A full-scale research project began in the year 2002 with Snohomish Public Utility (SPU) to develop efficient optimization methods for the location of electrical substations on rectangular grids, with the objective of minimizing installation and operation cost. Previous work on optimal power flow (OPF) under contingencies showed that any change to the circuit such as moving a substation from one node to another required a different set of powerflow equations. It is required to move all substations simultaneously to get better configuration while designing an algorithm. A compact representation of the problem resulted in the requirement to solve a mixed-integer quadratic program (MIQP). The impact on the reasonable integer solution from starting at very very different initial configuration has been analyzed by showing the performance on a realistic load distribution given by the SPU on a 24 × 46 grid.
All Science Journal Classification (ASJC) codes
- Control and Optimization
- Computational Mathematics
- Applied Mathematics