基于支持向量机的化工过程故障诊断

SVM-based Fault Diagnosis for Chemical Process

摘要: 引入了基于统计学习理论的支持向量机技术，以连续搅拌釜式反应器——CSTR模型为例，研究了非线性化工复杂反应过程的故障诊断问题。实验结果表明，支持向量机方法与传统故障诊断方法相比，具有更好的精度、速度以及适应性。

Abstract: The technique of the support vector machines, which are based on the statistical learning theory, is introduced into the fault diagnosis for non-linear chemical process. A series of experimental (results) of the CSTR model show that SVM has a better diagnosis performance than those traditional chemical process fault detection and diagnosis methods such as BPNN, wavelet-net.