Application Of Evidence Theory On Fault Line Detection For Small Current Grounding Power Systems

Fault line detection has become a troublesome problem when small current grounding power systems have the single-phase earth fault. so far , although many line detection methods have been proposed, any single line detection method is very difficult to adapt every kind of electrical network structure and the complex fault condition completely.Moreover, different time when fault appears, the faulted characteristic also performances differently in the degree of strength .If the signal in the process of the fault can be made full use of, then the condition of line detection will be improved greatly.In view of the questions above, A line detection method with multi-criteria and multi-period is proposed, which makes full use of fault information, and realizes the multi-criteria information fusion using the D-S evidence theory, therefore the performance of line detection can be improved greatly.It is usually abundant of transient components in single—phase earth fault. This paper proposes a fault detection method which uses wavelet transformation deriving from transient information. And based on this, a transient fault measure algorithm is established.This paper also gives a discussion on representation of fault detection with fault steady signal, and established fault measure algorithms for magnitudes and phase angles respectively.According to the detail of small current fault, the basic probability (bpa) function is designed, and then basic belief values of every line in each period can be obtained. The paper uses the distributional fusion algorithm and combines D-S combination rules, and makes final judgment of the line detection by the comprehensive belief values,Finally, a MATLAB simulation model of small current grounding power systems is established, and the simulations show that each kind of line detection criteria and the method which is proposed in this paper are correct and effective.