proposed considering requirements referred to the double‐fundamental dc voltage ripple. The feasibility of the proposed developments has been verified for three‐, five‐ and seven‐phase inverters by both numerical simulations and comprehensive experimental tests, always showing a good matching. [ABSTRACT FROM AUTHOR]

Recent developments in power electronics are making the multiphase machines a competitive alternative to the conventional three-phase counterparts. Due to their fault-tolerant features, multiphase drives represent a robust technology in high-power/high-current, safety-critical applications. Besides, their introduction to transportation electrification is gaining importance. Among the multiphase solutions, the multi-three-phase machines are receiving a lot of attention by the industry since they use the well-consolidated three-phase technology, thus reducing the design time and also the cost. Therefore, this article proposes a modular vector control scheme for multi-three-phase permanent magnet synchronous motors. The proposed solution uses a modular modeling approach for the independent and decoupled torque control of each three-phase unit, allowing the implementation of torque-sharing strategies among the three-phase sets of the machine. The developed modular control has been validated on a nine-phase permanent magnet machine. [ABSTRACT FROM AUTHOR]

When multiphase drives are used for specific applications, the modular solutions are preferred as they use consolidated power electronics technologies. The literature reports two modeling approaches for multiphase machines having a modular configuration of the stator winding. The first approach is the vector space decomposition (VSD) that models the energy conversion as for an equivalent three-phase machine. The main alternative to the VSD is the multistator (MS) modeling that emphasizes machine modularity in terms of torque production. Both approaches have advantages and disadvantages for multiphase machines with a modular structure. Therefore, this article aims to combine the VSD and MS approaches, defining a new matrix transformation and, hence, developing a new modeling approach for multiphase machines with a modular structure. The proposed transformation allows a decoupled and independent torque control of the sets composing the machine, preserving the torque regulation's modularity. Together with a new vector control scheme, it has been applied to a modular permanent magnet synchronous machine (PMSM) with a nonstandard spatial shift between windings. Experimental results are presented for a nine-phase PMSM prototype with a triple-three-phase stator winding configuration. [ABSTRACT FROM AUTHOR]

The article presents a novel carrier-based pulsewidth modulation technique for multiphase neutral point clamped converters. The technique is aimed to actively control the neutral point (NP) potential while supplying the desired set of line-to-line voltages to the load. Standard techniques are either based on the sole common mode voltage injection or on the sole multistep switching mode; contrarily, the proposed algorithm combines these two approaches to take advantage of their main benefits. The technique performs well for each number of phases, for each modulation index, and for each type of load. It can control in closed-loop the NP voltage to any desirable value with a reduced number of switching transitions. The proposed approach has been experimentally validated and compared with other carrier-based algorithms. [ABSTRACT FROM AUTHOR]

This article presents a generalized approach toward the dc-link voltage switching ripple analysis in the two-level multiphase pulsewidth modulation (PWM) voltage source inverters with a balanced load. Since the voltage ripple is one of the crucial sizing criteria for a dc-link capacitor, a simple and practical equation for designing the dc-link capacitor, based on the maximum (peak-to-peak) value of the dc-link voltage ripple, has been proposed for the multiphase inverters. The amplitude of the dc-link voltage switching ripple is analytically derived as a function of operating conditions. The effect of the number of phases on the dc-link capacitor size is investigated as well. It is found that considering the same total output current, the dc-link capacitor size is reduced by increasing the number of phases up to seven. However, from the point of view of the dc-link capacitor size, there are no benefits of further increasing the number of phases. Reference is made to two commonly used modulation strategies—sinusoidal PWM and continuous symmetric centered PWM (i.e., space vector). The mathematical models are derived with an aim to provide the precise dc-link capacitor sizing and hence improve the power density of the whole system. The comparison of different phase numbers has been made. Proposed theoretical developments are verified by the simulation and experimental tests. [ABSTRACT FROM AUTHOR]

