Risk assessment in electrical power network planning based on sparse least squares support vector machines

To assess risk of electrical power network planning in an effective and fast way, the forecasting model of least square support vector machine (LS-SVM) based on pruning algorithm is established. Relative to the classical SVM, the least square SVM (LS-SVM) can transform a quadratic programming problem into a linear programming problem thus reducing the computational complexity. But sparseness is lost in the LS-SVM case. And the pruning algorithm make LSSVM recur sparseness. For illustration, a real-world planning project dataset is used to test the effectiveness of sparse least squares support vector machines(S-LS-SVM).