As integrated energy systems become increasingly complex, their vulnerability to energy conversion units unavailability poses a considerable threat to system resilience and cost efficiency. This research is motivated by the need to address the growing threat of energy conversion unit unavailability and its significant impact on both system performance and operational costs. The development of a comprehensive vulnerability assessment framework, coupled with the integration of machine learning techniques, is hypothesized to significantly enhance the resilience and cost-efficiency of integrated energy management systems when faced with various unavailability scenarios. A novel methodology is introduced to enhance integrated energy management systems, focusing on the critical role of diverse energy conversion units. The methodology includes a new vulnerability index to quantify the impact of energy conversion unit unavailability on system performance, alongside an energy conversion unavailability attack model from the malicious actor perspective. Results indicate that under worst-case energy conversion unavailability attack scenarios, integrated energy systems operational costs can increase by up to 26.19%. The proposed solution, incorporating a Deep Q-Network into the integrated energy management system, achieves an 85.24% F1-score and a 96.68% reduction in additional costs associated with energy conversion unavailability attacks. This research demonstrates the varied impacts of different energy conversion units on integrated energy systems and proposes an enhanced integrated energy management system framework that improves system resilience and cost efficiency.

中文翻译：

---

以弹性为导向的综合能源管理系统方法：解决能源转换装置不可用和成本效益问题


随着综合能源系统变得越来越复杂，它们容易受到能源转换单元不可用的影响，对系统弹性和成本效益构成相当大的威胁。这项研究的动机是需要解决能源转换装置不可用的日益增长的威胁及其对系统性能和运营成本的重大影响。据推测，开发全面的脆弱性评估框架，再加上机器学习技术的集成，可以显著提高综合能源管理系统在面对各种不可用情况时的弹性和成本效益。引入了一种新的方法来增强综合能源管理系统，重点关注不同能源转换单元的关键作用。该方法包括一个新的漏洞指数，用于量化能源转换单元不可用对系统性能的影响，以及从恶意行为者角度出发的能源转换不可用攻击模型。结果表明，在最坏情况下的能源转换不可用攻击情景下，综合能源系统的运营成本最多可增加 26.19%。所提出的解决方案将深度 Q 网络整合到集成能源管理系统中，实现了 85.24% 的 F1 分数，并将与能源转换不可用攻击相关的额外成本降低了 96.68%。本研究展示了不同能源转换单元对综合能源系统的不同影响，并提出了一种增强的综合能源管理系统框架，以提高系统弹性和成本效率。