Task Availability Capability Assessment Based onCharacteristic Parameters

Abstract

To evaluate system task availability under the condition of real-time fault feature parameter input, a method of assessing task availability capability based on characteristic parameters was proposed. By introducing the neuro-fuzzy system(NFS), a vulnerability model of the aviation insurance system was established, and fuzzy rules were integrated into the framework of neural networks to establish an assessment model for system task availability capability. This method combined the reasoning ability of fuzzy logic and the infinite approximation function ability of neural networks to create an alternative model of the real system, which was more universal. Moreover, an intelligent optimization algorithm was used to make the alternative model approach to the real model, getting rid of the influence of unknown weight coefficients in the system that relied on experts or experience and endowing the fuzzy neural network(FNN) with learning capabilities. The experimental results and analysis show that the assessment model is comprehensive and reasonable and can be extended to the underwater field to assess the mission capabilities of navigation support systems and naval ship systems.

Keywords

• particle swarm optimization algorithm
• fuzzy neural network
• task availability capability assessment