Fault detection and isolation in flat navigation canals

Abstract

Abstract-Inland navigation networks are composed of several artificial canals that are characterized by no slope. These canals are particularly subject to resonance phenomena, which can create waves such that the navigation condition might not be guaranteed. It is therefore required to ensure dealing with freefault measured data and actuators. In this work, a fault detection and isolation method based on the Integrator Delay Zero model (IDZ) is designed for flat navigation canals. The proposed method is dedicated to the detection and isolation of sensor and actuator faults. Finally, it is tested by considering the Cuinchy-Fontinettes canal located in the north of France.

Modeling and fault diagnosis of flat inland navigation canals

“…Figure 7 shows the evolution of the two residuals: only r C (t) is activated when the fault is produced (at t = 250 min). Then, applying equation (19) and considering Tables 1-3, the Fontinettes sensor fault hypothesis provides a factortotal j equal to zero, and the Cuinchy sensor and actuator faults hypothesis provide a factortotal j equal to 3 and 1.5, respectively, and therefore, the most probable fault with a likelihood index P j (equation 22) equal to 6.67 is the Cuinchy sensor fault hypothesis. This hypothesis is confirmed after the waiting time T W = 7900 s, with no more activated signals.…”

“…Figure 8 shows the evolution of the two residuals: r F (t) is activated when the fault is produced (at t = 500 min). Then, applying equation (19) and considering Tables 1-3, the only fault that provides a factortotal j different from zero is the Fontinettes sensor fault and therefore the fault is correctly isolated.…”

“…CFr was also considered in order to develop a diagnosis approach for sensor and actuator faults. 19 The CFr has served in many works as the case study to illustrate the proposed approaches. Indeed, it constitutes an important reach in the inland navigation network in the north of France.…”

“…The first effort toward diagnosing sensor and actuator faults in the CFr was carried out: 19 some hints about the modeling step were given and the diagnosis approach was proposed, but it was only tested for a subset of all the possible faults that can impact the system. However, in this article the modeling methodology is detailed, and the expressions of the IDZ parameters are explicitly given by considering specific features of flat navigation reaches.…”

Digital Twin of Calais Canal with Model Predictive Controller: A Simulation on a Real Database

“…To do so, a set point known as the Normal Navigation Level (𝑁 𝑁 𝐿) is defined for each reach, together with a navigation rectangle defining an admissible water level interval. This rectangle has a High Navigation Level (𝐻𝑁 𝐿) and a Low Navigation Level (𝐿𝑁 𝐿): when the water level is outside of the navigation rectangle, the navigation must stop (Segovia et al, 2018). Another important objective regarding the management of large-scale waterways is minimizing operational costs.…”