Verification is a necessary part of Model-based systems engineering (MBSE) which is becoming a mainstream methodology for the design of complex systems. Verification in the early design stage has aroused widespread attention for its efficiency and cost-saving. Although there are numbers of researches on verification, some deficiencies still exist, such as the integration of design and verification needs to improve, the design problems are hard to trace and the behavior verification in the early design stage is often omitted. In this study, a novel ontology-based requirement verification method for complex systems is proposed to solve the above-mentioned problems. First, a requirement formalization method is proposed to avoid the ambiguousness of natural language, to make requirements easier to verify, and to make design problems easier to trace. Second, some transformation rules are defined to realize the automatic design ontology and rule generation. Based on these two steps, automated verification can be done through reasoning with ontology models and rules. This verification method is fully integrated with design tools and no additional expertise is needed for designers. To validate its feasibility and advantages, an example of a smart traffic light system is provided.

验证是基于模型的系统工程（MBSE）的必要组成部分，该模型正成为复杂系统设计的主流方法。设计初期的验证以其效率和节省成本引起了广泛关注。尽管关于验证的研究很多，但仍然存在一些不足，例如设计和验证的集成需要改进，设计问题难以追踪，早期设计阶段的行为验证常常被忽略。为了解决上述问题，提出了一种基于本体的复杂系统需求验证方法。首先，提出了一种需求形式化方法，以避免自然语言的含糊不清，使需求更易于验证，并使设计问题更易于追踪。其次，定义一些转换规则以实现自动设计本体和规则生成。基于这两个步骤，可以通过使用本体模型和规则进行推理来完成自动验证。此验证方法与设计工具完全集成在一起，并且设计师无需其他专业知识。为了验证其可行性和优势，提供了一个智能交通灯系统的示例。