^{1}

^{*}

^{1}

^{2}

In this paper an Enhanced Bean Optimization Algorithm (EBA) is used to solve optimal reactive power problem. Stimulated by the diffusion of beans in nature, a novel swarm intelligence algorithm-Bean Optimization Algorithm (BOA) has been projected previously. In the domain of incessant optimization problems solving, Bean Optimization Algorithm has exposed a first-class performance. In this paper, an Enhanced Bean Optimization Algorithm is presented for solving optimal reactive power problem. In this algorithm two novel evolution methodologies named population migration and deductive information cross-sharing are proposed to perk up the performance of Bean Optimization Algorithm. The projected Enhanced Bean optimization algorithm (EBA) has been tested in standard IEEE 30 bus test system and simulation results show clearly the enhanced performance of the projected algorithm in tumbling the real power loss.

Different algorithms are utilized to solve the Reactive Power Dispatch problem. Different types of numerical techniques like the gradient method [

Main aim of the reactive power dispatch problem is to reduce the active power loss in the transmission network, which can be described as:

where g_{k}: is the conductance of branch between nodes i and j, Nbr: is the total number of transmission lines in power systems.

For minimization of the voltage deviation in PQ buses, the objective function turns into:

where ω_{v}: is a weighting factor of voltage deviation.

VD is the voltage deviation given by:

The equality constraint of the Reactive power problem is represented by the power balance equation, and can be written as, where the total power generation must cover the total power demand and total power loss:

where,

Inequality constraints define the limitations in power system components and power system security. Upper and lower bounds on the active power of slack bus, and reactive power of generators are written as follows:

Upper and lower bounds on the bus voltage magnitudes are described as follows:

Upper and lower bounds on the transformers tap ratios are given as follows:

Upper and lower bounds on the compensators reactive powers are written as follows:

where N is the total number of buses, N_{T} is the total number of Transformers; N_{c} is the total number of shunt reactive compensators.

Stimulated by the diffusion mode of beans, Bean Optimization Algorithm (BOA) has been proposed previously to solve the various problems. In BOA, the position of an individual bean is articulated with real number vector and written as

where n is determined by the scale of problem .Bean group is comprises of large number of beans. The size of the bean group can be attuned depending upon realistic problems. In adding to the above, beans are propagated to the region and the area is defined by the type of problem. Father beans are those beans whose fitness value is greater than others. In BOA, the number and distribution of offspring beans will be placed according to their father bean’s fitness value. The fundamental equation of BOA is written as follows,

In the above equation,

Bean Optimization Algorithm (BOA) utilizes population evolution mechanism for solving optimization problems. Since most of the population evolution methods are continuous, they are complicated to solve discrete optimization problems. In this paper an Enhanced Bean Optimization Algorithm (EBA) is utilized for solving Reactive Power Problem.

The algorithm model can be described as follows,

1) Individual beans

The position vector of an individual bean is located as

The above indicates that there is a route as

2) Population progress

In the procedure of population migration, minimum two populations should be initialized. The father bean in each population will be mixed up in cross-species process through the interaction between populations in order to endorse the affluence of populations.

3) Cross-sharing of deductive information

In order to keep the deductive information of the father beans, there are cross operations between the father beans and the individual beans to create new offspring’s.

The explicit operation is shown as follows.

1) Pick an arbitrary position in the vectors of a father bean f and an individual bean s separately as a cross-re- gion.

2) Swap cross-region between f and s. Then remove the duplicate elements in f and s separately. Two new offspring individuals’ g and h will be produced.

In EBA, the first step is population has to be initiated (let the size of population be n). According to the fitness values of individual beans, choose the father beans (let the number of father beans be three): R_{1}, R_{2}, R_{3}. (n − 3)/3 individuals will be displayed as sub-populations “1” according to the Euclidean distance between individual beans and R_{1}. By using same method, sub-populations 2 and sub-population 3 will be produced. Then let R_{2} be the cross father bean of sub-population 3 and cross operations will be carried out between R_{2} and individual beans in sub-population 3. Choose the offspring with the most excellent fitness value to shift the previous individual bean in sub-population 3. Let R_{3} be the cross father bean of sub-population 1 and cross operations will be carried out between R_{3} and individual beans in sub-population 1. Pick the offspring with the most excellent fitness value to shift the former individual bean in sub-population 1. Let R_{1} be the cross father bean of sub-popu- lation 2 and cross operations will be carried out between R_{1} and individual beans in sub-population 2. Choose the offspring with the most excellent fitness value to relocate the previous individual bean in sub-population 2.

