Parallel Reduced Gradient Optimal Power Flow Solution

L WANG,N XIANG,S WANG,M HUANG
DOI: https://doi.org/10.1016/0378-7796(89)90025-4
IF: 3.818
1989-01-01
Electric Power Systems Research
Abstract:Two algorithms for the parallel solution of the reduced gradient optimal power flow by two-level computer networks are presented for the first time. The optimization method is the steepest descent method with an augmented Lagrangian function. In order to minimize the amount of information transmitted between the computers of the two-level computer networks, a new type of bus, a PQVθ separating bus, is introduced into the parallel reduced gradient optimal power flow calculation. Generalized power flow solution methods are used to solve the power flow problem. There are no approximations between the lower-level parallel computation and the higher-level coordination computation in the two parallel reduced gradient optimal power flow calculation algorithms. The proposed two algorithms have high parallelism and only a very small amount of data needs to be transmitted between the higher-level computer and the lower-level computers. Numerical results for a 30-bus two-area system are also presented.
What problem does this paper attempt to address?