Font Size: a A A

Low Voltage Alternating Series Arc Fault Detection

Posted on:2018-01-17Degree:MasterType:Thesis
Country:ChinaCandidate:Y F XuFull Text:PDF
GTID:2322330515991066Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of the Energy Internet,the production and utilization of electric power are diversified,while the weaknesses of the existing protection system of the electric power system are highlighted increasingly.Alternating Series Arc Fault(ASAF),as one of the weaknesses,has not been solved reasonablely,and the safety accidents which caused by ASAF occur frequently.Detecting ASAF in real-time and accurately,and cuting off the fault circuit is an effective way to avoid arc burning sustainingly,and leading to the accidents such as fire.Construct an experiment platform independently,and acquire the research data,the Output Voltage Signal of the Current Sensor(OVSCS)in the experiment circuit,by the oscilloscope.The digitized and intelligentized ASAF detection algorithm is mainly composed of feature extraction and fault recognition algorithms.Extract the short time zero crossing rate feature of the OVSCS of ASAF by short time zero crossing analysis.To weaken the impact that the noise makes on the feature,and extract valuable feature as much as possible,the information entropy theory is applied to optimize the threshold value which is used to reduce the noise.To compensate the insufficient that the result of short time zero crossing analysis contains little amplitude information,a series of the threshold values are used to extract the features which constitute the OVSCS short time zero crossing rate matrix.Pattern matching and Hidden Markov Model(HMM)are used to recognize ASAF.To make it easy to compute and model,define the statuses of ASAF in three different time scales according to the observation results of the experiment,and the four processes,signal frame,pattern matching,HMM recognition and macro status discriminant,are excuted in ASAF recognition sequencely.With a low cost microcontroller as the core hardware and the detection algorithm as the basic software,design and realize the Arc Fault Detection Device(AFDD)and the test device for ASAF detection.The AFDD is small,low cost and easy to use with few additional conditions.The functions of the test device are adjustment arc electrode spacing in high precision,automatic repeating experiments and discriminant ASAF type.Design experiment to test the AFDD.The experiment results show that the AFDD detects ASAF in real-time and accurately in uninterrupted power supply working state,and deals with Accidental Alternating Series Arc Fault(AASAF),and is of high reliability that no misjudgement is made when there is a ’good arc’ occurring.
Keywords/Search Tags:Series arc fault, Zero crossing rate, Pattern matching, Hidden markov model
PDF Full Text Request
Related items