518
P. Yang et al.
4]. However, there are also some problems, such as low learning efficiency, slow
convergence, and easy to fall into local minima. In order to solve these problems,
domestic and foreign scholars have continuously improved the BP neural network
from genetic algorithm, particle swarm optimization and adaptive particle swarm
optimization [5–7]. Particle swarm optimization (PSO) algorithm has faster calculation speed, the better global search, avoiding complex genetic crossover and mutation
algorithms, and fast convergence. The local convergence problem arises because the
particle swarm algorithm converges too fast.
In order to solve the above problem, an intelligent fault diagnosis method based on
Cuckoo search (CS) Algorithm optimization neural network is proposed. The Cuckoo
search algorithm is a parasitic and Levi fly parallel mechanism [8]. This mechanism
determines the efficient optimization mode, compensates for the local convergence
problem, and converges to the global optimum. Then, the characteristic parameters
of the acoustic emission of rolling bearing are trained, and the effectiveness of the
proposed algorithm is verified.
45.2 Cuckoo Search Algorithm Method
In the way of reproduction, the cuckoo bird is different from other birds. The cuckoo
bird does not nest and hatch, and does not feed the offspring. Instead, they produce
the eggs in other nests, deceive other hosts, and raise them from the host bird, a
breeding method called nest parasitism. Cuckoo search (CS) algorithm, the cuckoo
species combining with the Lévy flight behavior, proposed by Yang and Deb in 2009
[9]. In order to describe the behavior of cuckoo bird nesting, Yang and Deb proposed
three principles in CS [10]:
1. Each cuckoo lays one egg each time and dumps it in a randomly selected nest.
2. The best nests with high-quality eggs will pass onto the next generation.
3. The number of available bird nests is certain. Under this condition, the owner
of the bird’s nest will find the eggs. Once the egg is found, the host bird will
discard the cuckoo bird eggs or abandon the bird’s nest to build a new nest. In
other words, the bird’s nest will be replaced with a certain probability.
45.2.1 Cuckoo Search Algorithm Theory
Based on the three principles proposed by Yang and Deb, the cuckoo optimization
path and position update by the following equation:
x
t+1
i
= x
t
i + δ ⊕ levy(β)
(45.1)
where x
t
i is the position of the ith (i = 1, 2, …, n) nests in the t generation;δ is the
step factor, which is used to control the size of the step. δ is usually expressed as
P. Yang et al.
4]. However, there are also some problems, such as low learning efficiency, slow
convergence, and easy to fall into local minima. In order to solve these problems,
domestic and foreign scholars have continuously improved the BP neural network
from genetic algorithm, particle swarm optimization and adaptive particle swarm
optimization [5–7]. Particle swarm optimization (PSO) algorithm has faster calculation speed, the better global search, avoiding complex genetic crossover and mutation
algorithms, and fast convergence. The local convergence problem arises because the
particle swarm algorithm converges too fast.
In order to solve the above problem, an intelligent fault diagnosis method based on
Cuckoo search (CS) Algorithm optimization neural network is proposed. The Cuckoo
search algorithm is a parasitic and Levi fly parallel mechanism [8]. This mechanism
determines the efficient optimization mode, compensates for the local convergence
problem, and converges to the global optimum. Then, the characteristic parameters
of the acoustic emission of rolling bearing are trained, and the effectiveness of the
proposed algorithm is verified.
45.2 Cuckoo Search Algorithm Method
In the way of reproduction, the cuckoo bird is different from other birds. The cuckoo
bird does not nest and hatch, and does not feed the offspring. Instead, they produce
the eggs in other nests, deceive other hosts, and raise them from the host bird, a
breeding method called nest parasitism. Cuckoo search (CS) algorithm, the cuckoo
species combining with the Lévy flight behavior, proposed by Yang and Deb in 2009
[9]. In order to describe the behavior of cuckoo bird nesting, Yang and Deb proposed
three principles in CS [10]:
1. Each cuckoo lays one egg each time and dumps it in a randomly selected nest.
2. The best nests with high-quality eggs will pass onto the next generation.
3. The number of available bird nests is certain. Under this condition, the owner
of the bird’s nest will find the eggs. Once the egg is found, the host bird will
discard the cuckoo bird eggs or abandon the bird’s nest to build a new nest. In
other words, the bird’s nest will be replaced with a certain probability.
45.2.1 Cuckoo Search Algorithm Theory
Based on the three principles proposed by Yang and Deb, the cuckoo optimization
path and position update by the following equation:
x
t+1
i
= x
t
i + δ ⊕ levy(β)
(45.1)
where x
t
i is the position of the ith (i = 1, 2, …, n) nests in the t generation;δ is the
step factor, which is used to control the size of the step. δ is usually expressed as