Academia.eduAcademia.edu

Figure 4 – uploaded by Ahmed Aly

See full PDFdownloadDownload figure
From the results above, one can determine the individual best. Because that is the first iteration, this is already the individual best of each particle. The global best is given by the most ef- fective result of the population, which is given by particle 4 (0.2181). In order to determine a new population, one needs to take into consideration the individual best of each particle, the global best of the population, the own value of the population, the weight inertia, the accelerator coefficient, and also the par- ticle velocity. It is important to mention that, not necessarily, all the particles produce better results as the process goes on. This means that a particle may be taken to a position associated with a worse result than that observed in a previous iteration. This is the core of a cognitive process, since new positions are searched up to a stopping criterion is flagged. The value 0.2162 is the global

Table 4 From the results above, one can determine the individual best. Because that is the first iteration, this is already the individual best of each particle. The global best is given by the most ef- fective result of the population, which is given by particle 4 (0.2181). In order to determine a new population, one needs to take into consideration the individual best of each particle, the global best of the population, the own value of the population, the weight inertia, the accelerator coefficient, and also the par- ticle velocity. It is important to mention that, not necessarily, all the particles produce better results as the process goes on. This means that a particle may be taken to a position associated with a worse result than that observed in a previous iteration. This is the core of a cognitive process, since new positions are searched up to a stopping criterion is flagged. The value 0.2162 is the global

Related Figures (6)

The search space and initialization ranges for the experi- ments are listed in Table I. All experiments consisted of 100 executions. The PSO and HPSOM parameters were set to the values C; = Co = 2.0, and a linearly decreasing inertia weight starting at 0.7 and ending at 0.4 was used. The maximum velocity (Vmax) of each particle was set to be half the length of the search space in one dimension. The population size in
The search space and initialization ranges for the experi- ments are listed in Table I. All experiments consisted of 100 executions. The PSO and HPSOM parameters were set to the values C; = Co = 2.0, and a linearly decreasing inertia weight starting at 0.7 and ending at 0.4 was used. The maximum velocity (Vmax) of each particle was set to be half the length of the search space in one dimension. The population size in
SEARCH SPACE AND INITIALIZATION RANGES
SEARCH SPACE AND INITIALIZATION RANGES
RESULTS OF AVERAGE BEST FITNESS OF 100 EXECUTIONS
RESULTS OF AVERAGE BEST FITNESS OF 100 EXECUTIONS
‘ig. 2. Graphic of HPSOM behavior for IEEE 118-bus system.
‘ig. 2. Graphic of HPSOM behavior for IEEE 118-bus system.
The value 0.2158 [i.e., the particle (0.0686, 0.0919)] is the global best of this population, since it provides the smallest
The value 0.2158 [i.e., the particle (0.0686, 0.0919)] is the global best of this population, since it provides the smallest
Related topics:
Power SystemsReactive PowerEvolutionary ComputingReactive Power CompensationLoad FlowElectrical and Electronic EngineeringNonlinear OptimizationParticle Swarm OptimizerNonlinear Equations
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved