Harmonic pollution is one of the most important power quality issues in electric power systems. Correct location of the main harmonic disturbance source is a key step to solve the problem. This paper presents a method to detect the location of harmonic disturbance source in low voltage network through Bayesian inference. The harmonic state is estimated based on the measurement data from limited measurement points whereby the measurement error is also considered. The performance of the proposed method is assessed through a case study applied in a typical low voltage network. Monte Carlo simulation is used to obtain the statistical results. The influence of different parameters like disturbance level, measurement accuracy level, etc. is discussed. The proposed method shows the adequacy for the analysis of distribution networks.