Testing the high-power machines in general is not an easy task. One of the standard tests is the full-load test. This test typically requires another machine, of the same or higher power rating, to be coupled to the tested one. For multiphase machines, which are commonly designed for high-power applications, this test can be conducted in a different way. In order to simplify full-load test, this paper proposes a new method that is applicable for multiple three-phase machines with even number of neutral points. The method is based on indirect rotor-field oriented control. It enables evaluation of the efficiency and the thermal design in the case of synchronous machines and the segregation of the constant and stator variable losses for induction machines, without the need for coupling another machine as a load. In the presented method, the full-load test conditions on the stator are obtained by circulating the rated active power flow in a closed loop from one winding set to another. The only power used during the test is to cover machine and converter losses. The proposed control scheme is unique and is based around y-component from the vector space decomposition subspace. It is validated through the simulation and experimental results. [ABSTRACT FROM AUTHOR]

In this article, a new pulsewidth modulation technique is proposed for the output voltage control and the dc-bus capacitors voltage balancing of a multipoint-clamped multilevel converter. The voltage equalization is achieved by allowing each converter's leg voltage to switch through multiple levels in a single modulation period (multistep operation). The approach is generalized with respect to the number of levels and phases and can be applied regardless of the converter operating conditions. Nevertheless, the multistep behavior generally leads to an increase of the average switching transitions rate. In the proposed method, this drawback is mitigated with an online adaptive selection of the number of switching levels. The algorithm is validated through an extensive hardware-in-the-loop testing and compared to other approaches. [ABSTRACT FROM AUTHOR]

A space‐vector pulse‐width modulation (SVPWM) algorithm for a three‐level asymmetrical six‐phase drive based on vector‐space decomposition approach is presented in this study. A modification in zero‐plane of the transformation matrix is proposed in order to meet the requirement that the realisation of sinusoidal output phase voltages can be obtained through the chosen output leg voltage space vectors. Furthermore, a method of choosing the switching sequences based on all possible one‐level transitions of the leg voltages, i.e. a permutation method, is introduced. The algorithm is then validated experimentally and obtained results show that the developed method successfully achieves the desired fundamental phase voltage, though low‐order harmonics are present due to uncompensated inverter dead time. Finally, the performance of the proposed SVPWM algorithm is compared with several carrier‐based PWM algorithms including in‐phase disposition with 'double min–max injection'. This is a little‐known type of injection, which is verified to obtain identical performance as the presented multilevel algorithm. [ABSTRACT FROM AUTHOR]

This paper introduces a method which can accurately transfer active and reactive power between sets of three-phase windings of a multiphase induction machine. The machine has to have a multiple of three phases (3, 6, 9, etc.), with each three phases representing one set. The paper demonstrates first that the sets do not share the power in the same way as they share a torque. Therefore, the paper proposes a method that is capable of achieving accurate power sharing. This is ensured by considering the whole power that is transferred through the air-gap to/from a set, instead of considering only a power that is transferred between that set and a rotor shaft. A complete control algorithm of the proposed method is given. Experiments and simulations verify theoretical considerations and the proposed control. [ABSTRACT FROM AUTHOR]

A space vector pulse-width modulation (PWM) technique for a two-level asymmetrical six-phase drive is proposed for the first time in this study. The PWM technique is developed based on the modified vector space decomposition approach. The details regarding the process of choosing the switching sequences and the calculation of dwell times are presented. The PWM technique is verified through simulation using PLECS software and is further validated experimentally. Obtained results show the validity of the proposed PWM technique. [ABSTRACT FROM AUTHOR]

In multiphase machine drives, accuracy of the estimated machine parameters is crucial for effective performance prediction and/or control. While a great amount of work has been done on parameter estimation for three-phase machines, corresponding discussions for six-phase machine remain scarce. It has been proven in the literature that the effect of mutual leakage inductance between different winding layers has a significant impact on the equivalent machine reactance, which challenges the standard separation method of stator and rotor leakage inductances from the measured locked-rotor impedance. In this paper, parameter identification of an asymmetrical six-phase induction machine using six-phase no-load and locked-rotor tests is discussed. A zero-sequence test using an improved equivalent circuit is proposed to improve the accuracy of the estimated parameters. The concept is verified using experimental results obtained from a low-power prototype asymmetrical six-phase machine. [ABSTRACT FROM AUTHOR]

