基于数字孪生的光电探测系统性能预测

Application of digital twin model in performance prediction of electro-optical detection system

为解决光电探测系统由于多场耦合引起的性能退化建模难题,提出以光电探测系统的能量域表述为基础,建立数字孪生模型以解决其性能退化预测问题的方法。在该数字孪生模型的建立过程中,采用调制传递函数构建光电探测系统的静态性能物理模型;进一步采用动态贝叶斯网络表示调制传递函数随时间的演变规律,实现对其动态性能退化过程的图模型描述;最后通过粒子滤波算法实现系统的状态监测和性能预测,从而完成光电探测系统数字孪生模型的建立,并给出了具有不确定性估计的仿真验证结果,验证了数字孪生模型能够较好的解决光电探测系统的性能预测问题。

To solve the problem of electro-optical system for its performance monitoring,a digital twin model based on Dynamic Bayesian Network(DBN)was proposed.In this model,a system-level performance indicator from the perspective of energy domain using Modulation Transfer Function(MTF)was developed,which avoided tedious modelling of performance interactions between the multiple subsystems of electro-optical system.DBN was constructed from the evolution of MTF to denote the dynamic performance degradation process and the propagation of epistemic uncertainty.To make the digital twin model capable of tracking and predicting the system performance states,Particle Filter(PF)was proposed as the inference algorithm for DBN.A real dataset collected in the laboratory environment was used to validate the feasibility of the digital twin model and verify the effectiveness of PF inference algorithm.The results showed that the proposed method was effective for joint estimation of states and parameters,and the prediction of electro-optical system on-line health-status was achieved.