Improving Loss Minimization in 33kv Power Distribution Network Using Optimized Genetic Algorithm

: The epileptic power supply from the national grid due to instability is a concern to energy consumer. This instability in power supply experienced in power distribution network could be minimized by introducing Optimized Genetic Algorithm (OGA). It is achieved by characterizing 33KV distribution network, running the load flow of the characterized 33KV distribution network, determining the distribution losses from the load flow. Minimizing the determined losses in 33kv distribution network using (OGA), and designing SIMULINK model for improving loss minimization in 33kv power distribution network using OGA. Finally, validating and justifying the percentage of loss reduction in improving loss minimization in 33kv power distribution network without and with OGA. The results obtained are conventional percentage power loss in 33KV distribution network, 75%, while that when OGA is incorporated in the system is 72.9%. With these results obtained, the percentage improvement in loss reduction in 33KV distribution network when OGA is used is 2.1%. The conventional percentage of power loss in 33KV distribution network is 80%. The percentage power loss in the distribution network now is 72.9%; hence, power loss reduction in distribution network. Unmitigated power loss was 76.7% when OGA is introduced we had 74.63%. The percentage power loss in distribution network in bus 8 is 81.7% while that when OGA is applied is 79.49%. The percentage power loss in bus 9 of 33KV distribution network is 86.7%. Finally, when optimized genetic algorithm is incorporated in the system the percentage power loss in the network was reduced to 84.36%.


Introduction
Electricity consumers are increasing their demand for quality power supply more than what we had three years ago. It requires a modern technique to contain the situation. The growth of electricity demand is increasing rapidly which will require techniques or methods to enhance loss reduction in the distribution network. Many authors have proposed many types of ways to achieve a considerable reduction in power losses causing power outages. A closer review of known methods will be considered in the subheading below to see which of the techniques could reduce system energy loss and alleviates distribution congestion, as well as improving voltage profile a good method should be able to enhance reliability and provides lower operating cost. Distribution means the electric power from transmission being distributed to the final consumers in a safe and reliable manner.

Aim of the Study
This paper is aimed at using Optimized Genetic Algorithm (OGA) to improve loss minimization in 33kV Power Distribution Network in southern Nigeria.

Objectives
Frequent tripping of feeders and protective devices resulting in power failure as well power losses from copper conductors had become an endemic problem, therefore, the objective of this research work was to i. Collect data from the characterized 33Kv line from Abakaliki to Ugep. ii. Use the line parameters to run the load flow of the characterized 33kV distribution Network in order determine the distribution losses. iii. Minimize the determined losses in 33kV short transmission line (50kM). iv. Design a SIMULINK model for improving loss minimization in 33kV power distribution network using Optimized Genetic Algorithm (OGA). v. Validate in order to justify the percentage of loss reduction in improving loss minimization 33kV power distribution network without and with Optimized Genetic Algorithm (OGA)

Extent of Past Related Works
In a distribution network there is usually negative impact when Distributed generators tied to the system are wrongly placed. They can cause system surge due to reflection of mismatched impedance (Characteristic impedance). The resultant effect could lead to low power quality and reliability levels, the voltage regulation would not be effective as well [1,2]. The sizing and placement of DG's are determined from the optimization algorithms, which are read an analysed in the literature reviewed from different angles as in the study of [3,4] where the Kalman filter algorithm was used to reduce power losses. The authors in previous study [5] proposed a probabilistic power flow technique with embedded Genetic Algorithm to determine the problem of the total costs reduction. In that technique, the costs that has to do with the installation of DGs in a distribution system are made up of operating cost, running and investment costs. The method of Optimized Genetic Algorithm has been widely adopted to reduce transmission losses as it is simple and user friendly [6]. Local particle swarm optimization (PSO) is another useful technique but lacks the numerical accuracy since it is a population-based stochastic optimization technique [7]. We also have optimization algorithms for multi-objective as a technique which was proposed as well as pareto-front [8]. In [9], the authors talked about the use of an advanced Pareto-front non-dominated sorting multi-objective particle swarm optimization technique. The optimization problem considered two multi-objective functions; the constraints were the power loss reduction, voltage stability improvements with voltage profile and the power balance. The multi-objective performance index was discussed [10] to improve the voltage profile and minimise the system losses. In this work [11], the criss cross optimal algorithm and Monte Carlo simulation was equally presented as the desired tool to reduce the total costs and power losses. While the work by [12], the authors proposed an improved particle swarm optimization (IPSO) algorithm for reducing, electricity price, running cost and network loss. The authors [13] recommended improved analytical method to determine the optimal size of about four different DG typologies by utilizing an effective methodology. While in the study, [14], the optimum sizes and locations of DG units were determined selecting the power losses and voltage profile as objective functions. Shunt capacitors (SCs) are not exceptions as they could be used to minimize the variability of some DGs from a reliable energy source. Adoption of Genetic Algorithm (GA) was used in the placement and sizing to improve the bus voltage magnitude and minimize power losses as was described and implemented in the ETAP software [15]. In the review [16], a Sequential quadratic programming-based algorithm combined with Genetic Algorithm was proposed for mitigating the costs, power losses and network upgrading. A load concentration factorbased on analytical technique was presented to determine the optimal solution for power loss reduction and improvement in the voltage profile [17]. In this work, a relevant technique for reducing transmission losses in a 33kV distribution network explained in details the feasibility of the method. This proposed technique/method is developed on the Optimized Genetic Algorithm (OGA) to solve the endemic problem. The main original contributions of this work depended on the ability of the technique to deal with DGs to regulate the reactive power introduced in the network by solving the Optimal power factor (PF) of DGs while taking into consideration the contribution of the SCs and other limitations on the initial maximum installed capacity of the Synchronous Generators.