This paper studies pulse width modulation (PWM) techniques suitable for a four-level five-phase open-end winding (OeW) drive. The drive comprises a five-phase induction machine, supplied using two two-level voltage source inverters with isolated and unequal dc-link voltages, in the ratio 2:1. A decoupled carrier-based (CB) PWM modulation strategy, based on unequal voltage reference sharing between the two converters, is introduced in this paper. The stability of dc-link voltages in OeW drives is investigated next, using a novel analysis technique. Several modulation methods are analyzed and the results show that application of the coupled PWM technique, with carriers having in-phase disposition, leads to overcharging of the capacitor in the dc-link of the inverter intended to operate with the lower dc-link voltage. On the other hand, the proposed decoupled CB PWM scheme naturally eliminates the dc-link capacitor overcharging problem. These findings are verified experimentally, using open-loop V/f control. Two different decoupled CB modulation methods are compared and the best performing modulation method is selected and incorporated further into an OeW drive with field-oriented control. The presented steady state and transient experimental results demonstrate that the decoupled CB PWM technique is suitable for high-performance variable speed drive applications. [ABSTRACT FROM AUTHOR]

This study discusses the space vector pulse-width modulation (SVPWM) scheme for an eleven-phase two-level voltage source inverter, aimed at producing a sinusoidal output voltage waveform. Generalised space vector theory is used to realise the SVPWM. As per the general inverter switching theory, there are 211 = 2048 switching states that yield 2046 active voltage space vectors and one zero voltage vector, which results with two switching states. Out of the total of 2046 active voltage vectors, the most suitable set comprising 110 active voltage vectors is identified and is utilised in the implementation of the SVPWM. The sinusoidal voltage is obtained by controlling the duty cycles of the applied voltage space vectors in such a way that the non-zero reference voltage in the first (d-q) plane is achieved, while simultaneously zeroing the average voltage in the other four (x-y) planes in accordance with the zero references. The theoretical results are verified by experimentation using a passive resistive-inductive load. Finally, experimentally obtained total harmonic distortion values of the phase voltage and current for the eleven-phase SVPWM are compared with the corresponding values obtained using SVPWM for other odd phase numbers. [ABSTRACT FROM AUTHOR]

Three-phase multilevel (ML) inverters are used in many medium- and high-power applications such as motor drives and grid-connected systems. Despite the fact that numerous pulsewidth modulation (PWM) techniques for ML inverters have been developed, the impact of these modulation schemes on the peak-to-peak output current ripple amplitude has not been addressed yet. In this paper, analysis and comparison of current ripple for two-level (2L) and three-level (3L) voltage source inverters are given. Reference is made to optimal and popular modulation, so-called centered PWM, easily obtained by both carrier-based modulation (phase disposition, with proper common-mode voltage injection) and space vector modulation (nearest three vectors). It is shown that the peak-to-peak current ripple amplitude in 3L inverters can be determined on the basis of the ripple in 2L inverters, obtaining the same results as by directly analyzing the output voltage waveforms of the 3L inverters. This procedure can be readily extended to higher level numbers. The proposed analytical developments are verified by both numerical simulations and experimental tests. [ABSTRACT FROM PUBLISHER]

