Fuzzy neural network (FNN) is a kind of network with powerful self-learning and self-tuning functions combining artificial neural network and fuzzy logic system. It is a very active research field in the field of intelligent control theory.
The branch of leap, so the study of fuzzy neural network control is of great significance. This paper aims to analyze the advantages and disadvantages of fuzzy neural networks and their uses.
Introduction to Fuzzy Neural NetworkFuzzy neural network is a kind of technology that combines the powerful structural knowledge expression ability of fuzzy logic reasoning with the powerful self-learning ability of neural network. It is the product of the combination of fuzzy logic reasoning and neural network. Generally speaking, fuzzy neural network mainly refers to the use of neural network structure to achieve fuzzy logic reasoning.
Thus, the weight of the traditional neural network without explicit physical meaning is given the physical meaning of the inference parameter in the fuzzy logic. The following mainly discusses the fusion technology of neural network and fuzzy system, the preliminary research of fuzzy inference neural network, and the fuzzy inference neural network.
Advantages of neural network control
From a control point of view, neural networks have many characteristics and advantages for automatic control compared to traditional methods:
(1) Parallel distributed information processing. The neural network has a parallel structure for parallel data processing. This
Parallel mechanism can solve large-scale real-time calculation problems in control systems, and redundancy in parallel computing
The control system can be made to be highly fault tolerant and robust.
(2) The neural network is an intrinsic nonlinear system. In theory, neural networks can achieve arbitrary non-linearity with arbitrary precision.
Sexual mapping, the network can also achieve better system modeling than other methods. This feature makes the neural network
There are broad prospects for solving nonlinear control problems.
(3) Learning and self-adaptive skills. The neural network is based on past data records of the system under study.
Practice. A trained network has induction when input to the network is not included in the training set
ability. Neural networks can also be adaptively adjusted online.
(4) Multivariable systems. The neural network can handle many input signals and has a lot of output, so it is very
Easy to use in multivariable systems.
Fuzzy neural networks can be used for fuzzy regression, fuzzy controllers, fuzzy expert systems, fuzzy pedigree analysis, fuzzy matrix equations, and general approximators.
In the field of control, the concern is a fuzzy controller composed of a fuzzy neural network. In this chapter. Introduce the basic structure of fuzzy neural network, genetic algorithm, learning algorithm of fuzzy neural network, and application of fuzzy neural network
Fuzzy neural networks come in the following three forms:1. Logical fuzzy neural network
2. Arithmetic fuzzy neural network
3. Hybrid Fuzzy Neural Network A fuzzy neural network is a neural network with fuzzy weight coefficients or input signals that are fuzzy. The arithmetic methods performed in the above three forms of fuzzy neural networks are different. Fuzzy neural networks, both as approximators and mode memories, need to learn and optimize the weight coefficients. Learning algorithm is the key to fuzzy neural network optimization weight coefficient. For the logic fuzzy neural network, an error-based learning algorithm, that is, a monitoring learning algorithm, can be employed. For arithmetic fuzzy neural networks, there are fuzzy BP algorithms, genetic algorithms, and so on. For hybrid fuzzy neural networks, there is no reasonable algorithm at present; however, hybrid fuzzy neural networks are generally used for computation rather than for learning, and it does not necessarily have to be learned.
vape smoke pen,vape smok pen,vape pod pen,vape pens with case,vape pens in bulk 5000
Shenzhen Aierbaita Technology Co., Ltd. , https://www.aierbaitavape.com