The accurate and efficient intelligent planning of disassembly sequences plays a crucial role in ensuring the high-quality recycling of end-of-life power batteries. However, the solution space obtained by the metaheuristic algorithm is often incomplete, resulting in suboptimal sequence accuracy. Additionally, the complex and dynamic disassembly information associated with end-of-life power batteries poses challenges in analysis and reuse, leading to low efficiency in disassembly sequence planning. To address these issues, we propose a novel approach for planning disassembly sequences based on the knowledge graph representation of power batteries. Firstly, we construct an updateable and scalable disassembly information model using knowledge graphs to capture the dynamic information and assembly relationships among battery parts. Subsequently, we utilize a combination of topological sorting and backtracking algorithms on the constructed disassembly information graph to derive the optimal disassembly sequence. Finally, we demonstrate the feasibility and effectiveness of our approach through an illustrative case study involving an end-of-life power battery pack.

准确高效的拆解序列智能规划对于保证报废动力电池的高质量回收发挥着至关重要的作用。然而，元启发式算法获得的解空间往往是不完整的，导致序列精度不理想。此外，与报废动力电池相关的复杂且动态的拆解信息给分析和再利用带来了挑战，导致拆解顺序规划效率低下。为了解决这些问题，我们提出了一种基于动力电池知识图表示的规划拆卸序列的新方法。首先，我们利用知识图谱构建可更新、可扩展的拆解信息模型，以捕获电池部件之间的动态信息和装配关系。随后，我们在构建的反汇编信息图上结合拓扑排序和回溯算法来导出最佳反汇编序列。最后，我们通过涉及报废动力电池组的说明性案例研究证明了我们方法的可行性和有效性。