Benchmarking Project Development Case

Overall Case

• Freiberg laboratory: (Germany) DFG project of German scientific research association, artificial intelligence + offline model
• Morgan: (US, now Primetal) more than 100 sets of model development / off-line model development of manufacturing process
• Cascade Steel: (US) three sets of Level 2 development / online model development
• Evraz Oregon Steel: (US) model requirements ↑, developed a new generation of Level 2
• POSCO: (Korea) production line process / equipment optimization ← online intelligent design software
• NISCO / TISCO Level 2 optimization: application of a new generation of Level 2
• Taiyuan University of science and technology: manufacturing / research and development of lithium battery intelligent equipment
• BYD et al.: lithium battery manufacturing, Level 2 development, soft sensing technology
• Tesla: (US) lithium battery pole piece defect model (Nevada super factory + California Research and development)

German DFG Finite Element Artificial Intelligence Project: Model Development

The temperature calculation and force energy calculation of high-temperature machining are simulated by thermodynamic finite element method to realize the large deformation of materials. Before that, one deformation needs 3 days and 3 nights; This project reduces the calculation to 20 hours; There is another simplified model, which can calculate a pass in 5.5 minutes! This work creates a world precedent for the application of finite element in accurately determining large deformation in the world! Technology has passed through several major technology companies in the world, and then spread to factories all over the world! See Project details.
 

Model Development Case in Primetal (Morgan): Data Model And Finite Element Model

Based on the data collected by Morgan all over the world for more than 100 years, Morgan's five-year laboratory measurement results, the research and development results of its own research institute (more than 40 doctoral thesis results) brought by the team leader from Germany, and the results currently collected by Morgan team. For example, the data collection of Morgan engineer team for more than one month is available for only one plant in Cleveland. Based on a large amount of data and data collected from all over the world (based on their own R & D funds), the leader of our team has developed more than 100 sets of production process models for predicting material deformation, force and energy parameters, microstructure / mechanical properties, etc. See relevant parts of model system development.

Oregon Company: Micro Mechanism And Model Of Material Forming

The microstructure model and machine learning logic are combined into the modeling process, and combined into the large-scale software Level 2, which greatly improves the accuracy of the model, and makes the production line defective almost every day and optimized to be free of defects for half a year. See Oregon company in the development of new generation Level 2.

POSCO Project: Intelligent System Application Of Internet Platform

There are 9 restrictive factors that restrict each other. For example, the roll diameter of the equipment cannot be too large due to the capacity of the crown block, the width of the equipment is limited by the space, the diversification of products requires the equipment to have a specific shape, the products are a variety of stainless steel, the processing methods are very different, and some have extremely high strength and difficult deformation characteristics, etc. Using the developed app on Metal Pass (now www.Meta4-0.com), the design was successfully completed. Each step involves a large number of model calculations.

Model Based Production Optimization Of An Enterprise In Nanjing

Using computer programming and model calculation, more than 30000 data files of various materials in different temperature zones for Level 2 calculation are designed, and each data file has more than 100 data. The previous achievements of the new generation Level 2 are applied to this project. See relevant parts in the development of  new generation Level 2. 

Metal Pass Industrial Internet Process Optimization Based On Online Model

The app written by the finite difference method on Metal Pass network (now www.meta4-0.com) is calculated based on a large number of field measurement data, such as the model of heat transfer coefficient varying with various factors during water cooling. See the online application part of  model system development.

BYD / Tesla

Developed a burr early warning system for BYD, including model development and accuracy verification, online data acquisition of main influencing factors, soft sensing system for a series of data, tool notch measurement equipment (using machine vision, image data acquisition, thousand times imported microscope, measured tool notch processing, grouping use of tools based on initial quality...) And a series of projects; There are three contracts for lithium batteries alone; At the same time, the team is also the technical supplier of BYD cloud rail and other projects. Similar work also includes the projects of Tesla super factory (Nevada) and Tesla R & D headquarters (California); Go to the site several times respectively.

Electronic Manufacturing Industry

A series of projects in the electronic manufacturing industry in the Pearl River Delta, such as two of the world's top color TV manufacturers, SMT production line problems, stamping and injection molding sections, digestion of a series of introduced technologies, how to guide the technical design of Smart Manufacturing team

5G New Material Industry

PI film production line, 5g material production line (domestic equipment hardware, foreign equipment software and hardware), intelligent scheme of a series of investment projects of a large company

Others

There are also a large number of projects, such as various projects in the garment industry, electric vehicle improvement projects, and so on; Based on thousands of material data (rheological stress, Young's modulus, heat capacity, etc.), a model varying with temperature (some are also affected by deformation and deformation speed) is established, and the model error is usually less than 1%
 
Project Performance and List

The team has a German doctor of engineering and an American doctor of software, and has several years of working experience in seven or eight professional fields of engineering and software (process, product, equipment, automation, model, software architecture and development, database data, etc.). See the business list and business resources in the past 30 years.