The invention discloses a harmonic emission level evaluation method of an improved least square support vector machine. The method comprises the following steps: step 1, constructing an LS-SVM regression model based on experimental data; 2, removing outliers from the LS-SVM regression model by using a DP clustering algorithm; 3, performing weighted calculation on the LS-SVM regression model without the outliers according to the experimental data; 4, introducing a combined kernel function into the weighted LS-SVM regression model for improvement; and 5, calculating the harmonic impedance of thesystem according to the improved LS-SVM regression model, and evaluating the harmonic emission level according to the harmonic impedance of the system. The method solves the problems that an abnormalsample in an LS-SVM regression model influences the harmonic impedance calculation precision and the LS-SVM regression model is poor in performance. According to the method, the LS-SVM regression model is improved, the combined kernel function algorithm of the binomial function and the radial basis function and the DP clustering algorithm are combined, abnormal samples in the LS-SVM regression model are removed, and the calculation precision of the improved LS-SVM regression model and the algorithm performance are improved.