Reliable Distributed Optical Fiber Sensing System for Early Warning of Failure in Energy Storage Systems

Grid-tied battery energy storage systems (BESS) are versatile assets for power generators, enabling renewable energy and a more resilient, distributed grid. Grid-tied BESS will be a complex set of battery chemistries, and the control design is critical to ensuring the batteries are used properly. The control system usually includes a power conversion system (PCS), energy management system (EMS), and battery management system (BMS). These systems serve specific power delivery functions with sensor suites, which help create a clear picture of the systemĺs most important variables. Current sensor suites have limitations; for example, distributed battery cell temperature sensing leads to increased sensor complexity and many points of failure because of the large number of cells and sensors in a typical grid-tied energy storage system. Additionally, no solutions can physically measure variables such as component swelling, gas emissions, and leakage. The Fiber Optic sensing solution provided by Sensible Photonics hopes to fill the monitoring gap for these variables (i.e. component swelling, gas emissions/leakage) in combination with distributed temperature sensing in a single sensor system. Phase I activities will explore using fiber optic sensors for battery applications. The sensorĺs capability to provide early warning of failures in energy storage systems will be assessed. The ability to provide distributed temperature sensing, cell swelling, and cell leakage or gas emission would help isolate failures to single cells in lithium-ion battery failure scenarios. Similar failure pathways with other battery chemistries and systems will be researched. Failure pathways can be tested to monitor the effectiveness of the fiber optic sensors. We will measure early warning and the capability to isolate failures before catastrophic failures. Examining how system complexity could be decreased compared to incumbent solutions and direct performance comparisons will be evaluated. Phase II work will include sensor ruggedization, generating data sets for event classification, designing algorithms that prevent false positives and delivery of a high-fidelity signal. Additionally, applying the sensor technology to additional control systems such as the PCS will help understand how a fiber optic sensor system can monitor PCS failure pathways. This will help create a sensor system that can be applied to all modern battery chemistries and will harmonize sensor usage across the controls systems (BMS, PCS, etc.).