The invention relates to a vertical well water-content measurement method based on conductivity probe array information fusion, and belongs to the field of multiphase flow detection. The method comprises the following steps of: firstly, measuring the voltage response signal of each probe of a conductivity probe array; secondly, extracting the characteristic quantity from the voltage response signal of each probe through statistic analysis and wavelet analysis; thirdly, performing Z-score normalization, and then extracting principal components by a PCA (Principal Component Analysis) technology, wherein the extracted principal components are called as the PCA characteristic quantity; fourthly, respectively building a regression model, called as an SVR (Support Vector Regression) model, from the PCA characteristic quantity of the response signal of each probe to the oil-water two-phase flow water content by an SVR method; fifthly, optimizing the parameters of the SVR model by a particle swarm optimization algorithm; and finally, performing decision-making level information fusion of linear mean square estimation based on the minimum mean square error on the water content predicted by each probe. Compared with a vertical well water-content measurement method based on a single probe, the vertical well water-content measurement method based on conductivity probe array information fusion has the advantages that the robustness and the reliability of well logging are improved, and the measurement precision is improved.