Power network-planning optimization considering average power not supplied reliability index: Modified by chance-constrained programming

Abstract

In this paper, a mixed-integer linear formulation for optimization multi-stage power system is represented. In order to achieve the best results, all facilities should be considered simultaneously due to the interconnection between power network equipment. In this paper, power plant, HV and MV substations and Distributed Generation (DG) are regarded concurrently to create an integrated power system planning. Furthermore, a reliability index called Average Power Not Supplied (APNS) in load point is modelled. Stochastic programming in the Chance-Constrained approach is also utilized to overcome uncertain parameters containing demand in load points, nominal voltages and facility failure rates. The mathematical model is tested on a real life case study in four schemes containing with or without considering DG and in certain or uncertain environment. Using both DG and MV substation in schemes planning leads to the establishment of a parallel system to improve network reliability. Since each scheme has specific requirements, it’s mathematical formulation is discussed in detail. The proposed formulation determines the optimum system configuration, including feeder routing, siting and sizing for power plant, HV/MV substation and DG. The numerical results show that planning in uncertain environment has a higher cost in comparison with certain model, but the results are more close to real world operations. Based on the obtained results a planner can estimate the cost components in different environments. Also in uncertain schemes, the growth in confidence levels increases the grid total cost. Therefore, planners must create an appropriate balance between confidence level and system total cost. Numerical results suggest that using DGs leads to a decrease in the required substations, their power to service load points and APNS cost. Thus using DGs can be an economic alternative to decrease system total cost. The achieved results indicate that the proposed planning for designing a power network has a desirable performance in practice.