High speed permanent magnet synchronous motors (PMSMs) are used in electric vehicles because of their intense power density. The high speed implies a significant electromotive force and requires flux weakening. The usual control algorithms realize flux weakening by adding a negative I d current component when the voltage required by the current regulation exceeds the maximum voltage depending on the battery. If the magnet can be totally defluxed, then it is better to use a maximum torque per volt strategy. Furthermore, there is no speed regulation in the control and the driver gives a torque reference. This reference value has to be limited by the attainable operating points; therefore, the battery power limit has to be taken into account in addition to the voltage and current limits. The d-q current references are calculated to minimize the total current magnitude required to reach the reference torque. This paper proposes a strategy to control a PMSM operating continuously since the speed zero up to the maximum speed without the switching algorithm, in order to take into account the different limitations (current, voltage, and power) and to expand the overspeed zone. In order to validate the proposed strategy, experimental results are shown for a low power machine.Index Terms-Flux weakening, high speed permanent magnet synchronous motor (PMSM), maximum torque per volt (MTPV), torque-speed characteristic, vector control. NOMENCLATURE LPFLow-pass filter. MinCPT Minimal current per torque. MTPAMaximum torque per ampere.

High speed Permanent Magnet Synchronous Motors (PMSM) are used in electrical vehicles because of their strong power density. The high speed implies a big electromotive force and requires flux-weakening. Usual control algorithms do flux-weakening by adding negative Id current when the required voltage by the current regulation exceeds the maximum voltage depending on the battery. If the magnet can be totally defluxed then it is better to use a Maximum Torque Per Volt (MTPV) strategy. Furthermore there is no speed regulation in the control and the driver gives a torque reference that has to be limited to the reachable operating points; the battery power limit has to be taken into account in addition to the voltage and current limits. The d-q current reference are calculated to minimize the total current magnitude required to reach the reference torque. This paper presents a salient pole PMSM control that calculates the minimum magnitude current references to obtain the reference torque while respecting the voltage, current and power limits with a flux-weakening strategy and including MTPV operation in a unified algorithm. The effectiveness of the proposed method is observed through simulations.

The digital control of high speed Permanent Magnet Synchronous Motor (PMSM) is carried out as part of mass reduction for embedded systems. This paper focuses on improving the stability of high speed PMSM digital control. A study of the stability of the closed-loop system shows the importance of taking into account the constraints of digital control (delay and discrete characteristics) in the calculation of decoupling terms and inverse Park transformation. The proposed decoupling is done using the discrete model of the system and a prediction of the evolution of the currents during the delay. The voltage, calculated by vector control, is applied to the machine by a three-phase voltage-source Power Width Modulation PWM converter. This voltage is modified because the rotor reference d-q frame turns during the delay and the switching period. The voltage error caused by this rotation can be significant at high speeds and more generally when the ratio of the switching frequency over the electrical output frequency becomes too high. To stabilize the system, we suggest predicting the average electric angle that the motor will have during the application of the voltage vector. The efficiency is shown through a time simulation.

Synchronous motors with Interior Permanent Magnets (IPMSM) are particularly effective for operation at high speed due to their high power density. However the speeds generate large electromagnetic forces and it is necessary to set up a flux-weakening. Classical control algorithms do flux-weakening by injecting a negative Id current in respect of current norm. The value of this current is usually defined on the LUTs (Look Up Tables) depending on the speed and the torque required reflecting open-loop operation. This article presents continuous control law in closed loop ensuring the operation from the speed zero up to the maximum speed. The generation of the current reference Id is based on strategy to Minimum Current per Torque (MinCPT) followed a Maximum Torque per Volt (MTPV) strategy. The goal is to maximize the speed for a given power. The effectiveness of this unified algorithm is tested on an experimental platform.

No abstract