Please use this identifier to cite or link to this item:
Title: Optimal Placement of Dispersed Generators for Practical Distribution Network
Authors: Kaur, Navdeep
Supervisor: Jain, Sanjay K.
Keywords: Dispersed Generators;Radial Distribution Network;Optimal Placement;Multiobjective Optimization;Probabilistic Allocation
Issue Date: 28-Aug-2019
Abstract: The usage of dispersed generators (DGs) are proving a promising and environmental friendly sources to supply electrical energy near to load centers. The interest in assimilation of DGs in distribution networks (DN) is because of increasing load demand and availability of renewable generation. However, the concept of integrating DGs to low voltage DN is a paradigm that differs from conventional central generation based paradigm. The integration of dispersed generation is directed to complement the efforts in improving the energy efficiency and reliability of power systems. With the usage of DGs, the increasing load demand can be met without expanding the existing transmission systems. The DNs are characterized by large number of buses and lines with a high R/X ratio and the power flow from the higher to lower voltage levels to supply loads. In radial distribution network (RDN), the voltage of buses drop when moved away from the source and results into high losses. Moreover, the DNs are mostly subjected to growing or varying load demand. The integration of DGs benefits the DNs by diminishing losses and enhancing the voltage of all buses while meeting the load demand. Such integration is beneficial only, if DGs of appropriate sizes are placed at suitable locations. The use of shunt capacitors in DNs is expected to reduce losses and yield lower reactive burden. The optimal placement of DG and capacitor simultaneously can be investigated to supply load demand. For the practical DNs, the load cannot be assumed simply as constant power (CP) load. The bus voltage has strong influence on load associated with that bus. The prevalent practical loads can be voltage dependent as found in commercial, residential and industrial loads. Further, the inherent uncertainty in practical DNs shall be analyzed by some probabilistic approach in modeling and analysis. In this research work, the effectiveness of DG integration is investigated for diminution of losses in DNs using analytical expressions based approach and Particle Swarm Optimization (PSO) for practical voltage-dependent loads. Based on the characteristics, four types of DGs are considered for integration. The analytical expressions based on equivalent current injection are derived by utilizing topological structure of RDN to obtain optimal size of DG for minimum loss. In the presented formulation, the optimal DG placement is obtained without repeatedly computing the load flow. The proposed formulation finds the optimal size of all types of DGs. The optimal size and site of DGs are also determined by PSO. The potential candidate buses for allocation of DGs are obtained through sensitivity factors. It limits the search space without compromising on the quality of the solution. The allocation obtained from analytical expressions and PSO are compared. The investigations are carried out on 33-bus and 69-bus RDNs. The Type-I and Type-III DGs are allocated optimally at multiple locations in DNs for optimization of a multi-objective function (MOF). The MOF is derived by combining indices of active power loss, reactive power loss, voltage deviation and overall economy with fuzzy decision approach. This MOF is optimized using PSO to find the optimal sizes and sites of DGs for voltage dependent practical loads. The objective function is optimized under operating constraints of bus voltage and MVA limit. Using base case power flow solution, the clusters of buses are formulated to limit the search space. The true Pareto based multi-objective optimization is also analyzed for VDI and PLI. Theses approaches are tested on a 69-bus RDN. The optimal allocation of DG and capacitor is determined through single and multi-objective optimization of power loss index and voltage deviation index. The placement of DG and capacitor is attempted separately and concurrently. The fuzzy decision approach is implemented to solve multi-objective optimization by assigning fuzzy memberships to the objective indices and optimizing the MOF. This formulation is implemented under PSO to attain the optimal sizes and locations of DG and capacitor for voltage-dependent practical loads. The multi-objective allocation of DG and capacitor is attained through true Pareto based approach. The proposed approaches are tested on 33-bus and 69-bus RDNs. The uncertainty in loads present in practical DN is expressed through probabilistic measures and analysis is carried out through Latin Hypercube Sampling and Monte Carlo Simulation techniques. The probable size of DG and capacitor are obtained through probabilistic approach. In this probabilistic approach, the size of DG and capacitor for a pre-defined optimal location is computed corresponding to each load sample obtained by two techniques. The optimization is achieved through PSO for minimizing power loss and voltage deviation indices individually. Moreover, these indices are combined to form a MOF, to be optimized by multi-objective PSO. This probabilistic approach is employed for single and multi-objective optimization problem on 33-bus RDN. The results obtained through two sampling techniques and results between the deterministic and probabilistic allocation of DG and capacitor are compared.
Appears in Collections:Doctoral Theses@EIED

Files in This Item:
File Description SizeFormat 
Thesis_950904019.pdf3.29 MBAdobe PDFThumbnail

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.