Fault occurred in transmission line can cause many problems such as failure of equipment, instability in power flow, and economical losses. Many of the techniques proposed for fault classification in transmission line have applied steady state component as it is easily affected by the surroundings. Then, protection scheme based on fault generated transient that can offer an accurate result for fault classification in power system should be proposed. This paper presents the fault classification scheme using Fuzzy-KNN (Fuzzy k-Nearest Neighbor) classifier and wavelet features. Two wavelet features were calculated in this study which are Wavelet Mean (μ) and Wavelet Standard Deviation (σ). Then, the Fuzzy-KNN classifier was tested with three datasets categories: Ideal, 30 dB noise, and 20 dB noise datasets. The overall results in accuracy performance show that the Fuzzy-KNN classifier performed better than the KNN (k-Nearest Neighbor) classifier.