Research on Security Protection Measures of the Perception Layer of Power Distribution Internet of Things

Power Grid Multi-protocol Network Topology Based on Data Exchange Strategy

“…According to the multi-protocol marked grid initial network topology shown in Figure 2, the networking analysis of grid network link is carried out [6]. Assuming that the edge vector of the grid core network is Li{i=1，2，…，Cl}, the directed graph G={V,E} is used to represent the random link distribution model of the grid, where V is the edge set of the grid, E Is a multi-protocol marker point [7]. The location update equation for the multi-protocol intelligent data exchange node is as follows [8].…”

“…In the grid communication environment, the node C does not know the delivery message, and the exchange time of the three broadcast packets received by the node B and the node C is described as follows. (7) The message complexity of each packet along the path is recorded. The node of the multi-protocol label is NP, and the energy cost of the data exchange core in the life cycle of the grid communication environment can be gotten.…”

“…The node of the multi-protocol label is NP, and the energy cost of the data exchange core in the life cycle of the grid communication environment can be gotten. 8), the coefficient α2≥1 infers the coverage of the node B to the node A link until the loop termination condition is satisfied, and the iteration is completed to realize the intelligent data exchange protocol optimization design and the multi-protocol intelligent data exchange node of the data exchange network [6][7].…”

Abnormal Data Detection and Identification Method of Distribution Internet of Things Monitoring Terminal Based on Spatiotemporal Correlation

“…As an important part of the ubiquitous power Internet of Things, the distribution Internet of Things can effectively improve the automation and informatization level of the distribution network [1,2] and provide users with diversified and differentiated energy services; thus, further enhancing the level of electricity safety for customers [3,4]. The architecture of the distribution Internet of Things is divided into a perception layer, a network layer, a platform layer, and an application layer.…”

“…At present, there are few studies on the detection of abnormal data of LTUs in the distribution Internet of Things, but there are many abnormal data detection methods for IoT terminals, which can be divided into four categories: (1) anomaly detection methods based on statistics [6,7], which need to establish a complete dataset in advance and understand the prior information, resulting in limited detection conditions and poor real-time performance; (2) distance-based anomaly detection methods [8,9] for detecting the top n outliers based on the distance function; however, this method increases the network communication overhead and is not suitable for dynamic changes in network topology and multidimensional data; (3) density-based anomaly detection methods [10,11], which take a long time to calculate; if the size of the dataset is m, the time complexity is O(m 2 ), making this method unsuitable for the detection of power distribution IoT data with a large amount of data requiring real-time monitoring; (4) anomaly detection methods based on pattern recognition [12,13]. Due to the nonlinearity, complexity, ambiguity, and randomness of abnormal data during the operation of LTUs, it is difficult to express with precise mathematical equations, and the complex working environment of LTUs puts forward higher requirements for the real-time performance and robustness of the detection method.…”