基于可视化的固态电解质材料机器学习筛选与预测

【目的】随着“碳达峰”和“碳中和”目标的提出，能源消费领域电气化进程将进一步加快，其中在储能技术领域，锂电池是当前最具发展潜力的技术之一，已被广泛地应用在国民生活的方方面面。传统的锂电池所采用的液态电解质存在漏液、易燃和爆炸等多方面的潜在安全隐患，能量密度和安全性更高的固态电解质被认为是代替液态电解质的理想解决方案。当前，寻找具有高离子电导率等特性的固态电解质材料仍然是当前的研究热点。
【应用背景】传统的材料研究采用“试错”模式，基于已知经验与材料物理化学特性进行假设，然后进行实验验证，通过对上述过程的反复迭代，最终找到目标材料。上述过程耗时费力，限制了相关材料的研发进程。近年来，机器学习等方法被广泛引入并用于新材料的研究中，但却缺少辅助工具帮助材料领域专家分析和理解机器学习模型，并实现对满足特定性能需求的材料预测。
【方法】在这种背景下，我们基于可视化相关技术，建立了材料数据可视分析系统，期望促进材料科学家更高效地寻找高性能固体电解质材料。
【结果】我们基于可视化技术对多种机器学习算法的结果进行重构和展示，并通过不同视图对材料之间的关系进行可视化对比和分析，结合我们实验分析得到的一些案例，最终给出了预测。
【结论】最终，经过材料实验反馈，证实了部分预测材料的优良性能，验证了该系统的有效性。

[Objective] It is a hot research topic to find the ideal solid electrolyte material with high ion conductivity, and replace the liquid electrolyte which has safety concerns as the electrolyte material of lithium batteries.
[Context] In recent years, methods such as machine learning have been widely used in the prediction of new materials. However, there are few aids to help materials experts analyze and understand machine learning models to predict the composition of materials that meet performance requirements.
[Methods] Under such background, we built a visual analysis system based on visualization technology, trying to help experts in the field of materials analyze the results of machine learning, predict and look for high-performance solid electrolyte materials.
[Results] We compare the results of several machine learning algorithms and use visualization techniques to display the results. We visually analyze the relationship between materials through different views and finally give the prediction based on some cases we summarized.
[Conclusions] Many material experiments have verified the excellent properties of some predicted materials and have confirmed the effectiveness of our system.