Distribution System Load Flow Method


Abstract

The distribution system load flow method need to be modified because of the increasing penetration of distributed generators (DG) in the power of distribution system. Thus, for the load flow is accuracy, we need to careful in the selection of the suitable model of DG. In this research, we consider DG as constant negative load in a simple algorithm of the distribution system load flow with DG. We will examine and show the results for the test of distribu

1. Introduction
First of all, we need to understand about the load flow analysis. We should know that load flow analysis is an important tool for power system to plan and operate. However, we also have the conventional load flow methods such as Newton-Raphson or Fast Decoupled. These methods are typically designed for electrical transmission system. They are not suitable for the power distribution system load flow analysis. We know that the distribution networks are typically radial in the nature. And, their feeders usually have high R/X ratio. Thus, they are not capable for the load flow analysis method. In the last two decades, several load flow analysis methods have been deployed for various applications in distribution automation (DA). These methods are known as distribution system load flow (DSLF) methods. For the radial distribution networks (RDN), these methods are exploiting their special topological characteristics. For most of the DSLF methods, the forward and backward sweep (FBS) based methods can be considered as the most simple and fast methods to carry out the distribution system cash flow [3].  

In recent times, the distribution system has gained the focus from the industry due to the increasing penetration of the distributed generators (DGs). We will see that the integration of the DGs into the distribution systems have altered the basic configuration of passive systems to let to active one of them. [7] This can be given some of the benefits as well as the challenges. Some of their benefits are reducing the loses of lines, improving the profile of vantage, reducing the emissions of pollutants, increasing the overall of energy, enhancing the reliability and security of the system, improving the quality of the power, and the relieving the congestion of the T&D.
 
[11] In this research, we will entitle the fourteen challenge relevant to the analysis of DG in the distribution system such as:
	The providing the screening applications,
	The solution of the power flow
	The analysis of multiphase
	The model size of the circuit
	The dynamics, harmonics, determining the value of DG
	The modeling, assessing distribution reliability
	The loss analysis
	The protective device coordinations and the transformation of the connection devices.
Therefore, the DSLF analysis with DG attracts most of researchers to the model of DGs in the best possible way for the suitable inclusions. For the development of lights, in this research papers, we will discuss the DSLF analysis in the addressing of the DG modeling issues. We also consider the FBS based on the DSLF methods with some of the novel bus identification scheme and constant negative for the power of the the DGs model. The first three results for the distribution test systems will be presented [2].

We are realized that the penetration of DG into the distribution systems  have a fast expansion and commercialization. In the normal way, we see that the distribution systems are passive systems. In these systems, the flow of both real and reactive power is always from the higher to lower voltage levels. However, the power flows may becomes reversed in the penetration distributed generation system. This is because of the system is no longer a passive circuit. The circuit that will supply loads but does not give an active system with power flows and voltages. Therefore, there are many changes on the natire of the distribution systems with the distributed generation [8].     

We see that the change within the distribution systems in the integration of the DGs have created some of the provisions. These provisions have the positive as well as the negative impact to the distribution systems and the customers. There are some benefits of DGs as the voltage profile improvement, the line loss reduction and the reduction impact in the environment. [14] They also have the multi objective performance index for the distribution network with distributed generation in. This research papers will focus on the DG planning models, considering some of the technical requirements as the rating of thermal, the rising of the voltage, the system fault levels, the losing of the power, We also examine on the maximize method to maximize the distributed wind power generation for distributed systems. 

The maximization of the distributed generation can be taken into the conventional of the electric distribution systems to reduce the power loss. [1] If the distributed generators are not placed properly, the power loss may be shown. We need to find a method to identify the optimal bus to place the DGs in a system based on the bus admittance matrix, generation information and the load distribution of the system. The limitation of this technique is in the way that it will make to loss sensitivity factor that may not lead to the best placement for loss reduction [5]. 


2. Literature Review
For the literature review, we will examine many methods are available to carry out the load flow in the RDN. For this research paper, we will study the FB based load flow algorithm in details [9].

2.1 The Modeling for the Balanced RDN
We assume that we have the balance of RDN in the nature. We have an equivalent single phase feeder that be shown in the Figure 1. In this Figure, there is the sending end bus a’ in the receiving of the bus ending with Ia is the feeder branch current. And, we have Z is the impedance of the branch feeder. This will send the end voltage V and the receiving end voltage VT as [4]:

 
Figure 1: Equivalent Single Phrase Feeder

In the Figure 1 above, we see that the load connected to various buses of the distributed systems. We are assumed that they are constant power loads. We can implement the load flow the load connected at the bus to represented as the same as current injection. Next, we will consider the arbitrary bus-i of the RDN in the Figure 2 to have a constant complex power load is connected. We see in the Figure 2 the bus i is corresponding to load the current injection or the equivalent current injection h. We will compute this as a function of the bus voltage as [6]:

In the equation above, the P and Q are equal to PLi and QLi at bus-I as shown the in Figure below:
 
Figure 2: Current Injection at bus-i

2.2 The Scheme for Bus Identification 
For a fast implementation of FBS based load flow, we need to proposed several arrays for bus identification. These arrays are proven to be extremely useful for its reduced search time and fast implementation of load flow. Thus, we will do some reviews of the various arrays. An array of dimension can be the double the number of branches (nbr) of a radial distribution system. It can be namely as adb(2*nbr). And, it can store all the adjacent or neighboring buses of each of the buses of the RDN. Two other arrays are mf and mt of dimension `nb'. These arrays nb (number of buses in a RDN) act as pointers of the adb array. These arrays in turn govern the reservation of allocation of memory locations for each bus. It means that mf and mt point to the starting and end addresses in the adb array respectively. Similarly, all the previous buses are identified and stored in an array pb. These arrays can be formed from the general system data of a RDN [10].

