As the core driver of the digital economy, digital technology has gradually become a new driving force for enterprise development and promoted the evolution of core competitiveness of enterprises. In this context, digital transformation has become a necessary option for all enterprises and a prerequisite for sustainable development, and embracing the digital economy has become a common choice for enterprises. However, from the actual situation, C-end industries such as retail e-commerce, finance and other fields are in the forefront of digitalization, while the digitalization process of traditional real economy industries represented by manufacturing and energy heavy industries is relatively slow. As the pillar of the national economy and the key areas of policy support, it is urgent for the real economy to accelerate the digital transformation.

Taking the manufacturing industry as an example, the transformation and upgrading of China’s manufacturing industry in the past focused on the construction of information systems and the opening up of internal information of enterprises, mainly reflected in the construction and upgrading of large business systems such as ERP, and paid more attention to process driving. With the diversification and individuation of downstream demand becoming the mainstream trend, data-driven began to become the mainstream mode of transformation and upgrading of manufacturing enterprises, the digitalization of product design, research and development, production and manufacturing has become the core competitiveness of enterprises, and foreign industrial software service providers continue to deepen the Chinese market based on advanced technology and profound industry understanding.

01

CAE technology and AI technology dual integration

Help enterprises take off in digital transformation

As the core component of industrial software, R&D and design software such as CAE (Computer Aided Engineering) is the main tool for the digital transformation of manufacturing enterprises and the key competitive field of software providers. Based on three-dimensional solid modeling, CAE provides basis for product development and design by simulating the performance of products in structure, fluid, heat, electromagnetic field, etc., and is widely used in manufacturing, energy heavy industry and other fields.

Based on rich simulation models and industry data, the application of CAE can effectively help manufacturing enterprises reduce or even avoid the repetitive work of multiple recalls and optimization in the product design stage, and help enterprises reduce costs and improve efficiency. Driven by “intelligent manufacturing”, the importance of CAE for manufacturing enterprises continues to increase.

At the same time, the global market competition is becoming increasingly fierce, taking the automobile manufacturing industry as an example, the car manufacturing cycle has been shortened from the past 3-5 years to the current 1-2 years, which will inevitably put forward higher requirements for the efficiency of each link, especially in the product research and development design link will pay more and more attention to the accuracy and output efficiency of the simulation model. However, the traditional CAE 3D modeling technology is gradually difficult to meet the requirements of enterprises for such high timeliness and realistic model effects, forcing service providers to constantly explore better solutions.

With the continuous evolution of AI technology, AI-based machine learning can obtain more accurate prediction models by training neural networks based on existing large amounts of data, and AI has become a key application technology for manufacturing enterprises in research and development design. The deep integration of AI technology and CAE technology, using the large amount of data accumulated by CAE in the manufacturing industry as the basis for deep learning, will enable the continuous optimization of CAE modeling paradigm and further reduce computing costs. Observing this trend, the world’s leading CAE service providers have begun to explore the integration of AI technology and their own products, and actively embrace more possibilities of AI+CAE.

As the world’s leading CAE service provider, Altair initially focused on helping automotive enterprises apply engineering simulation technology. After observing the pain points of digital transformation in product development and design of traditional enterprises, Altair gradually improved the solution of integration of simulation, high-performance computing and artificial intelligence technology through active research and development, mergers and acquisitions.

Altair also noted the development opportunity of AI+CAE, “Through the deep integration of simulation technology and AI technology, combined with the rich data accumulated internally, it can provide customers with simulation results closer to the real needs and better user experience.” Liu Yuan, general manager of Altair Greater China, said in an interview with Yiou.

On the one hand, the deep integration of AI and CAE can better achieve “what you want is what you get”, that is, to achieve the productization of ideas and needs. This is in line with Altair’s internal Physics AI concept, which quickly builds machine learning models based on a large number of existing simulation results, and can help enterprise customers quickly build new models and output results.

On the other hand, from the perspective of digital twins, there are two ways to build digital twins inside Altair. One is based on traditional 3D modeling. Although this method can accurately depict the model, it is slow in practical application and cannot be displayed in real time. The second path relies on the romAI tool to achieve deep integration of CAE technology and AI technology, and uses machine learning to reduce the three-dimensional model to one dimension, so as to display the simulation results more quickly. In fact, Altair, through the integration of CAE technology and AI technology, can achieve minute-level automotive crash test model results output.