The analysis and assessment of the pulsewidth modulation (PWM) techniques is commonly based on the comparison of the total harmonic distortion (THD) results. THD is usually calculated by application of the Fourier transformation and by taking a limited number of harmonics into the consideration. In this paper, derivation of analytical formulas for the phase voltage THD is presented. The considered system is a symmetrical multiphase star-connected load, supplied from a multilevel pulsewidth-modulated voltage-source inverter (VSI, three-phase case is also covered). The solution is based on the Parseval's theorem, which links frequency spectrum and time domain through the average power (i.e., root-mean-square (rms) squared value) of the signal. The assumption throughout the derivations is that the ratio of the switching to fundamental frequency is high. Derivations are based on the integration of the power of the PWM signal in a single switching period over the fundamental period of the signal. Only ideal sinusoidal reference voltages are analyzed, and no injection of any type is considered. Formulas for phase voltage THD for any number of phases are derived for two- and three-level cases, for the most commonly used carrier-based methods. Comparison of the analytically obtained curves with simulation and experimental results shows a high level of agreement and validates the analysis and derivations. [ABSTRACT FROM AUTHOR]

Nowadays, three-phase multilevel inverters are widely employed in medium and high-power applications, increasing the power ratings, improving the output voltage quality and reducing the conducted electromagnetic interferences. Despite of numerous pulse-width modulation (PWM) techniques have been developed for multilevel inverters, a detailed analysis of the output current ripple amplitude has not been reported yet. In this study, the peak-to-peak current ripple distribution over a fundamental period is analysed in details specifically for three-level three-phase voltage source inverters for both motor-load and grid-connected applications. In particular, the peak-to-peak amplitude of the current ripple is determined analytically as a function of the modulation index. The centred PWM strategy is considered in all the developments, implemented either by carrier-based or space vector (SV) PWM methods. With this modulation, the dc bus utilisation is maximised in a simple and effective way, and a nearly-optimal behaviour is obtained to minimise the current ripple rms. The results obtained in different cases and sub-cases identified in the proposed analytical approach are verified by experimental tests with reference to threephase three-level neutral-point clamped configuration. [ABSTRACT FROM AUTHOR]

Multilevel inverter supplied multiphase variable-speed drive systems have in recent times started attracting more attention, due to various advantages that they offer when compared to the standard three-phase two-level drives. For proper functioning of such systems good pulsewidth modulation (PWM) strategy is of crucial importance. Control complexity of multiphase multilevel inverters increases rapidly with an increase in the number of phases and the number of levels. This paper deals with a three-level neutral point clamped (NPC) inverter supplied five-phase induction motor drive and analyses five PWM strategies: three are carrier-based (CBPWM) and two are space vector based (SVPWM). The aim is to provide a detailed comparison and thus conclude on pros and cons of each solution, providing a guideline for the selection of the most appropriate PWM technique. Experimental results are provided for all analysed PWM methods. The comparison of the PWM techniques is given in terms of the voltage and current waveforms and spectra, as well as the total harmonic distortion (THD) in a whole linear modulation index range, which is used as the global figure of merit. Properties of the common mode voltage (CMV) are also investigated. Complexity of the algorithms, in terms of the computational time requirements and memory consumption, is addressed as well. It is shown that the performance of the PWM techniques is very similar and that one CBPWM and one SVPWM technique are characterised with identical performance. However, using the algorithm complexity as the main criterion, space vector techniques are more involved. [ABSTRACT FROM PUBLISHER]

This paper presents a novel modulation strategy for a seven-phase three-level inverter, aimed at variable-speed drive applications. Developed space vector pulsewidth modulation (SVPWM) strategy is based on vector space decomposition (VSD) approach and an analysis of multidimensional space vector projections in mutually orthogonal planes. The algorithm represents an extension of a recently developed five-phase three-level SVPWM algorithm and is considerably more complex, due to the existence of three mutually orthogonal planes in a seven-phase system. The difficulties of generalization of space vector strategies, which stem from the nature of the multiphase systems, are highlighted. Increasing the number of phases from five to seven increases the number of switching states from 3 ^5 = 243 to 3^7 = 2187 and the number of orthogonal planes from two to three. This considerably affects the complexity of the offline calculations. However, the final implementation of the algorithm is simple. Developed SVPWM algorithm is compared to an equivalent carrier-based strategy and it is shown that they yield identical performance. The complexity of the algorithms for real-time implementation is compared. Simulation and experimental results, obtained using neutral-point clamped inverter, verify the theoretical developments. [ABSTRACT FROM AUTHOR]