Cooperative identification of advection–diffusion processes with spatially varying coefficients based on a multi-model structure

Abstract This work develops a practical real-time identification strategy for advection–diffusion processes with spatially varying coefficients. A novel multi-model structure that represents the processes in practical applications is proposed. The multi-model structure is parameterised as blended linear partial differential equation models. Both the offline centralised and the online distributed identification algorithms using mobile sensor networks are proposed. Trajectories for mobile sensor networks in the advection–diffusion field with spatially varying parameters are found, and distributed control laws to guide the mobile sensor networks to follow the trajectories while keeping desired formations are designed. The effectiveness of this proposed method is demonstrated in simulations.

摘要：这项工作针对具有空间变化系数的对流 - 扩散过程开发了一种实用的实时识别策略。提出了一种在实际应用中表示这些过程的新型多模型结构。该多模型结构被参数化为混合线性偏微分方程模型。提出了使用移动传感器网络的离线集中式和在线分布式识别算法。找到了在具有空间变化参数的对流 - 扩散场中移动传感器网络的轨迹，并设计了分布式控制律，以引导移动传感器网络在保持期望队形的同时遵循这些轨迹。通过仿真证明了所提方法的有效性。