LESSENING OF ACTUAL POWER LOSS BY MODIFIED ALGORITHM

This paper presents a Modified Teaching-Learning-Based Optimization (MTLBO) algorithm for solving reactive power flow problem. Basic Teaching-Learning-Based Optimization (TLBO) is reliable, accurate and vigorous for solving the optimization problems. Also, it has been found that TLBO algorithm slow in convergence due to its high concentration in the accuracy. This paper presents an, Modified version of TLBO algorithm, called as Modified Teaching-Learning-Based Optimization (MTLBO). A parameter called as “weight” has been included in the fundamental TLBO equations & subsequently it increases the rate of convergence. In order to evaluate the proposed algorithm, it has been tested in practical 191 test bus system. Simulation results reveal about the better performance of the proposed algorithm in reducing the real power loss & voltage profiles are within the limits.


Introduction
Optimal reactive power dispatch problem is one of the difficult optimization problems in power systems & various mathematical techniques [1][2][3][4][5][6][7] have been utilized to solve the problem. Recently many types of Evolutionary algorithms [8][9] have been used to solve the reactive power problem. This paper presents a Modified Teaching-Learning-Based Optimization (MTLBO) algorithm for solving reactive power flow problem. Basic Teaching-Learning-Based Optimization (TLBO) [10][11][12][13][14][15][16] is reliable, accurate and vigorous for solving the optimization problems. Also it has been found that TLBO algorithm slow in convergence due to its high concentration in the accuracy. This paper presents an, Modified version of TLBO algorithm, called as Modified Teaching-Learning-Based Optimization (MTLBO). A parameter called as "weight" has been included in the fundamental TLBO equations & subsequently it increases the rate of convergence. In order to evaluate the proposed algorithm, it has been tested in practical 191 test bus system. Simulation results reveal about the better performance of the proposed algorithm in reducing the real power loss & voltage profiles are within the limits.

Active Power Loss
The objective of the reactive power dispatch is to minimize the active power loss in the transmission network, which can be described as follows: Where F-objective function, PLpower loss, gk-conductance of branch,Vi and Vj are voltages at buses i,j,Nbr-total number of transmission lines in power systems.

Voltage Profile Improvement
For minimizing the voltage deviation in PQ buses, the objective function becomes: Where VD -voltage deviation,ω v -is a weighting factor of voltage deviation.
Voltage deviation given by: Where Npq-number of load buses

Equality Constraint
The equality constraint of the problem is represented by the power balance equation, where the total power generation must cover the total power demand and the power losses: Where PG-total power generation,PD -total power demand.

Inequality Constraints
The inequality constraints in the power system as well as the limits created to ensure system security. Upper and lower bounds on the active power of slack bus (Pg), and reactive power of generators (Qg) are written in mathematically as follows: Upper and lower bounds on the bus voltage magnitudes (Vi): Upper and lower bounds on the transformers tap ratios (Ti): Upper and lower bounds on the compensators reactive powers (Qc): Where N is the total number of buses, NT is the total number of Transformers; Nc is the total number of shunt reactive compensators.

Basic Teaching-Learning-Based Optimization
Based on the consequence of the influence of a teacher on the output of students in a class, Teaching-Learning-Based Optimization (TLBO) optimization algorithm has been framed. It is a population-based method & there are numbers of different design variables. Different subjects offered to learners and the learners' result is analogous to the "fitness" & it act as different design variables in TLBO. The most excellent solution is analogous to Teacher in TLBO. The algorithm consists of first part "Teacher Phase" and the second "Learner Phase". Learning from the teacher is the "Teacher Phase" means and learning through the interaction between learners is the "Learner Phase". The execution of TLBO as follows,

1) Initialization
Following are the notations used for describing the TLBO: L: "class size "of the learners; C: list of the courses offered to the learners to learn; MAX IT; number of maximum iterations. In search space bounded the population Y is arbitrarily initialized by a by matrix of L rows and C columns. In ith learner . the jth parameter is assigned values arbitrarily by equation Within the range (0,1) "rand" represents a uniformly distributed arbitrary variable, minimum and maximum value for jth parameter is represented by and . For the generation g parameters of the ith learner are given by,

2) Teacher Phase
At generation g the mean parameter Eg of each subject learners in the class is given as, The teacher ℎ with the minimum objective function value of the learner is considered as for respective iteration. Shifting the mean of the learners towards its teacher is done by Teacher phase. An arbitrary weighted differential vector is formed to obtain a new-fangled set of improved learners from the current mean, desired mean parameters and is added to the existing population of learners.
The value of mean to be changed is decided by " " -teaching factor. Value of can be either 1 or 2. With equal probability the value of is decided arbitrarily as, The value of value is arbitrarily decided by the algorithm using Equation (14). In generation g if ( ) is superior learner than ( ) , than it swap the inferior learner ( ) in the matrix.

3) Learner Phase
In this phase the mutual interaction tends to augment the knowledge of the learner. The arbitrary inter-action among learners improves the knowledge. For a given learner ( ) another learner ( ) is arbitrarily selected( ≠ ). In the learner phase the ith parameter of the matrix Ynew is given as,

4) Algorithm Termination
After MAXIT conditions satisfied the algorithm is terminated.

Modified Teaching-Learning-Based Optimization (MTLBO) Algorithm
The principles of teaching-learning approach is imitated in Teaching-Learning-Based Optimization (TLBO) & draw analogy with the real class room. Teaching-learning process is an iterative process where in the continuous interaction takes place for the transfer of knowledge. A parameter known as "weight" is added in the Equations (13) and (15)  value is considered and decided by a weight factor "wf' in our Modified Teaching-Learning-Based Optimization (MTLBO) algorithm. During the early stages of the search Individuals are encouraged to sample diverse zones of the exploration space. It is important to adjust the movements of trial solutions finely & they can explore the interior of a relatively small space in the later stages. Value of the weight factor reduced linearly with time from a maximum to a minimum value by, The maximum and minimum values of weight factor w are and , "I" -iteration is the current iteration number and max iteration is the maximum number of iterations. & are selected between 0.9 -0.1, respectively. New set of improved learners in the teacher phase can be, And in learner phase a set of improved learners are,

Simulation Results
Modified

Conclusion
In this paper a novel approach Modified Teaching-Learning-Based Optimization (MTLBO) algorithm used to solve reactive power problem, considering various generator constraints, has been successfully applied. Also it has been found that TLBO algorithm slow in convergence due to its high concentration in the accuracy. The performance of the proposed Modified Teaching-Learning-Based Optimization (MTLBO) algorithm has been has been tested in practical 191 test bus system and simuation results expose about the decrease of real power loss & volatge profiles are within the limits.