The 2D-BPSO algorithm presented in Algorithm 2, is an ex-tension of the original 2D-BPSO algorithm defined in [9], [10]. For completeness, we explain the key aspects of the algorithm. The algorithm starts with a population of possible solutions all randomly generated. During each generation, the algorithm computes the fitness of each solution in the population. Then as presented in line 7, the velocity of each particle is updated properly, according to an inertia weight a, the individual knowl-edge of each particle, and the global knowledge (i.e., best solu-tion known by all particles). The value of a can help to con-verge; the algorithm starts with a maximum value and linearly decreases the value to a minimum on each generation. Such de-crease causes the search for solution to explore more at the be-ginning, and refine the search to exploit the local discoveries, at later stages of the evolution. The algoritlun contains N populations of solutions, one for each green energy source (e.g., solar panels, wind energy, nat-ural gas, etc.). With these populations, it becomes possible to achieve mutual exclusion of loads, that is, no load should be connected to mom than one energy source at any particular time. Such mutual exclusion, can be performed using simple logical operations, like AND and XOR in lines 17-22. The performance (fitness) of solutions are evaluated on each energy source, and then added together such that the total benefit of the solution can be compared among the possible solutions. The smart home system can then switch the energy source use per appliance given the 2D-BPSO solution.
The 2D-BPSO algorithm presented in Algorithm 2, is an ex-tension of the original 2D-BPSO algorithm defined in [9], [10]. For completeness, we explain the key aspects of the algorithm. The algorithm starts with a population of possible solutions all randomly generated. During each generation, the algorithm computes the fitness of each solution in the population. Then as presented in line 7, the velocity of each particle is updated properly, according to an inertia weight a, the individual knowl-edge of each particle, and the global knowledge (i.e., best solu-tion known by all particles). The value of a can help to con-verge; the algorithm starts with a maximum value and linearly decreases the value to a minimum on each generation. Such de-crease causes the search for solution to explore more at the be-ginning, and refine the search to exploit the local discoveries, at later stages of the evolution. The algoritlun contains N populations of solutions, one for each green energy source (e.g., solar panels, wind energy, nat-ural gas, etc.). With these populations, it becomes possible to achieve mutual exclusion of loads, that is, no load should be connected to mom than one energy source at any particular time. Such mutual exclusion, can be performed using simple logical operations, like AND and XOR in lines 17-22. The performance (fitness) of solutions are evaluated on each energy source, and then added together such that the total benefit of the solution can be compared among the possible solutions.ระบบบ้านอัจฉริยะสามารถเปลี่ยนแหล่งพลังงานที่ใช้ต่ออุปกรณ์ที่ได้รับ 2d-bpso โซลูชั่น
การแปล กรุณารอสักครู่..
