Blog Author: Emna_Azek
There are several distinct ways in which controller firmware for power electronics (PE) applications can be developed and tested. Of course, the full-scale or scaled-down prototypes are still the favorite playgrounds for all PE engineers, but those toys are quite expensive and more importantly, they are a nice (sic) compendium of Murphy’s law examples – if something can go wrong it certainly will and it’s going to take a lot of time to make the prototype operational again. So, PE engineers must tread carefully and the prototypes are usually allocated for the final testing phases.
For various objective and subjective reasons, software-in-the-loop (SIL) and controller-hardware-in-the-loop (C-HIL) have emerged as the two approaches that take the cake for the software development and various testing stages.
The outline of one platform for SIL and C-HIL tests is depicted in Figure 1. It consists of an emulator (Typhoon HIL emulator HIL602+), an interface board with the controller (AURIX™ TC275 Lite Kit or AURIX™ TC375 Lite Kit, for example), and a PC.
The emulator mimics the behavior of the power electronics power stage. The interface board adjusts analog and digital signals exchanged between the emulator and the controller. Two sets of software tools are used on the PC – one relating to the emulator and one relating to the controller.
The power stage to-be-executed in real-time on the emulator(s) is designed using the Typhoon HIL Control Center tool called Schematic Editor. Moreover, power stage-related data is collected and depicted using the Typhoon HIL Control Center tool called HIL SCADA.
The controller is programmed using AURIX™ Development Studio and the internal variables are observed using OneEye graphical user interface.
Figure 1. C-HIL platform for the development and testing of EV applications.
Figure 2. The setup consists of the emulator, interface card and controller, and PC (PC not depicted).
Figure 2 depicts the hardware setup that corresponds to the scheme shown in Figure 1.
Figure 3 shows the Schematic Editor window in which the elements of the power stage are inserted, arranged, and adjusted. Once finished, the schematic is compiled and downloaded to the emulator for real-time execution. This approach corresponds to the C-HIL paradigm.
Alternatively, in Schematic Editor, besides the power stage, users can implement control structures using the Signal Processing toolbox (drag-and-drop, graphical type of the control structures synthesis). Then the schematic, containing the power stage and the pertaining control structures, is compiled and executed on the PC via the virtual HIL (VHIL) toolset (admittedly not in real-time). This roughly corresponds to the SIL paradigm.
Furthermore, if the control structures are not too complex, they can also be deployed on the emulator. In this configuration, both the power stage and the control structures are executed on the emulator in real-time, and an external controller is not necessary. With two clicks, the user can choose whether the schematic is going to be downloaded to the emulator or if it is going to be executed on the PC.
Figure 3. Schematic editor window.
The controller can be programmed using two approaches. In the first, “traditional manner”, the software is developed using solely AURIX™ Development Studio. In the second approach, colloquially called rapid control prototyping is used - the firmware that configures peripherals and interrupt routines can be implemented in AURIX™ Development Studio, but the content of the interrupt routines (regulators, modulator, etc.) can be exported from the Schematic editor once such structures were developed in an aforementioned manner (using Signal Processing toolbox). The simple copy/paste function of the (auto)generated .c and .h files into the AURIX™ Development Studio project completes the project creation and the code is ready for compilation and execution on the controller.
The former approach was extensively used by Ivan Todorović from the Faculty of Technical Sciences (Novi Sad, Serbia). The latter paradigm is the topic of ongoing development. Still, a series of tests were already conducted, mostly by the students from the Faculty of Technical Sciences, covering various applications – from machine control, battery chargers to energy management schemes. One set of responses is shown in Figure 4 - recorded for the power stage shown in Figure 1, before the dead-time violation problem was solved.
Now, why is this setup a great environment for the development of EV applications (among others)?
Since it is equivalent to a sandbox concept, extensively used for software development, it represents a rather safe environment for students and young researchers to learn through their own mistakes. There is quite literally nothing that can be destroyed. Next, both the AURIX™ - related toolchain and Typhoon HIL toolchain are easy to use and can be mastered in a manner of weeks. Moreover, the HIL 6 series emulators and AURIX™ TC275/TC375 are capable of “handling” fairly complex and diverse power stages, in real-time. Finally, the environment is the epitome of versatility – in one moment battery chargers can be tested and in another energy management schemes.
Figure 4. Responses for recorded in a setup consisting of the battery, inverter and PMSM.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
By technically required cookies we mean cookies without those the technical provision of the online service cannot be ensured. These include e.g. cookies supporting essential services like a smooth reproduction of video or audio footage. So called ‘functional cookies’ are also assigned belonging to this category. Functional cookies store information in order to provide you comfortable use of our online services (e.g. language selection). The legal basis for the processing of personal data by means of cookies of this category is Infineon’s legitimate interest. This includes, among other things, the interest in having a professional external presentation as well as an optimal balancing of the loads on the server due to technical reasons.
By performance and marketing cookies we mean cookies which are technically not required. We use performance and marketing cookies only if you have given us your prior consent. With such cookies, we collect information about how users interact with our website and which pages have been visited. This helps us to understand user activity on our website on an aggregated as well as on a personal level to provide you relevant content and services.