Large scale hydrothermal optimal power flow problems based on interior point nonlinear programming

This paper presents an interior point algorithm for hydrothermal optimal power flow problems (HTOPF) which is derived from the perturbed KKT conditions of the primal problem. Moreover, the algorithm is extended successfully to solve approximate HTOPF problems (A-HTOPF) to find a suboptimal solution with much less execution time. For large scale systems, A-HTOPF can reduce CPU time by half and can guarantee more than 99% accuracy in most cases. By properly exchanging rows and columns of a correction equation, the reduced equation with novel 4T×4T block diagonal submatrices can be derived, where T is the number of time interval. Its topological structure is identical to that of the nodal admittance matrix, thus enabling an efficient algorithm. Numerical tests have been executed on six test systems of up to 1047 buses for HTOPF over 72 time intervals. The computational burden of the maximum test system is equal to that of 75384 (1047×72) bus OPF problem. The simulation results have verified that the proposed algorithm possesses good convergence property within reasonable execution time, and hence, the algorithm is quite promising for large scale applications