Water affair Internet of Things abnormity self-detection method and system based on edge calculation

Abstract

The invention discloses a water affair Internet of Things abnormity self-detection method and system based on edge calculation, and relates to the technical field of water affair Internet of Things data processing. The method is used for solving the problems of equipment monitoring delay and false alarm in a weak network environment. The method comprises the following steps: firstly, carrying out space-time alignment and denoising on heterogeneous data through a sliding time window and discrete Kalman filtering, and constructing a real-time working condition state matrix; then, a logic engine is used for executing Boolean operation based on an equipment interlocking rule, and a hard fault risk value is generated; then, a prediction deviation is calculated in combination with a time sequence reasoning model, a gain coefficient is dynamically adjusted according to the network packet loss rate, a comprehensive decision index is calculated in a weighted mode, and weak network self-adaptive decision making is achieved; finally, when the index exceeds the limit, a relay is directly driven to act, and the nonvolatile memory is addressed to solidify fault data. According to the invention, full-closed-loop autonomous protection independent of the cloud is realized.

Images