Reiterate the above procedure until the termination condition is met.

EBA for solving Optimal Reactive Power problem

Set the number of iterations be S.

Arbitrarily produce n initial beans.

Compute the fitness value of the preliminary beans and Select S father beans.

Create z sub-populations by using clustering algorithm.

While (the number of iterations < S)

For i = 1:S

For j = 1:n

Cross operations are carried out between Y_{j} and R_{(i+1)};

The bean with the best fitness value is recorded as Y_{j}_{1};

Y_{i} = Y_{j}_{1};

End

Modernize the Father beans;

End

End

Output the finest solution.

Enhanced Bean Algorithm has been tested in IEEE 30-bus, 41 branch system. The system has 6 generator-bus voltage magnitudes, 4 transformer-tap settings, and 2 bus shunt reactive compensators. Bus 1 is considered as slack bus and 2, 5, 8, 11 and 13 are considered as PV generator buses and the other buses are taken as PQ load buses. Generators buses (PV) are 2, 5, 8, 11, 13 and slack bus is 1. Control variables limits are listed in

List of Variables | Min. Value | Max. Value | Category |
---|---|---|---|

Generator Bus | 0.90 | 1.08 | Continuous |

Load Bus | 0.90 | 1.01 | Continuous |

Transformer-Tap | 0.91 | 1.00 | Discrete |

Shunt Reactive Compensator | −0.10 | 0.30 | Discrete |

Bus | Pg. | Pgmin | Pgmax | Qgmin |
---|---|---|---|---|

1 | 90.00 | 47 | 121 | −20 |

2 | 82.00 | 18 | 75 | −20 |

5 | 50.00 | 10 | 41 | −11 |

8 | 20.00 | 10 | 32 | −13 |

11 | 20.00 | 10 | 19 | −10 |

13 | 20.00 | 11 | 35 | −13 |

Control Variables | EBA |
---|---|

V1 | 1.0612 |

V2 | 1.0503 |

V5 | 1.0312 |

V8 | 1.0417 |

V11 | 1.0814 |

V13 | 1.0601 |

T4, 12 | 0.00 |

T6, 9 | 0.01 |

T6, 10 | 0.90 |

T28, 27 | 0.90 |

Q10 | 0.11 |

Q24 | 0.11 |

Real power loss | 4.2781 |

Voltage deviation | 0.9057 |

Iterations | 25 |
---|---|

Time taken (secs) | 4.32 |

Real power loss | 4.2781 |

Methods | Real power loss (MW) |
---|---|

SGA [ | 4.98 |

PSO [ | 4.9262 |

LP [ | 5.988 |

EP [ | 4.963 |

CGA [ | 4.980 |

AGA [ | 4.926 |

CLPSO [ | 4.7208 |

HSA [ | 4.7624 |

BB-BC [ | 4.690 |

EBA | 4.2781 |

In this paper, Enhanced Bean Optimization Algorithm (EBA) has been efficiently solved the Optimal Reactive Power Dispatch problem. The projected algorithm has been tested in standard IEEE 30 bus system. Simulation study shows the robustness of projected Enhanced Bean Optimization Algorithm (EBA) method in providing improved optimal solution by decreasing the real power loss. The control variables values obtained after the optimization by Enhanced Bean Optimization Algorithm (EBA) are well within the limits.

Kanagasabai Lenin,Bhumanapally Ravindhranath Reddy,Munagala Suryakalavathi, (2016) Enhanced Bean Optimization Algorithm for Solving Reactive Power Problem. Open Access Library Journal,03,1-8. doi: 10.4236/oalib.1102464