Degradation tests represent a significant bottleneck in electrochemical technology development, occasionally requiring tens of thousands of hours. Thus, reliable degradation forecasting in a short time frame is a game-changer in accelerating the establishment of future electrochemical devices. Herein, we show a multidimensional kinetic model for electrocatalyst degradation by quantifying the relationship among potential, current, and time, applicable under various conditions. Aiming to predict reliable degradation behaviors in shorter experimental timeframes and inspired by modern weather forecasting methods, we integrated Bayesian data assimilation with our model to expedite multidimensional parameter optimization. Consequently, we achieved accurate and robust forecasting of electrocatalyst lifetime by employing oxygen evolution reaction as a representative system: it takes just 300 h to obtain the final lifetime of close to 1000 h even with environmental noise. This data-driven approach can accelerate our understanding of the microscopic electrochemical mechanisms and simultaneously directly bridge this understanding to develop next-generation energy technologies.

中文翻译：

---

通过贝叶斯数据同化对多维动力学模型进行准确和稳健的预测，加速电催化剂降解测试


降解测试是电化学技术发展中的一个重大瓶颈，有时需要数万小时。因此，在短时间内进行可靠的退化预测是加速建立未来电化学设备的游戏规则改变者。在此，我们通过量化在各种条件下适用的电位、电流和时间之间的关系，展示了电催化剂降解的多维动力学模型。为了在更短的实验时间框架内预测可靠的降解行为，并受到现代天气预报方法的启发，我们将贝叶斯数据同化与我们的模型相结合，以加快多维参数优化。因此，我们采用析氧反应作为代表性系统，实现了对电催化剂寿命的准确和稳健预测：即使有环境噪声，也只需 300 小时即可获得接近 1000 小时的最终寿命。这种数据驱动的方法可以加速我们对微观电化学机制的理解，同时直接弥合这种理解以开发下一代能源技术。