Methodology
The collection of required data for analysis and determination of losses was the first step. It is done in this manner, characterizing 33KV distribution network, running the load flow of the characterized 33KV distribution network, determining the distribution losses from the load flow. Minimizing the determined losses in 33kv distribution network using optimized genetic algorithms, and designing SIMULINK model for improving loss minimization in 33kv power distribution network using optimized genetic algorithm. Finally, validating and justifying the percentage of loss reduction in improving loss minimization in 33kv power distribution network without and with optimized genetic algorithm.

Power Flow Solution by Newton-Raphson Method Maximum Power Mismatch=2.83789e-007 No. of Iterations=10
The total power is 600MW. On the other hand the faulty buses are bus 3, 5, 6 8 and 9 because their P.u.volts did not fall within 0.95 to 1.05p.u. Volt. Their respective power losses are 150MW, 120MW, 140MW, 110MW and 80MW. Where X1 is no of loss buses X2 is p.u.volts that atributes to the power losses in the distribution network.

To determine the Distribution Losses from the Load Flow
X3 is distributed power losses (mw) in these buses P is percentage of loss in the distribution network    Figure 1 is the Load flow analysis of the 33kV distribution network under consideration while Figure 2 shows the step by step technique of using Optimized Genetic Algorithm. Figure 3 depicts the designed SIMULINK model for improving loss minimization in 33kv power distribution network using optimized Genetic algorithm. Figure 4 is a comparison of percentage power loss in bus 3 of 33KV distribution network with and without Optimized genetic algorithm, Figure 5, Compares percentage power loss in bus 5 of 33KV distribution network with and without Optimized genetic algorithm Table 2 shows determined distribution losses.

Results and Discussion
The results obtained at different faulty buses in the distribution network shows that there is reduction in percentage of power losses in distribution network as detailed in figures 4 and Figure 5 respectively.
In figure 4, the Percentage power loss in bus 3 of 33kV distribution network with and without Optimized Genetic Algorithm was compared, and the result presented here showed that the conventional percentage power loss in 33KV distribution network is 75% while that when optimized genetic algorithm is incorporated in the system is 72.9%. With these results obtained, the percentage improvement in loss reduction in 33KV distribution network when optimized genetic algorithm is imbibed in the system is 2.1%. Figure 5 shows the comparison between percentage power loss in bus 5 of 33KV distribution network with and without Optimized genetic algorithm; the result presented revealed that the conventional percentage of power loss in 33KV distribution network is 80% while the percentage power loss in the distribution network when Optimized genetic algorithm is incorporated in the system is 72.9%. This shows that there is power loss reduction in distribution network when optimized genetic algorithm is introduced in the system.

Conclusion and Recommendation
The intermittent power supply in our distribution network has liquidated some establishment that solely depend on power to run their daily work. This is due to power loss in the distribution network. This irregular power supply in the distribution network is overcome by improving loss minimization in 33kv power distribution network using optimized genetic algorithm. It is done in this manner, characterizing 33KV distribution network, running the load flow of the characterized 33KV distribution network, determining the distribution losses from the load flow.
Minimizing the determined losses in 33kv distribution network using optimized genetic algorithms, and designing SIMULINK model for improving loss minimization in 33kv power distribution network using optimized genetic algorithm. Finally, validating and justifying the percentage of loss reduction in improving loss minimization in 33kv power distribution network without and with optimized genetic algorithm. The results obtained are conventional percentage power loss in 33KV distribution network is 75% while that when optimized genetic algorithm is incorporated in the system is 72.9%. With these results obtained the percentage improvement in loss reduction in 33KV distribution network when optimized genetic algorithm is imbibed in the system is 2.1%. The conventional percentage of power loss in 33KV distribution network is 80%. On the other hand, the percentage power loss in the distribution network when Optimized genetic algorithm is incorporated in the system is 72.9%. This shows that there is power loss reduction in distribution network when optimized genetic algorithm is incorporated in the system. The conventional power loss in distribution network is 76.7% while that when optimized genetic algorithm is inculcated in the system is74.63%. The conventional percentage power loss in distribution network in bus 8 is 81.7% while that when optimized genetic algorithm is imbibed in the system is 79.49%. The conventional percentage power loss in bus 9 of 33KV distribution network is 86.7%. Finally, when optimized genetic algorithm is incorporated in the system the percentage power loss in the distribution network reduced drastically to 84.36%. With these results, it shows that the improvement in minimizing power loss in 33KV distribution network is 2.34%.

Recommendations
To ensure optimum performance reliability of electricity supply in 33kV power distribution, the following recommendations are suggested based on the findings: 1. Losses could be minimized using Sychronous phase modifiers. 2. Capacitor banks should be placed in paralle to load centers to improve power factor. 3. Solid State var compensators should be encouraged in the distribution substations. 4. Preventive maintenance should be implemented quarterly to improve the integrity of power system components.

.4 Corona technical losses and non-technical losses
could be minimize with timely replacement of dilapidated and old power system equipment. The Government should make provision for training technical personnel in the industry.