Planning of Unbalanced Radial Distribution Systems with Reactive and Distributed Energy Sources Using Evolutionary Computing Techniques

Abstract

The distribution system plays a key role in power system as it provides energy to the consumers safely, reliably, and economically. However, due to high R/X ratio, and low operating voltages, most of the losses occur in the distribution system. Moreover, distribution systems are generally unbalanced due to unequal single phase loads at the three phases of the system, and also additional unbalancing is introduced due to non-equilateral conductor spacing. This, causes the voltage, the current, and the power unbalance in the system. Further, the total neutral current of the system increases causing unwanted tripping of the relay. Hence, the service quality and the reliability of the distribution system reduces. Therefore, a suitable phase balancing strategy is required to mitigate the phase unbalancing in the unbalanced distribution systems. Also, apart from reducing the phase unbalancing in the unbalanced distribution systems, a suitable strategy is required to minimize the system power loss. In this regard, it is necessary for the distribution engineers to plan the unbalanced distribution systems in order to reduce the losses, voltage unbalances, and neutral current of the system for safe and reliable operation. Most of the approaches for the planning of the unbalanced distribution systems are based upon metaheuristic algorithms. Moreover, the recent research has focused only on either phase balancing or simultaneous phase balancing and conductor sizing optimization in unbalanced distribution systems using metaheuristic algorithms. However no work has been carried out to study the impact of the simultaneous optimization of the phase balancing, the conductor sizing, the capacitor location and sizing, the DG location and sizing, DSTATCOM location, and rating on system power loss, voltage unbalance, etc. utilizing these algorithms. As the metaheuristic algorithms are random in nature, the convergence is not guaranteed in a single simulation run. Hence, it is necessary to perform a statistical comparison among them in order to understand their relative merits and demerits for multiple simulation runs.

In this thesis, the impact of the simultaneous optimization of the phase balancing and the conductor sizing on the planning problems/objective functions of the unbalanced distribution system such as; the power loss, the voltage unbalance, the total neutral current, and the complex power unbalance studied using various metaheuristic algorithms such as the DE, the CSA, the PSO, and the GA. In the first step, these objective functions are optimized separately; then they are aggregated with weights into a multi-objective optimization problem. Further, a performance comparison in terms of the mean value of the objective functions and standard deviation (SD) carried out.

The reactive power compensating devices, such as the Capacitor, and the DSTATCOM has been integrated into the planning problem for the power loss minimization, the voltage profile improvement, and the voltage unbalance mitigation of the unbalanced distribution systems. Moreover, a three phase unbalanced modelling of the DSTATCOM has been developed. In this thesis, the effect of the simultaneous optimization of the phase balancing, the conductor sizing, the capacitor sizing, and the simultaneous optimization of the phase balancing, the conductor sizing, and the DSTATCOM sizing on the planning problem investigated. Both, single and multi-objective optimization approach are used in order to solve this problem. Also, statistical performance among the metaheuristic algorithms such as; the DE, the CSA, the PSO, and the GA in terms of the mean value of the objective function and SD carried out.

Further, the renewable sources such as the DG and a combined DG and DSTATCOM has been incorporated into the unbalanced system in order to study their impact on various planning problems.