The Research On The Fault Diagnosis Of Condenser And Development Of Application Software

In the turbine of condenser, condensing system as a cooler is an important auxiliary installation. Its thermodynamic characteristic has a heavy impaction on the security and economy of a running turbine .It is important that deeply studies the working mechanism and operating characteristic, and establish the monitoring and optimization model for fault diagnosis and improving the operation level.In this paper, condensing system's performance is discussed, and condensing system's common faults are analyzed and the cause and signs of condenser's vacuum falling are emphasized.The judgment of condenser's faults is not easy for it has many kinds of reasons and its signs are fuzzy to some extent. In this paper, fuzzy neural network, which is fairly advanced, is applied in condenser's fault diagnosis. After introducing fuzzy logic theory and neural networks basis, their complex—fuzzy neural networks is drawn forth. Fuzzy neural network(FNN) merits of fuzzy logic and NN, it can described fuzzy question, has great capabilities of studying and data directly disposing, has great capabilities of expression about knowledge structure, has great capabilities of fault tolerant. In a word, FNN can automatically attain to knowledge, adapt to the change of environment, and adopt collateral calculation of numerical value, so that it is an effective method which solves localization of traditional fault diagnosis.With MATLAB, application software for the condensing system fault diagnosis was constructed based on fuzzy back-propagation neural network. The software is easy for operators to operate and has a friendly interface.