Voltage optimization and control on distribution networks is of uttermost importance to distribution system operators. The performance of the entire distribution network depends on the voltage profile of the system. The aim of this dissertation is to develop threefold advanced solutions that can effectively address and mitigate the voltage control and optimization issues of the smart grid. The first objective of this thesis is to develop accurate methods to quantify and analyze the expected variability individual and aggregated PV systems which significantly impact the voltage profile of grid-interfaced PV systems. We develop an improved statistical formulation, propose the use power spectral density and implement a new wavelet design using least squares optimization for PV aggregation and variability analysis. Using these proposed methods, the aggregation results show different percentages in reduction in the variability of the aggregated pairs of PV plants as well as the aggregation of the entire fleet of four PV sites. It also shows the impact of the sizes of the aggregated PV sites on each other. The results show that the relative sizes of the PV fleets have significant impacts on the reduction in intermittencies of the aggregated system by reducing the aggregated PV variability by as much as 57%. The correlations between these sites shows significant influence in the reduction in variability of the aggregated PV systems. Also the larger the aggregated system, the more the reduction in variability. Mathematical modeling-based and Heuristic-based voltage control and optimization algorithms with coordinated use of SIs, OLTC/VR, and capacitor banks for optimal power flow (OPF) is presented as a second objective in this thesis. These algorithms were validated using standard IEEE test feeders. The results of the proposed OPF algorithms show the effectiveness of optimally coordinating SIs with legacy voltage control devices with multi-mode, multi-droop settings SIs on the same feeder for effective voltage control. Based on the case study, the results show that prioritizing the use of SIs, can help optimize and efficiently use the existing legacy devices, reduce the number of tap changes by 28% and improve the capacitor bank utilization. by 17%. Finally, design optimization algorithms for solid-state transformers for distribution feeder voltage regulation and control are presented. The proposed algorithms for medium/high-frequency transformer optimization achieves efficiency above 98% with appropriate selection of the design variables with power density of around 4W/cm3 while the proposed simplified and cost-effective voltage regulation controllers were able to effectively regulate the output voltage around the desired value. Laboratory scale experimental setup is implemented to validate the proposed optimization algorithms and controllers with the experimental results in good agreement with the simulation values.
