As a symbol of the peaceful use of nuclear energy by mankind, nuclear power has become an indispensable category in the world's energy structure with its clean and efficient features since it was developed in the last century. Safety and stability are the prerequisites for the sustainable development of nuclear energy, and how to ensure the safety of reactor operation has become the focus of research. As an important branch of computer network science and engineering, artificial intelligence technology has been widely used in industrial production due to its high efficiency, good at repeating tedious mechanical work, high reliability, and low cost.  Now, as the nuclear industry continues to pay attention to less-manned and unmanned on-duty monitoring, the application of artificial intelligence in the field of nuclear engineering will become more and more diverse. This article will summarize the application of artificial intelligence in the nuclear field. First, it will give a basic introduction to artificial intelligence, and then introduce the application status of artificial intelligence in the nuclear field from the aspects of working condition prediction, fault identification and operation and maintenance according to the application. Some discussions and suggestions are presented for further research in nuclear engineering.
Artificial intelligence (AI) technology, a pivotal branch of computer network science and engineering, has found widespread application in industrial production owing to its attributes of high efficiency, proficiency in repetitive and mundane tasks, exceptional reliability, and cost-effectiveness. Presently, with the nuclear industry progressively emphasizing unmanned and minimally manned monitoring, the application of AI in nuclear engineering is poised to diversify. This article aims to encapsulate the multifaceted applications of artificial intelligence in the nuclear domain. The pressurized water reactor (PWR) is the most widely used reactor type in the world, and its safe operation requires many considerations. Among them, the pipe rupture accident is a serious accident that may cause pressure loss in the reactor, coolant leakage, and core meltdown. Therefore, timely and accurate judgment of PWR piping rupture is an important means to ensure nuclear safety. This study takes a typical pressurized water reactor as the research object and uses the random forest algorithm. The method involves the collection of actual data and accident simulation, combined with artificial intelligence algorithms, so that the algorithmic model can be adapted to the operating conditions. The training results show that within the test range, the accuracy of the random forest algorithm for identifying pipe breaks is over 95%, which provides valuable help for the identification of reactor pipe breaks.
 

Artificial Intelligence,Parameter Prediction,Fault Identification