Physical modeling solution for the development of

Physical modeling solution for mechatronics system development

mechatronics system development and its challenges

the concept of "mechatronics" has been put forward for about 40 years. Generally speaking, it refers to a comprehensive technology combining mechanical technology, electronic technology, sensor testing technology, interface technology, information technology, computer technology, automatic control technology, etc. Mechatronics system is a new type of mechatronics system with mechatronics technology. We divide the mechatronics system into four parts: mechanical part, electronic part, control part and software part. Mechatronics system embodies integrity, which is a collection of parts that are organically connected by mutual differences and interactions to complete certain functions for the same purpose. It is not a simple combination or patchwork of these technologies. Based on different concepts, the industry has different views on this definition

the development of mechatronics system will face many challenges. First, manage the complexity of different fields; Second, any system development is based on subsystems and more subsystems, and these subsystems are the collection of these four fields; Third, the traditional approach is to make the prototype in four directions, then integrate it, and finally see whether it works. The defect of this method is obvious. If we find errors after integration and then repair them in the early stage, it will cause a great waste of cost. Therefore, we need to find design defects in the early design process as much as possible; Fourth, the optimization of design process. It is likely that we can get an optimization or a better design scheme in any local area, but after integration, we find that its overall performance cannot be optimized. So the most important challenge is to draw it in the whole design process, while considering the flexibility and cost effectiveness of each subsystem. We can reuse the designed or verified models by selective means, and then add different constraints, so as to apply them in new fields

mechatronics system development solution

the mathworkstm provides two solution platforms, one is the industrial standard language Matlab based on scientific computing The platform, MATLAB, was first released in 1984 and has a history of 25 years so far. It provides an open interactive environment integrating analysis, visualization and advanced programming for algorithm developers. Based on MATLAB and related application toolbox, it can help engineers complete data acquisition (from various software, hardware and databases), information mining (analysis and visualization) Scientific research work in different industries and fields such as system algorithm development and result sharing (reporting and Publishing). The second core product development platform is simulink, It was first released in 1990. It is based on the MATLAB platform and adds functions based on model design, system level simulation and embedded system implementation. Simulink implements a modular design environment for modeling, simulation and complex systems, especially control systems, DSP and communication systems. When MATLAB and Simulink are combined together to serve customers for a long time, the whole process from the design and development of Mechatronics System Based on model design concept to simulation implementation is realized

when implementing algorithm development on this platform, our physical modeling of the research object and the simulation of these objects under various environments are model-based development. We can combine the model environment and algorithm for simulation, so as to put forward guiding opinions on the design. After verification, we can download the algorithm directly to the chip level and hardware level. For example, we can automatically generate embedded C code and put it on DSP. People with electronic engineering background may be more concerned about VHDL code generation. C code generation is a mature product 15 years ago. In these two years, our algorithm can directly generate VHDL code. Verification at different levels is very important in the whole development process, rather than evaluation at a later stage. From the early model to the generation of embedded level code, we all need tools for validation and verification

how to develop mechatronics system under this platform? The previous practice is to study the system first, and get 1 Adjust the experimental length of the collet position of the experimental machine device (the average distance between clamping wires) to 180mm ± 1mm. After its attribute or dynamic response, extract its physical equation. This requires us to put forward the friction pressure 120 ± 0.2N equation manually. The derivation of this equation is very difficult, but we can do so on the new platform. The application case of this method is that Agfa company controls the flow of printer paper, and makes a control development in this way. Remote control technology is to complete the derivation and modeling of models and equations on the Simulink platform, then do a simulation, and finally generate code. This was actually achieved 15 years ago

machine alternating test box has the function of constant test box. How can the electrical integration system combine different fields? Simulink, especially physical modeling products, has provided joint simulation solutions since 1990. Themathworks has good cooperation with many professional tool manufacturers in both mechanical and electronic markets, such as CAD software and EDA software. Because these tools need to be embedded and integrated with each other, we support this kind of joint simulation. Some of these joint simulation solutions are provided by the MathWorks, and some are provided by third-party software providers. Joint simulation means that on a platform, the controller development uses the MathWorks' own products, and the controlled object, such as aircraft, comes from other products. The more detailed design or structure of the controlled object is completed by a third-party professional software, and the MathWorks only provides some accessories and interfaces to complete the transmission of parameters. The advantage of this solution is that it can reuse existing models and support design and verification. One drawback of this solution is that we need engineers to understand different products of different companies, and may even need to understand the interdependence between different products. In fact, in industrial applications, the efficiency of this joint simulation is very low. Whether it is the development of control algorithm or the generation of embedded software, the direct generation of embedded code from our platform has many practical applications in aviation, aerospace, automobile and train

physical modeling solutions and simscape TM language

the MathWorks can provide a better method - physical modeling solutions

the MathWorks has used physicalmodeling technology in physical modeling for more than ten years, and physical modeling products have corresponding tools in various fields. Among them, simscapetm is a very important technology that has emerged in the past two years. It is the underlying platform for these multi domain physical modeling. The biggest advantage is that it can enable users to avoid the tedious work of deriving equations. Simscape provides some basic libraries to complete the modeling of mechanical, electronic, hydraulic, thermal and other fields. We only need to connect the libraries together to complete a multi domain modeling. When integrating multi domain physical models, the equations in the background can be automatically generated to solve these models. Why do we use these multi domain modeling solutions? For example: we know that some system development is done on the Simulink platform, such as the construction of control system modules. The connecting line between them is called signal line, and the arrow direction of signal line represents the direction indicated by signal flow. This is a very traditional or classic application. Especially in the field of control and signal processing, its flow direction or module construction represents the direction of data flow and information flow. However, some applications in other fields are not so simple, such as the design of a shaking table. Because it is a multi degree of freedom link, it is difficult to describe it with data flow instructions, so new tools are needed to describe this physical modeling. In physical modeling products, we use different modules to build them. The middle signal line has no direction, but only the interaction of data. So when doing system simulation, we will face two kinds of problems, the first is causal description, and the second is non causal description. Products developed on Simulink and simscape platform are to solve such mixed problems, so that some problems can be solved in its appropriate solution environment, which can improve the development efficiency. Its main advantage is that when we do multi domain physical simulation on a single platform, we can find and solve problems before the real hardware is built

the simscape language we launched today is a new programming function, which can write components, domains and function libraries of physical modeling in the Simulink environment. The new language is included in simscape. It uses physical network or non causal modeling methods to extend the Simulink system to the modeling and Simulation of mechatronics system and other multi domain physical systems. Engineers can use it to develop reusable component models and systems to promote the rapid development of technology, such as fuel cells, wind power systems and hybrid electric vehicles. With this new feature, end users and collaborators can expand simscape, create and share reusable models, so as to improve team efficiency and strengthen communication. With these shared physical models, the design team can more accurately simulate the system behavior, develop more robust control strategies, make design tradeoffs, and find problems in the system performance early in the development process. (end)