基于LE-ISTOA-SVM的聚合釜化工过程故障诊断

聚合釜是制备高分子化合物的最主要设备，其能否稳定的运行关系到产品的质量以及人员、设备的安全。但是，现代化工生产流程非常复杂，化工过程诊断数据具有高维非线性的特点。提出了基于LE-ISTOA-SVM的聚合釜化工过程故障诊断方法。首先，使用拉普拉斯特征映射算法（Laplace Feature Mapping Algorithm,LE）对故障数据进行降维。然后，使用改进乌燕鸥优化算法（Improved Sooty Tern Optimization Algorithm，ISTOA）优化SVM（Support Vector Machine，SVM）的参数来提高其性能。最后，利用聚合釜的实验数据做仿真测试。测试结果表明，该方法有较好的诊断效果。

Fault Diagnosis of Polymerization Kettle Chemical Process Based on LE-ISTOA-SVM

Polymerization kettle is the most important equipment for preparing polymer compounds. Its stable operation is related to the quality of products and the safety of personnel and equipment.However, the modern chemical production process is very complex, and the chemical process diagnosis data has the characteristics of high-dimensional nonlinearity.This paper proposes a fault diagnosis method based on LE-ISTOA-SVM.Firstly, the Laplace feature mapping algorithm is used to reduce the dimension of fault data.Then, the improved stooty tern optimization algorithm is used to optimize the parameters of SVM to improve its performance.Finally, the experimental data of the polymerizer is used for simulation test. The test results show that the diagnosis effect of this method is good.