Internet-of-things equipment behavior analysis method based on Flink

The invention provides an Internet-of-things equipment behavior analysis method based on Flink. The invention belongs to the technical field of Internet-of-things, according to the method, the data ofthe equipment are collected based on the Internet-of-things platform and are sent to the MQTT; the flink obtains equipment data from the MQTT and sends the equipment data to the MQTT; firstly, data cleaning is carried out to obtain required characteristic values; the acquired feature value is calculated by using a machine learning algorithm to obtain a device behavior model; analyzing, screeningand aggregating are performed on the device behavior model through a flink time window mechanism to obtain a device behavior analysis result, the device behavior analysis result is stored in an HDFS,and dynamically displaying the device behavior analysis result in real time. Wherein the characteristic value is determined by an actual demand. The finally formed equipment behavior analysis result can be used for equipment fa