Methodology for planning tie lines between interconnected power systems

Abstract

Power exchanges between utilities improve reliability and increase economical savings by accessing cheaper generation resources, taking advantage of peak diversity and time zone difference between interconnected systems and time difference of forced outages of generating units of two systems, reducing the overall spinning reserve and minimizing interruption cost. As such, interconnection between utilities has become a part of power system planning. Interconnection planning requires the evaluation of location and capacity of tie lines that ensures optimum operating cost and optimum rate of return on investment. The problem is combinatorial i.e. it is a problem of selecting a solution from among a finite set of possible solutions, which precludes closed form solution. Current practices in interconnection planning use transmission expansion planning techniques that do not meet these requirements. A methodology, first of its kind, is introduced for tie line planning between interconnected systems. The methodology is a heuristic one, using a full nonlinear power system model. A cost function is used to evaluate the performance, during the planning period of different tie line candidates. Instead of a single peak load, correlated demand is used as the load model. This decomposes the planning problem into a number of smaller optimization problems that are solved using a full ac optimal power flow (OPF) technique. At first, at every load level, the OPF is used to search for local minima and to evaluate the optimized capacity and location for a tie line. The cost function includes both the tie line cost and the energy production cost for a particular load level. The energy production cost for the whole planning period is evaluated in the next stage considering the evaluated optimum tie -line obtained for a -particular load level. Appropriate penalty function is introduced to evaluate penalty cost, which calculates the cost that would be incurred if the tie capacity determined at a load level is over or under sized with respect to the optimum tie capacity obtained for other load levels. This penalty cost is also added to the total cost. Proceeding in the same way the total cost for the whole planning horizon for all optimum tie capacities for all load levels are evaluated. Comparing these costs the tie set with the minimum cost is selected. The proposed methodology is applied to the IEEE Two Area Reliability Test System 1996 and Bangladesh Power System. The results obtained using the proposed methodology is compared with those obtained using the conventional transmission line planning approach. The results obtained using the proposed methodology with multilevel load model is also compared with those obtained using single peak load model.