Battery AI 2.0 project

Battery AI promises better battery management

Aimed at providing online health monitoring and residual lifetime prediction for battery assets, Battery AI 2.0 utilizes artificial intelligence and semi-physical methods. The tool is already in use on DNV’s Veracity platform, eliminating impractical, time-consuming and destructive testing for industry stakeholders.

Battery performance has been a big concern, as batteries are widely deployed in stationary, mobility, and maritime applications as part of the energy transition. This performance is affected by many operational factors, which makes monitoring the accurate battery state of health, a key parameter reflecting available capacity, a challenge for battery asset management. 

The failure of a system in service could have negative or dangerous consequences. Battery state of health is normally determined by offline and time-consuming testing, but this is not an optimum solution as the testing cannot always replicate real operating conditions and battery variability.   


Online battery health monitoring solution

Battery AI 2.0 is a major upgrade and expansion of Battery AI 1.0, which focused on the development of the battery degradation analysis tool. 


By utilizing artificial intelligence to learn from more than one million channel-hours of lab testing data, Battery AI 2.0 is an effective tool that can determine the relationship between operating conditions and battery degradation and provides a bespoke model for battery state of health estimation.  


The combination of AI modelling and cycle counting method also provides a practical solution to efficiently and accurately access the actual battery state of health under intended operating conditions. 


Using a real-time data stream from physical operating systems, the solution can also be used to provide online health monitoring and residual useful life prediction. This permits relevant stakeholders to know the exact performance of the battery asset and thus optimize their asset management in an informed manner. 

Battery testing services increasingly require in depth analytics to obtain insights from raw data. Battery AI provides the ability to simulate real world operating conditions for different application scenarios based on limited test data and project expected battery life and productivity more accurately, thus vastly improving financial planning security.

  • Martin Plass ,
  • Business Development Leader ,
  • New York Battery & Energy Storage Technology (NY-BEST) Test Center
The benefits

Battery AI 2.0 provides a real-time and highly efficient method to assess the battery health status and to eliminate costly, time-consuming and destructive testing procedures. With third-party assessment, customers can gain an objective and independent understanding of asset health and performance, thus enhancing safety, and can ensure warranty conditions are met. In addition, monitoring under operating conditions determines the suitability of the battery system and changes needed for future projects. 

As a proven energy storage source, batteries have been widely used in energy storage solutions, electrical vehicles and vessels and represent an important component in the power solution. At the same time, battery asset health management has been the most important issue in design, operating and recycling stages across the whole life cycle of batteries. Battery suppliers, investors and operators are working continuously to improve energy utilization efficiency and prolong battery life to obtain the best return on investment. Based on its Battery AI tool, DNV can provide advisory and assurance services for relevant stakeholders.