Introduction to digital control for electric machines

In recent years, the control of electrical machines has undergone significant progress. These advances are essentially due to the technological revolution in digital electronics, which has enabled the development of efficient digital solutions with the possibility of implementing more complex algorithms. The first implementations of electrical machine control algorithms were based on analog solutions. These solutions ensured high-bandwidth, high-resolution control. However, they lacked reliability, given their sensitivity to disturbances and variations in control parameters linked to the thermal constraints of analog control circuits. Subsequently, digital solutions naturally emerged to overcome these drawbacks. Today, most digital implementations are based on microprocessors and DSPs (Digital Signal Processors). These digital solutions are equipped with ALUs (Arithmetic Logic Units) dedicated to carrying out the arithmetic and logic calculations of control algorithms. They also incorporate peripherals such as analog/digital converters and timers, well suited to the control needs of electrical machines. Their use has solved the problems associated with analog controls. In addition, they offered significant cost benefits and design flexibility. It should also be noted that these digital solutions are purely software-based. Indeed, the designs associated with them are conceived using predefined architectures. The designer cannot therefore act on the hardware part of the design, and only has access to the software part. On the other hand, despite the advantages offered by these digital solutions, certain advantages offered by analog implementations are lost. This is mainly due to the fact that the discretization and quantization of the control algorithms to be implemented, as well as computation time delays deteriorate control performance in terms of correction speed and control resolution.

New digital hardware solutions such as FPGAs (Field Programmable Gate Arrays) or ASICs (Application Specific Integrated Circuits) can also be considered as suitable solutions for improving control performance and regaining some of the advantages of analog control [D 2 902] . The inherent parallelism of these new digital solutions, as well as their high computing capacities, mean that calculation times are negligible, despite the complexity of the algorithms to be implemented. What's more, compared with software solutions, hardware solutions give the designer access to the hardware architecture,...