Effect of distributed generation installation on power loss using genetic algorithm method

Injection of the generator distributed in the distribution network can affect the power system significantly. The effect that occurs depends on the allocation of DG on each part of the distribution network. Implementation of this approach has been made to the IEEE 30 bus standard and shows the optimum location and size of the DG which shows a decrease in power losses in the system. This paper aims to show the impact of distributed generation on the distribution system losses. The main purpose of installing DG on a distribution system is to reduce power losses on the power system.Some problems in power systems that can be solved with the installation of DG, one of which will be explored in the use of DG in this study is to reduce the power loss in the transmission line. Simulation results from case studies on the IEEE 30 bus standard system show that the system power loss decreased from 5.7781 MW to 1,5757 MW or just 27,27%. The simulated DG is injected to the bus with the lowest voltage drop on the bus number 8.


Introduction
The need for electric power is now increasing due to the increasing demand for load due to the rapid economic growth. This economic growth is driving the acceleration of the industrial world to use equipment that requires more electrical energy. Therefore, the supply and quality of electrical energy needs to be improved.Electrical energy supply that there is generally a conventional power plant. Conventional power plants are generally designed on a large scale, centralized, and built away from load centers requiring transmission and distribution networks to supply electricity. Conventional electric power system consists of three parts, ie generation, transmission and distribution associated with the load.
Long transmission and distribution networks from power plants to load centers result in greater power losses, so technical efforts are required to reduce such losses such as network shortening, reconductor, insertion of the substation, capacitor installation, automatic voltage regulator (AVR ), Replacement of connectors and so on.
In recent decades, distributed generation began to be developed in the developed countries to support the electricity needs of the country. Distributed generation is considered as an appropriate solution to overcome the shortage of energy supply and to overcome the problem of power distribution systems such as power loss, the balance of the system, and also address the critical load that is experiencing drop voltage.
This scattered plant is a small-scale plant connected to a local distribution system often called "Distributed Generation" (DG). DG characteristic is small scale usually between 50 kW to 400 MW, distributed and close to load center (closed to load), interconnection In the power flow study, many methods can be used, but the common and widely used method is Newton Raphson.

Newton Raphson method in Power Flow Calculation
The Newton Raphson method is used to solve nonlinear algebraic equations simultaneously from some unknown variables with a linear approximation. Generally formulated with: In equation above bus 1 as reference bus. The Jacobian matrix provides a linear comparison between the changes in the voltage angle ∆หܸ () ห and the voltage magnitude (∆ܲ () ) with changes in active power (∆ܲ () ). and reactive power (∆ܳ () ). In the simplest form of equation (4) can be written as follows:
The value of ∆ܲ () and ∆ܳ () is the difference between the scheduled value and the calculated value of power residuals: Correction bus voltage magnitude and angle with the general iteration (i + 1) is: This process stops until convergence:

Genetic Algorithm
In a genetic algorithm, a set of parameters (an individual) or in biology is called a chromosome for a problem in this case is an object function, formed or encoded in binary form. And in each generation, a number of individuals (population) are evaluated in parallel to their matching, as the price of the object to be minimized. New and improved populations are generated from the old through the application of genetic operators such as selection, crossover, and mutation. Selection is a simple operator to get the new chromosome from the most powerful. Strings that have a stronger fitness value have a greater chance of being selected and can follow the operations of other operators.
Crossovers randomly select a pair of parent and form two offspring through the corresponding segment exchange from the parent. Implementation of crossover is done on the genes of the two parent chromosomes which result in the genes from the two different mains combined into their new offspring.
Mutation is a random change from the position of the string. In a binary string presentation, a simple change means 0 to 1 and vice versa.
Operations are repeated until the specified number of generations has been reached. Genetic structure of the algorithm in out line as shown in Figures 1-3.

Research Steps
The steps in this research were conducted in two steps, first to determine the location of DG placement and the second is to determine the optimum DG capacity. 1. The first step calculating the power flow to the system tested using Newton Raphson method to get the condition of the voltage profile on the system bus. 2. From this first step can be determined the location where DG will be paired on the system with the indicator refers to the voltage profile on the bus that decreased below 0.9500 pu. From the second step of this optimization calculation can already determine the capacity of DG to be installed on the distribution system.     Figure 4 shows graphically the voltage profile that occurred after the power flow analysis before the optimization, it is seen that the bus number 8 voltage profile below 0.9500 pu. It shows that under these conditions it is necessary to maintain the voltage profile of the bus going back well above 0.9500 pu. So that on bus 8 which is a load bus that has the lowest voltage magnitude is selected as the main location of DG installation. It is also possible to install DG on another bus which also has low voltage magnitude, but still on bus 8 also installed DG. Power losses that occur before optimization can be shown in Table 2. Where it can be seen that the lowest loss of active power occurs in channel number 20 ie from bus 14 to bus 15 of 0.0057 MW. While the highest occurred in channel number 6 that is from bus 2 to bus 6 of 0.8029 MW. Total active power loss is 5.7781 MW.

Determination of Optimal Capacity DG
For this optimization step the maximum capacity referred is from DG type, ie from the intermediate type whose maximum capacity is 50 MW. Referring to Table 3 it can be seen that by injecting DG on three buses are buses 8, 7, and 19, indicates that DG increases the system voltage profile. We can see in table 3 that there is no voltage profile on any bus that is below the allowed standard limit of 0.9500 pu. We can also note on bus number 8 which previously experienced a decrease of 0.9424 pu, after the optimization using genetic algorithm to be 1.0029pu. All voltage profiles are within the permitted limits of between 0.9500 pu to 1.0500 p.u.

1234567890''""
10th International Conference Numerical Analysis in Engineering IOP Publishing IOP Conf. Series: Materials Science and Engineering 308 (2018) 012034 doi:10.1088/1757-899X/308/1/012034 Figure 5. Graph of the voltage profile after optimization In Figure 5 we can see that after the treatment with the installation of DG, seen voltage profile on bus number 8 is to rise above 1.0029. This shows that by adding DG to the voltage dropped bus and several other buses, this can improve the voltage profile of the bus, as well as fix some of the buses that have decreased as well. From the results of this power flow analysis also found that the total generation, loading and loss of power (losses) on the network shown in Table 4 The total amount of power generation is Pgen = 166.6938 MW and Qgen = 33.0763 MVAR. Total loading for is Pload = 165,1181 MW and Qload = 63,5628 MVAR. While the total loss of power (losses) we need to see is the loss of active power only that is equal to Ploss = 1.5757 MW (27.27%). And it is seen that there is a decrease in power losses when compared with before the optimization of 5.7781 MW.

Conclusions
The results of the power flow analysis show that indicating the existence of a critical bus that is bus number 8. This indicates that the plant is in the system can not serve the load burden. So it should be understood that there should be an additional generator with DG installation. Installation of DG from the simulation result is bus 8. The improvement of voltage profile obtained is that all buses are within the permissible limits and nothing is below or above the standard value. From the installation of DG can also be known that the power losses on the channel on the system to decrease to 1.5757 MW or only 27.27% of the previous value of 5.7781 MW.

Suggestion
The suggestions that can be put forward for future research, use and planning are as follows: 1. This research may be used as a good basis for advanced research related to Distributed Generation, genetic algorithms, and other issues related to this study. 2. It is possible that the research will be used in the power system as a reference in planning the installation of DG to solve the problem of decreasing the voltage magnitude on the system and also can reduce the power losses.