ABSTRACT
Today, with the deepening of artificial intelligence research, various intelligent algorithms have been widely used in various fields, leading to a series of revolutionary technological and scientific research innovations. It has no exception in the intelligent monitoring research of shortwave transmitter. Compared with the traditional manual monitoring disposal, the automatic monitoring control and intelligent state discrimination and disposal also need to be introduced. Artificial intelligence is a very broad field, involving a lot of technical concepts and research directions. One important aspect is the artificial neural network. Artificial neural network can well solve the problem of pattern recognition which is difficult to be distinguished by traditional means because of its own features such as parallelism, self-learning, selforganization and associative memory. Among the monitoring technologies based on neural network, BP neural network is a widely used model. In the intelligent monitoring of the state of shortwave transmitter, due to many types of equipment state, unapparent information difference and so on, relying solely on BP algorithm may cause problems, such as slow convergence rate of discrimination model, easy to fall into local extremum. Therefore, it is necessary to optimize and adjust BP neural network to a certain degree. In this paper,
actual calculation. The number of nodes in each layer of the network structure is set according to the types of input and output data as well as related requirements of network construction, and the number of nodes on the network is the number of neurons. Data enters into the network in the input layer and moves forward along with the arrow direction, then moves forward after calculated at each node, and finally the result is output on the output layer.
