Fuzzy Analysis of Ground Humidity Sensor Readings on a Crop Area Managed Using IoT-Technology
摘要
Most modern IoT software boards support real-time analytics – stream aggregation, filtering, operations with the accumulated data set. Interactive analytics of data accumulated in IoT-platforms is implemented using predictive models of time series based on various methods of statistical analysis and machine learning. Another approach to time series forecasting is based on a fuzzy paradigm, which assumes a weakly structured data generated in the IoT-platform in the form of aggregation of sensor readings from web-devices. The paper discusses a method of fuzzy modeling and forecasting of segments of averaged sensor readings from web-devices capable of providing information support for predictive and prescriptive analytical solutions based on a unified IoT-information platform. As an example of the application of IoT technology, a crop area was selected in the Zangelan region of Azerbaijan, where various web-devices are installed, the sensors of which monitor climatic conditions, air and soil humidity, the ground chemical composition, the state of vegetation, etc.