Model-Based Design for STM32 Nucleo boards applied to motion control

In the field of Power Electronics, several types of motor control systems have been developed using microprocessors. Many of them are based on systems of differential equations, use state observers and need to handle multiple tasks in order to execute an optimal operation of the motor. The more complex is the system, the harder is to implement. Traditionally, many steps are carried out from the modelling stage to the real operation of the control systems, and require iterative tests to ensure the absence of error. A major part of the effort is focused on embedded software representation of the control system, which demands a large amount of code lines that may contain hard-to-detect bugs. The Model-Based Design is a method in which most of the design is centred in the model, while the code representation and embedded integration is automatically carried out. In this work, simplified implementation and validation methods are proposed using Model-Based Design for motor control systems, exemplified with an I-Hz control. Taking advantage of the Matlab/Simulink graphical tools, a control system can be modelled as a combination of block diagrams and state machines to reduce the mathematical complexity and schedule multiple tasks. Also, the STM32 Embedded Target toolbox is used for automating the generation and integration of the modelled control algorithm to an embedded software representation compatible with the STM32 Nucleo boards, avoiding to write any code manually and carry errors. Validation of the designed control system is eased when exactly the same simulated control algorithm is integrated to a microprocessor because the names of the variables are shared. The variables involved in the operation of the control system can be read from the microprocessor's memory using a serial monitor with USART communication protocol. Minimal differences took place when the real and simulated results where confronted due to some nonidealities, but under certain tolerance ranges that allowed the model to be endorsed. In conclusion, the proposed method for designing motor control systems allowed to skip or to simplify many steps that are traditionally carried out, specially what regards to the mathematical modelling and code programming, so time and effort reduction was successfully accomplished.