Next, we will consider about two more arrays, namely, nsb and sb. The array nsb stores the number of subsequent buses corresponding to various branches of a RDN and has a dimension equal to nbr. The sb array stores all the subsequent buses to each of the the branches of the system. Two pointer arrays mfs and mts are also introduced to point to the start and end memory locations in sb array. The arrays mfs, mts, nsb and sb can be formed from the input data of a RDN. The general storage and pointer operations of these arrays are explained later with reference to a fictious RDN of nb buses and nbr branches [11].

2.3 The method of FBS based on DSLF 
FBS based DSLF is a well established method. This method exploits the various topological specialties of the RDN and is quite fast as well as simple to implement. This is an iterative method. In the first step, starting from any end bus the RDN branch currents are gradually calculated till the substation bus or the root bus. In the second step, starting from the substation bus all subsequent buses up to the end buses are updated using. This equation uses to calculate branch currents in the backward sweep. This process continues till voltage magnitudes converge. The current in any branch of a RDN can be calculated using as[12]: 
 

3. Research Design 
Distributed Generation (DG) is an electric active power source connected directly to the distribution network or customer side of the meter. However, DG is not a new concept but it is an emerging approach for providing electric power in the heart of the power system. It encompasses several technologies which can be broadly categorized as renewable or nonrenewable. [15] Renewable DG includes small hydro plants, wind turbines, photo voltaic cells, fuel cells, geothermal power plants, biomass power plants, tidal power plants, wave power plants etc., whereas non renewable category includes conventional fossil fuel based generators, micro turbines, etc. In this research papers, a broader term such as Distributed Energy Resources (DER) can be defined, which refers to the electric power generation resources that are directly connected to the medium voltage (MV) or low voltage (LV) distribution systems and it includes both generation units
One of the challenges to consider DG in the analysis and design of distribution systems is the power flow solution. That is taken into account the proper modeling of embedded DGs. There are three types of mathematical models of DGs for load flow analysis as [8]: 
	A constant power factor model for synchronous generator and power electronics based DGs 
	A reactive power model for induction generator based DGs 
	A constant voltage model for large scale controllable DGs 
Depending on the control, the DG may be set to output power at either constant power factor for small DG or constant voltage for large DG. Thus, two types of DG models need to be developed: constant PQ modeled as negative loads with currents injecting into the node and PV nodes. The model of DGs as PV or PQ depends on its operational mode and control characteristics. 

3.1 Distributed Generation Model PQ
Since DGs are normally smaller in size when compared with the conventional power sources, the constant PQ model is commonly found to be sufficient for the distribution system load flow analysis. This model is adequate because DGs typically not permitted to regulate the voltage. Instead, they regulate power and power factor, hence modeled as negative loads. [14] Most of the DGs are equipped with automatic voltage regulators (AVR) and operate in constant power output mode. Therefore, voltage output level of the DGs are same as system voltage. Hence, it is preferable to handle the interconnection nodes of DGs as the PQ node model rather than the PV node model as.

3.2 Distributed Generation with FBS based Load Flow 
With DGs modeled as negative loads, the equivalent loads at bus-i can be expressed as: 
Pt = PLi - Pgi
Qt = QLi - Qgt
PE and Qu are the constant power loads connected at bus -i and Pg, and Qg, are the real and reactive powers injected by the DG connected at bus-i respectively. The ECIs then can be calculated at all the buses using. The backward and forward sweeps are then followed to execute the load flow [9].

3.3 Distributed Generation Testing System
Three test distribution systems are considered to validate the model and the method described in this research paper. The Figure below shows a 12 bus radial test system with DGs connected at bus -6 and 12 and the system data is presented in Table-I. The Figure aslo shows the second test radial distribution system of 33 bus and DGs are connected at bus 5 and 18. Pg and Qg of these DGs are considered to be 10% and 7% of total active power and reactive power load of the corresponding system respectively. The third test distribution system is a 69 bus radial system  with six DGs connected at bus-11, 22, 31, 38, 53 and 58 [7].
 
Figure 3: Distributed Generation bus test system


4. Conclusion 
Finally, in this research papers, a simple algorithm for DSLF with DG has been proposed. Although, the basic FBS based method has been followed but the proposed bus identification scheme has been used for a faster implementation. The DGs have been modeled as constant negative power loads. Results for three radial test distribution systems have been presented. 

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.