The application of genetic algorithm in diagnostics of metal-oxide surge arrester

Experimental verification of monitoring techniques for metal‐oxide surge arrester

“…This section briefly presents the methodology proposed in [22,29]. The optimisation task of this method is to minimise the sum of squared deviations between waveforms of measured leakage current and estimated leakage current…”

“…In (22), OF is the objective function, T is the signal period, i tm is the measured value of total leakage current, i ts is the simulated value of total leakage current, and x is the vector of variables.…”

“…More details about the method can be found in [22,29]. It has been shown that acceptable results of GA are achieved after only 100 generations, running with a population of 500 chromosomes, mutation rate of 3% and crossover rate of 80%, using the roulette wheel as a selection method.…”

“…2 presents the flowchart of the algorithm that uses a GA to find the optimal solution. [22,29] In For the base case, to which the other cases will later be compared, continuous operating voltage U c , without added voltage harmonics (case 1) at the ambient temperature of 25°C is selected. For all measurements, the same experimental setting is used (Fig.…”

Extracting the Resistive Current Component from a Surge Arrester’s Leakage Current without Voltage Reference

“…The obtained curves that are shown in Figure 5 can be approximated by sectional linear function. Because I t,mn (ϕ) is the consequence ofÎ r (ϕ), the intersection of the linear function appears at the same argument, when ϕ = ϕ i (calculating intersection of (15) and (17) gives ϕ i = 164.2 • , I t,mn,i = 363 µA). Hence, the trend lines, represented by the linear function, are:…”