Multiobjective capacity planning of photovoltaics in smart electrical energy networks: improved normal boundary intersection method

摘要

Incentive-based regulations, toward higher performance networks, are the main driver for minimising losses in distribution systems. On the other hand, more renewable generation is needed to achieve environmental targets. Hence, a multiobjective model is introduced in this study seeking to minimise energy losses as well as maximise renewable generation in radial distribution systems (RDSs). Two alternative control strategies of future smart grids such as reactive power management using adaptive power factor control and coordinated voltage control are considered in the optimisation problem. The problem is subjected to the various technical constraints such as voltage limits, thermal limits and reactive capability limits of photovoltaic (PV) penetration, power factor regulations and underload tap changer adjustment. Also, the uncertainties of load and renewable generation are considered, too. Then, the obtained non-linear programming problem is relaxed and reformulated as a well-suited and computationally efficient second-order cone programming problems. To obtain more efficient and evenly distributed Pareto set to help decision-making process, a modified normal boundary intersection method is introduced for solution methodology. The implementation of the proposed framework on IEEE 33-bus RDSs shows the gains that the flexibility provided by innovative control strategies can have on energy loss reduction and PV capacity.