Self-Organizing Sensor Data Collection for Adaptive Manufacturing

The patent segments the data collection system into multiple independent data collectors, each responsible for specific sensors or sensor groups. Each data collector operates autonomously to collect, preprocess, and manage data from its assigned sensors, reducing the complexity burden on any single component while maintaining comprehensive monitoring coverage across all industrial equipment.

The patent introduces a hierarchical data management architecture that adds temporal and spatial dimensions to data organization. Data is structured across multiple levels (sensor level, collector level, central level) and time scales (real-time, historical, aggregated), transforming the flat complex data structure into a multi-dimensional organized system that improves manageability and diagnostic capability simultaneously.

The patent implements local quality by enabling each data collector to perform preliminary data processing, filtering, and analysis locally rather than centralizing all computational tasks. This distributed approach allows real-time responsiveness at the edge of the network while reducing the computational burden on central systems, achieving real-time adaptability without proportionally increasing overall system complexity.

The patent applies preliminary action by preprocessing and analyzing data locally at data collectors before transmission to central systems. This includes filtering raw sensor data, performing initial diagnostic analysis, and preparing summarized information in advance, which reduces the complexity of real-time processing required at higher system levels while maintaining fast response capabilities.