journal article·2022-09Open access

Enhanced physics-constrained deep neural networks for modeling vanadium redox flow battery

QiZhi He, Yucheng Fu, Panos Stinis, Alexandre Tartakovsky

Abstract

Numerical modeling and simulation have become indispensable tools for advancing a comprehensive understanding of the underlying mechanisms and cost-effective process optimization and control of flow batteries. In this study, we propose an enhanced version of the physics-constrained deep neural network (PCDNN) approach [1] to provide high-accuracy voltage predictions in the vanadium redox flow batteries (VRFBs). The purpose of the PCDNN approach is to enforce the physics-based zero-dimensional (0D) VRFB model in a neural network to assure model generalization for various battery operation conditions. Limited by the simplifications of the 0D model, the PCDNN cannot capture sharp voltage changes in the extreme SOC regions. To improve the accuracy of voltage prediction at extreme ranges, we introduce a second (enhanced) DNN to mitigate the prediction errors carried from the 0D model itself and call the resulting approach enhanced PCDNN (ePCDNN). By comparing the model prediction with experimental data, we demonstrate that the ePCDNN approach can accurately capture the voltage response throughout the charge--discharge cycle, including the tail region of the voltage discharge curve. Compared to the standard PCDNN, the prediction accuracy of the ePCDNN is significantly improved. The loss function for training the ePCDNN is designed to be flexible by adjusting the weights of the physics-constrained DNN and the enhanced DNN. This allows the ePCDNN framework to be transferable to battery systems with variable physical model fidelity.
Source: openaire · Quality: candidate unverified · Source record ↗

Research context

OpenAlex topics: Advanced battery technologies research · Advanced Battery Technologies Research · Supercapacitor Materials and Fabrication

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (40)

Show source-provided references
1. Carbon neutrality: Toward a sustainable future
2. Flow batteries: current status and trends10.1021/cr500720t ↗
3. The chemistry of redox-flow batteries10.1002/anie.201410823 ↗
4. Redox flow batteries: a review10.1007/s10800-011-0348-2 ↗
5. Development of a redox flow battery system
6. Optimization studies on a Fe/Cr redox flow battery10.1016/0378-7753(92)80133-V ↗
7. Novel vanadium chloride/polyhalide redox flow battery10.1016/S0378-7753(03)00621-9 ↗
8. Kinetics of the chemical dissolution of vanadium pentoxide in acidic bromide solutions10.1023/B:JACH.0000031168.03880.1a ↗
9. A study of the discharge performance of the Ti/Fe redox flow system10.1016/0378-7753(84)80054-3 ↗
10. Recent progress in redox flow battery research and development10.1002/adfm.201200694 ↗
11. Characteristics of a new all-vanadium redox flow battery10.1016/0378-7753(88)80005-3 ↗
12. Preparation and characterization of Nafion/SPEEK layered composite membrane and its application in vanadium redox flow battery10.1016/j.memsci.2008.08.025 ↗
13. A high power density and long cycle life vanadium redox flow battery10.1016/j.ensm.2019.07.005 ↗
14. Performance improvements and cost considerations of the vanadium redox flow battery10.1149/08901.0029ecst ↗
15. Data-driven hierarchical optimal allocation of battery energy storage system10.1109/TSTE.2021.3080311 ↗
16. Data driven battery modeling and management method with aging phenomenon considered10.1016/j.apenergy.2020.115340 ↗
17. Data-driven prediction of battery cycle life before capacity degradation10.1038/s41560-019-0356-8 ↗
18. Physics-informed machine learning10.1038/s42254-021-00314-5 ↗
19. Physics-constrained deep neural network method for estimating parameters in a redox flow battery10.1016/j.jpowsour.2022.231147 ↗
20. A simple model for the vanadium redox battery10.1016/j.electacta.2009.06.086 ↗
21. A dynamic unit cell model for the all-vanadium flow battery10.1149/1.3561426 ↗
22. An enhancement to Vynnycky’s model for the all-vanadium redox flow battery10.1016/j.electacta.2013.12.074 ↗
23. Low grade heat recovery for power generation through electrochemical route: Vanadium Redox Flow Battery, a case study10.1016/j.apsusc.2018.02.025 ↗
24. Nano-catalytic layer engraved carbon felt via copper oxide etching for vanadium redox flow batteries10.1016/j.carbon.2019.07.072 ↗
25. Laser-perforated carbon paper electrodes for improved mass-transport in high power density vanadium redox flow batteries10.1016/j.jpowsour.2014.03.007 ↗
26. Symmetry-breaking design of an organic iron complex catholyte for a long cyclability aqueous organic redox flow battery10.1038/s41560-021-00879-6 ↗
27. Effect of discharge cut off voltage on cycle life of MgNi-based electrode for rechargeable Ni-MH batteries10.1016/j.jallcom.2005.03.069 ↗
28. Data-driven electrode parameter identification for vanadium redox flow batteries through experimental and numerical methods10.1016/j.apenergy.2020.115530 ↗
29. Verified reduction of dimensionality for an all-vanadium redox flow battery model10.1016/j.jpowsour.2015.01.019 ↗
30. Open data, models, and codes for vanadium redox batch cell systems: A systems approach using zero-dimensional models
31. Open circuit voltage of vanadium redox flow batteries: Discrepancy between models and experiments10.1016/j.elecom.2011.01.020 ↗
32. Newman · 2012
33. Transport phenomena10.1115/1.1424298 ↗
34. Machine learning coupled multi-scale modeling for redox flow batteries
35. A limited memory algorithm for bound constrained optimization10.1137/0916069 ↗
36. Adam: A method for stochastic optimization
37. Understanding the difficulty of training deep feedforward neural networks
38. Kinetic study of electrochemical treatment of carbon fiber microelectrodes leading to in situ enhancement of vanadium flow battery efficiency10.1149/2.1091609jes ↗
39. Analytical modeling for redox flow battery design10.1016/j.jpowsour.2020.228817 ↗
40. Yang · 2021