Research On Power Quality Aviation Ground Power Supply

The power quality detection technology studied in this paper is the key technology of the power quality detection system for aviation ground power supply.This technology can effectively improve the accuracy and intelligence of the detection system.This paper studies the following contents:1.Accurate detection of power quality disturbance signal is realized.In this paper,the mathematical models of voltage sag,voltage swell,harmonic and transient pulse are constructed by analyzing the power supply value of aviation ground power supply.Then the modified S-transform algorithm is used to realize the accurate detection of power quality disturbance signal.The simulation results show that the modified S-transform can accurately detect the voltage amplitude,sudden change time and frequency of power quality disturbance signal.2.Accurate classification of power quality disturbance signals is realized.Three features of power quality disturbance signal are extracted by modified S-transform and classified by Support Vector Machine.The simulation results show that the method can classify and recognize all kinds of disturbances(7 kinds of disturbances and fundamental frequency signal)appearing in aviation ground power supply.3.Accurate diagnosis of voltage sag fault source is realized.The modified Stransform is used to extract three features of voltage sag fault source signals(3 kinds of fault sources and no fault signals)appearing in aviation ground power supply,and then the Random Forest classifier is used to classify them.Finally,the intelligent diagnosis function is realized.The simulation results show that the classification method can accurately diagnose the voltage sag fault sources.4.Engineering experiments have proved that the modified S-transform algorithm,Support Vector Machine classifier and Random Forest classifier were adopted in this paper,have good effect in the detection and classification of the output signals of aviation ground power supply.Modified S-transform algorithm can improve the detection accuracy of amplitude,time and frequency;Support Vector Machine classifier and Random Forest classifier can improve equipment performance and detection efficiency,and also reduce detection cost.In summary,this paper starting with the establishment of the mathematical model of power quality disturbance signal of aviation ground power supply,and uses modified S-transform algorithm as signal analysis tool,Support Vector Machine classifier as disturbance recognition tool,and Random Forest classifier as automatic diagnosis tool,and ultimately achieves the purpose of improving detection accuracy,signal classification and recognition and automatic diagnosis of fault sources.