Failure diagnosis of water supply network by immune system

Recently, several immune network information models are proposed, and they are applied to process diagnosis as well as sensor fault diagnosis of a cement process. These models dynamically react to the on-line date from sensors. They can self-identify the unreliable sensor and unreliable constraint by moving from one equilibrium to another, reacting to the change of the relationship among data. In this paper, a simple network model for the water supply system of waterworks is diagnosed using such as immune network information model, and it is tried to find the cause of a sensor fault diagnosis and process faults diagnosis (water leak, etc.). Through some numerical examples the effectiveness and applicability of immune network information model is shown.