Method and device for abnormal real-time detection and processing of agricultural Internet of Things data stream

Abstract

The invention provides a real-time agricultural Internet of Things data stream abnormality detection and processing method and a real-time agricultural Internet of Things data stream abnormality detection and processing device. In the method, the size of a sliding window is first determined according to a characteristic cycle of acquired data and a time interval between acquisitions, and a measured value acquired by a sensor at the current moment and a prediction interval are then predicted according to historical acquired data in the sliding window; an actual measured value which is actually acquired at the current moment is then compared with the precision interval, if the actual measured value does not fall into the prediction interval, then data which are actually acquired at the current moment are determined as abnormal data, and thereby the detection of abnormal data is fulfilled. In addition, after the method provided by the invention determines that the data at the current moment are abnormal data, the obtained predicted value is utilized to replace the abnormal data, and thereby the processing of the abnormal data can be fulfilled. The accuracy of acquired data streams is effectively increased, and powerful data support is provided for the automatic control of equipment and effective data analysis.

Description

technical field [0001] The invention belongs to the field of computer technology, and in particular relates to a real-time detection and processing method and device for an abnormal data stream of the agricultural internet of things. Background technique [0002] The Internet of Things has been widely used in agricultural production such as livestock and poultry breeding, facility gardening, and aquaculture, and has become one of the important means of obtaining data. In the agricultural Internet of Things, sensing equipment regularly measures equipment status and environmental parameters, and the time series data collected by it has the characteristics of infinity, dynamics, and time-space correlation, and is transmitted to the data processing system in the form of data streams. However, the environment in which the agricultural IoT is deployed is relatively harsh. For example, there is a lot of dust in livestock and poultry breeding sites. The aquaculture IoT is mainly dep...

AI Technical Summary

Problems solved by technology

[0005] An object of the present invention is to provide a method and device for abnormal real-time detection and processing of agricultural Internet of Things data streams, which are used to overcome the shortcomings of insufficient data accuracy and real-time performance obtained by using existing abnormal data detection methods

Images