MathWorks, a leading global developer of mathematical computing software, announced today that Professor Zhang Jilei’s team from the School of Information Science and Technology at Southwest Jiaotong University, with the support of the National Natural Science Foundation of China and major national strategic demands, has relied on the engineering technology solutions of MATLAB® and Simulink® A new type of distributed cooperative braking control system has been developed. This system resolves the bandwidth and communication issues of electronically controlled pneumatic (ECP) in traditional schemes and enhances the safety and reliability of heavy-haul train operation. By calculating the adaptive braking force based on the characteristics of each carriage and the terrain they are located in, the braking distance is shortened and the longitudinal impact is reduced.
The project adopted a workflow based on Model-Based Design (MBD), breaking through the technical bottleneck in China’s heavy-haul train safety technology. The integrated MBD workflow covers the entire process from concept development to detailed design, hardware implementation and test verification, supporting key activities such as algorithm design, dynamic simulation and hardware verification, and can significantly shorten the development cycle.
Professor Zhang’s team used MATLAB to establish a multi-body dynamics model of heavy-haul trains containing over 200 differential equations, precisely simulating complex factors such as carriage coupling, track slope, and air resistance, achieving joint simulation of control algorithms and mechanical systems, and designing a distributed braking algorithm Later, the physical model of the braking system will be built through Simscape™ to ensure compatibility. By leveraging Simulink to build a real-time simulation platform, the algorithm is deployed to the STM32 microcontroller for hardware-in-the-loop (HIL) testing. In the laboratory, the full-load emergency braking scenario of 80-ton to 100-ton class trains is simulated, replacing traditional CAE software and high-risk real vehicle tests. Through batch simulation, the combined optimization control parameters of four types of working conditions – uphill, flat, downhill and curved – were rapidly verified, which shortened the average braking distance and reduced the couhook force to within 150kN, effectively lowering the longitudinal impact force and reaching the international leading level.
I am currently in charge of a National Natural Science Foundation of China general project: Research on Adaptive Optimization Control Method for Distributed Cooperative Air Braking of Autonomous Heavy-Haul Trains. Professor Zhang from Southwest Jiaotong University stated, “The MATLAB/Simulink toolchain is the core engine for us to overcome technical difficulties.” From algorithm iteration to hardware deployment, the model-based design approach enables us to focus our main efforts on innovation itself rather than repetitive engineering verification. We conduct desktop simulation and HIL testing to quickly verify the control implementation concept and iteratively update the algorithm, making our project implementation faster, more efficient and more cost-effective compared to traditional methods.”
Mr. Zhang Hui, General Manager of MathWorks China, emphasized: “We have always adhered to the concept of accelerating the progress of science and engineering, providing high-quality localized services for users of industry, academia and research in China, and building a ‘bridge’ for the transformation of scientific research achievements from the academic circle to the industrial circle.” The successful case of Professor Zhang’s team from Southwest Jiaotong University proves that model-based design continuously helps researchers in both academia and industry solve engineering problems of complex systems and promotes the intelligentization process of the rail transit industry.
Illustration: HIL test setup, including oscilloscope, cylinder, intake and exhaust valves, and STM32 control board
