Prediction of Electromagnetic Shielding Effectiveness of DCS Cabinet Based on LS-SVM

Due to the limitations of measuring the electromagnetic shielding effectiveness (SE) of the cabinet through sensors, a least square support vector machine (LS-SVM) model is introduced to predict the SE of the cabinet without sensor measurement points. The effectiveness of this method is verified using the HPR1000 safety-class DCS cabinet as an example. In engineering equipment, the factors that affect the SE value mainly include the size of the structure, materials, frequency, and testing points. Considering the actual situation of the DCS cabinet, the spatial coordinates XYZ and frequency of the test points are used as inputs to the model, and SE is used as output to establish an LS-SVM model. According to the actual application situation, experimental verification is carried out under two conditions: specific frequency and non-specific frequency. Through the established LS-SVM model, the predicted value of SE is obtained. From the experimental data and prediction results, it can be seen that the LS-SVM based method for predicting the SE of cabinets has good practical effects.