冷水机组数字孪生模型及故障诊断研究

Research on Digital Twin Model and Fault Diagnosis of Chiller

本文采用了基于数字孪生的冷水机组故障诊断方法,研究了冷水机组制冷剂泄漏、冷凝器水流量减少、蒸发器水流量减少故障的诊断策略,分析了物理建模、经验建模和数据驱动建模的联合建模方式生成数据的可靠性和对故障诊断的提升效果,并通过实验验证了所提出方法的可靠性。结果表明:数字孪生方法可以有效解决故障诊断模型在新工况点下诊断能力下降的问题,利用数字孪生方法后系统仿真模型的所有热力参数最大相对误差均不超过4%,且故障诊断模型的准确率、查全率、平衡F分数分别提升9.7%、29.4%和16.6%。

In this paper,a fault diagnosis method of chiller based on digital twin is adopted to search the fault diagnosis strategy of refrigerant leakage,condenser water flow reduction,and evaporator water flow reduction faults in the chiller.The reliability of data generated by the joint modeling methods of physical modeling,empirical modeling and data-driven modeling and the improvement effect on fault diagnosis are analyzed,and the reliability of the proposed method is verified by experiments.The results show that:digital twin method can effectively solve the problem that the diagnostic ability of the FDD model decreases under the new operating point.After using the digital twin method,the maximum relative error of all thermal parameters of the system simulation model does not exceed 4%,and the accuracy,recall,and F1-score of the fault diagnosis model were increased by 9.7%,29.4%,and 16.6%,respectively.