RETRACTED ARTICLE: Forest management and resource monitoring based on AMI intrusion detection algorithm and artificial intelligence

Abstract

Smart grid is the combination of computer network and traditional power system to realize the intelligence of power grid. AMI is an advanced measurement system, which connects the power system and users, realizes the two-way interaction of data and information between power supply enterprises and users, and promotes the development of smart grid. Therefore, the safe use of AMI system is the key to the development of smart grid. With the increasingly close connection between smart grid and computer network, the number of network attacks against AMI system continues to increase. However, the current security defense technology of AMI system is passive defense technology represented by protocol and encryption, which can not resist the invasion of unknown network attacks. The active defense technology represented by intrusion detection is an important barrier of AMI security defense. Now, AMI intrusion detection algorithm based on machine learning is often proposed. This paper will focus on the above problems in the process of AMI intrusion detection. Because the data size is small and the data function is not complete, the detection accuracy of DBN-OS-RKELM algorithm is low. An improved generalized regression neural network (GRNN) intrusion detection algorithm DBN-FOA-GRNN is proposed. The algorithm uses the generalized regression neural network (GRNN) with excellent nonlinear mapping function and high convergence rate. In order to improve the performance of GRNN intrusion detection, population optimization algorithm (FOA) is used to optimize the unique random parameter of GRNN to reduce the possibility of falling into local optimum. This paper uses the public intrusion detection data set NSL-KDD to verify and analyze the algorithm. Experimental results show that the two intrusion detection algorithms proposed in this paper have some improvement in detection effect compared with the traditional machine learning methods. The combination of the two algorithms can solve the problem of intrusion detection of various data, and meet the requirements of AMI intrusion detection.