As the interface between wind power generation systems and the power grid, inverters play a crucial role in power conversion and control. They are also one of the most vulnerable components in the system, often leading to failures. The quality of power delivered to the grid or load heavily depends on the performance of the inverter. Ensuring stable grid operation and improving power quality makes fault diagnosis of the inverter a critical task. In recent years, research into inverter fault diagnosis has become a major focus for scholars worldwide.
The TMS320F28335 DSP, introduced by Texas Instruments, is a 32-bit floating-point digital controller with a main frequency of 150 MHz. It offers rich peripherals, high cost-performance, large memory capacity, and fast processing speed [4-5]. This processor has been widely used as the core controller in intelligent control and fault detection systems for transformers.
Inverter systems are complex hybrid systems that involve multiple interdependent components [6]. Traditional programming methods for DSP systems are time-consuming and labor-intensive. To address this, MathWorks and TI developed the TSP tool, which allows modeling, simulation, code generation, and debugging within the Simulink environment. This significantly improves engineering efficiency. This paper implements automatic code generation for an inverter system, streamlining the development process.
Automatic code generation technology involves using software like MATLAB or specific toolboxes to build a system simulation model that automatically generates embedded application code based on a target configuration [7-8]. Embedded Coder, a powerful tool from MathWorks, supports embedded system development. The TSP TI C2000 toolbox, jointly developed by TI and MathWorks, integrates seamlessly with TI’s CCS IDE, enabling efficient development for C2000 series DSPs [9-10].
This tool provides direct interface modules for DSP peripherals, transforming the system model into optimized, portable, and customizable embedded C code [11-12]. The model source and signal receiving modules are replaced with I/O ports, while the system.tlc file coordinates the entire code generation process. This approach eliminates the need for manual coding, making debugging easier and reducing development time, cost, and effort.
The development process based on code generation starts with defining system design standards and building a simulation model in Simulink. Parameters and the simulation environment are configured according to the system requirements, and intelligent algorithms are integrated. After simulation, if results deviate from expectations, the model or parameters are adjusted until the results align with theoretical predictions. Once the simulation is complete, the target environment is configured, and the code is generated, downloaded to the hardware, and tested.
A three-level inverter is a common power electronic topology formed by combining two-level converter bridge arms in series or parallel [13]. Unlike traditional two-level inverters, it can output more than two voltage levels. A single-phase bridge arm of a three-level inverter can be constructed, and three identical bridge arms connected in parallel with a DC power supply form a full-bridge structure. IGBTs are controlled using a modulation algorithm to produce a three-level AC voltage waveform. The modulation method reduces voltage stress and du/dt, resulting in a waveform closer to a sine wave compared to two-level inverters.
Generating three-level PWM pulses manually in CCS is a complex and error-prone task. To improve efficiency and reduce errors, automatic code generation is used. A three-level PWM model is built in Simulink, and the Digital Output module in TSP defines the output ports. The PWM module from the Power Systems toolbox is used, with parameters such as frequency, phase, and sampling period set to generate a 50 Hz output. The 12 pulses required for control are distributed across two Digital Output modules using Demux and Mux blocks.
After setting up the model and target environment, the code is compiled and generated. The .out file is then downloaded to the DSP via CCS. The entire process simplifies development, reducing manual coding and improving accuracy and efficiency.
A test system was designed using the TMS320F28335 as the main controller, consisting of a PC, power supply, expansion module, optical isolation module, control module, and inverter module. The system can support both two-level and three-level topologies, providing a platform for researching intelligent control and fault diagnosis. The generated three-level PWM code was tested on the physical system, with observed pulse signals and line voltage waveforms matching the simulation results.
The conclusion highlights that the proposed code generation method significantly reduces development time, lowers error rates, and increases efficiency. It provides a practical solution for verifying inverter control and fault diagnosis algorithms, offering high value for real-world applications.
Chamfered Radiator,Galvanized Chamfered Radiator,Corrosion Resistant Chamfered Radiator,Power Transformer Chamfered Radiator
Shenyang Tiantong Electricity Co., Ltd. , https://www.ttradiator.com