Fuel cell faulty predicting system and its establishing method

The disclosure relates to a fuel cell faulty predicting system which including many sensors, a digital signal processor (DSP), a grey-predicting module, an extension neural network module and a displayer. The sensors are applied to sense many state signals of a fuel cell system. The DSP is applied t...

Full description

Saved in:
Bibliographic Details
Main Authors TSAI, HAN-HSUEH, WANG, MENG-HUI
Format Patent
LanguageChinese
English
Published 16.12.2011
Subjects
BASIC ELECTRIC ELEMENTS
ELECTRICITY
MEASURING
MEASURING ELECTRIC VARIABLES
MEASURING MAGNETIC VARIABLES
PHYSICS
PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSIONOF CHEMICAL INTO ELECTRICAL ENERGY
TESTING
Online AccessGet full text

Cover

More Information
Summary:The disclosure relates to a fuel cell faulty predicting system which including many sensors, a digital signal processor (DSP), a grey-predicting module, an extension neural network module and a displayer. The sensors are applied to sense many state signals of a fuel cell system. The DSP is applied to generate a set of characteristic values based on the state signals. The grey- predicting module is applied to generate a set of predicting values based on the characteristic values. The extension neural network module is trained by many matter-element modules. The extension neural network module generates a predicted faulty condition based on the characteristic values. The displayer is applied to show out the sensing data and the predicting faulty condition.
Bibliography:Application Number: TW20100118209