A Network Programming Approach for Investment Planning in Large Power Systems

Abstract

Electric power plays a crucial role in the development of an economy. Availability of adequate electric power has become essential not only in industrial and urban areas but the rural society is also getting dependent on the use of electricity for agricultural and other needs: Due to the high rate of demand growth for electricity and prohibitive costs of shortages, investment planning in electric power generation and transmission systems has received increasing attention during the last two decades. Whereas uncertainties regarding future demand growth and its temporal variation, technological innovations in the power generation and transmission field and future fuel prices make long term investment planning a challenging task, the gains from planning investments within the framework of a least-cost investment planning model can be quite considerable both in terms of economy and reliability of future power supply.The purpose of this paper is to describe a Network Programming Model for least-cost investment in electric power generation and transmission system and illustrate it by a case study involving application of this model to northern region of India for planning the elecbic power system to meet projected peak demand in the target year. The model described. in this paper is essentially an optimization model based on the power system network in which electric power frows from generating nodes to load centres through existing or proposed transmission lines and costs of the generation of power and transmission including power and energy losses are accounted for. The model with the aid of a network computer programme obtains a least-cost flow pattern in the network which specifies .the optimal selection of generating plans and transmission lines and the pattern of power flow. The optimal solution Should be tested with load flow studies to determine system reliability under various contingencies.During the study of the northern region of India, the Network model was used to determine the optimal location of a nuclear plant from four alternative locations. It also provided the optimal system expansion plan by choosing the leastcost additions of thermal and hydro plants to the system and suggested additional transmiss10n links~ The network model uses a simple, but integrated representation of the generation and transmission system and its spatial nature. It would be possible to include a more sophisticated representation of the power system bvt at additional computational cost. Currently the network algorithm is highly efficient and it is possible using an iterative approach to handle nonlinear cost functions associated with capital costs and transmission losses in the network.The Network model as formulated represents the power generation and transmission system operation during the peak period, hut it is possible to study the system operation during other periods or assuming certain maintenance schedule of power plants. The situation of a dry year could also be simulated by corresponding deration of hydro plant capabilities. The model formulation is simple but it could become a very useful tool for the power system planner who is interested in analysing the large number of alternative location and network expansion possibilities in a large power system and obtain economically optimal system configuration.