An unborn battery pack has spent its "life" in the virtual world ". In the digital twin system, engineers simulated it from cell sorting to module welding, from PACK integration to vehicle loading, from daily charging and discharging to retirement and dismantling ten years later. Every possible fault, every inflection point that can decay at one time, and every concentration of structural stress are discovered and optimized in advance in the virtual space. This ability of "rehearsing life" is exactly the core competitiveness of the intelligent manufacturing system of energy gathering new energy.

Virtual trial and error in the design phase ". Traditional battery pack development relies on repeated production and testing of physical prototypes, with a cycle of 6-8 months. The digital twin platform that gathers new energy resources will compress this cycle to 8 weeks. Before the introduction of new products, engineers carved the entire production line in the virtual space 1:1 to complete thermal simulation, structural strength simulation, electrochemical simulation and vibration simulation. The design of a liquid-cooled runner of AGV Battery pack found the problem of insufficient heat dissipation of edge cell in the virtual simulation. By adjusting the reducing ratio of runner, the temperature difference of the whole package is compressed from 5℃ to 3℃, the rework of the physical prototype stage is avoided.

The real-time image of the manufacturing phase ". The operating parameters of production line equipment-laser power, welding speed, gluing track, environmental temperature and humidity-are imported into the digital twin model in real time and compared with the virtual ideal state. When deviation occurs, the system automatically alerts and proposes to adjust the plan. In a certain welding process, the digital twin system detected abnormal fluctuation of molten pool temperature, and traced it back to protect the drift of gas flow sensor, and the maintenance team completed the replacement within 30 minutes, the welding quality risk of the whole batch of products is avoided.

Predictive prophet in the O & M phase ". The digital twin model runs continuously on the cloud at battery pack of the factory, receiving the actual running data synchronously. AI algorithm comparison virtual preTest and actual performance to identify early signals of attenuation trend. A customer's energy storage is battery pack. After 18 months of operation, the digital twin model predicts that the internal resistance of a string of cell will exceed the threshold within 3 months. The new energy source will issue maintenance suggestions in advance, the customer arranges planned shutdown and replacement to avoid the loss of unplanned power failure.

"Circular planning" in retirement stage ". When battery pack approaches the end of design life, the digital twin model evaluates its remaining capacity and health status, providing decision-making basis for echelon utilization or recycling and dismantling. Capacity attenuation to battery pack of 80%, it is recommended to transfer to energy storage power station to continue service; Attenuation to 60%, enter the recovery process, lithium, nickel, the recovery rate and economic value of cobalt and other metals are calculated in advance in the virtual model.

The value of digital twins lies not only in shortening the development cycle or reducing the failure rate, but also in changing the battery pack "life cycle management" paradigm-from "forgetting at birth" to "tracing the whole process", from "passive maintenance" to "active prevention", from "disposable products" to "circulating resources". Gathering energy and new energy believe that the future battery Manufacturing instead of producing physical products, manage digital assets.