Adaptive Sensor Data Storage in Fluid Conveyance Networks

The patent segments data management by creating separate data lakes for different data types (sensor data, maintenance data, operational data) and organizing them into structured zones (raw data zone, processed data zone, archived data zone). This segmentation allows independent management of each data category, reducing overall system complexity while handling large quantities of sensor data from multiple sources.

The patent introduces data lakes as intermediary storage structures between sensor data collection and analysis systems. These data lakes serve as buffering intermediaries that temporarily hold, organize, and pre-process data before it is consumed by analytical applications, thereby simplifying the interface between data sources and data consumers.

The patent implements dynamic sensing configurations where sensors can be selectively activated or deactivated based on real-time operational conditions. The system adjusts which sensors are active, what sampling rates are used, and which data parameters are collected based on current system state, thereby adapting to variance while maintaining operational simplicity through automated control.

The patent changes operational parameters such as sampling rate, data collection frequency, and sensor activation status based on system conditions. During normal operation, lower sampling rates are used to reduce data volume, while during critical events or anomalies, the system automatically increases sampling rates and activates additional sensors, providing adaptability without requiring complex manual reconfiguration.

The patent performs preliminary data processing actions during the data collection and storage phase. Data is organized, tagged, and pre-filtered as it enters the data lake, with metadata generated and initial quality checks performed. This preliminary processing reduces the burden on later analysis systems and enables faster retrieval and processing of relevant data when needed.

The patent implements continuous data processing pipelines that operate concurrently with data collection. Data flows continuously through processing stages including filtering, aggregation, and preliminary analysis, rather than being batch-processed later. This continuous processing reduces overall time loss by eliminating idle periods between data collection and analysis initiation.

The patent segments analysis functions between edge devices and central systems. Simple, real-time analysis is performed at the edge using embedded processors and sensors, providing immediate monitoring and alerting. Complex, precision analysis is performed centrally on aggregated data batches. This segmentation enables both real-time responsiveness and high-precision analysis without requiring all data processing to occur at a single location.

The patent adds a spatial dimension to data processing by distributing analysis capabilities across multiple locations (edge devices, local gateways, central data lakes). Instead of a single centralized processing point, the system creates a multi-dimensional processing architecture where different types of analysis occur at different spatial levels, enabling parallel processing that improves both real-time responsiveness and overall analysis precision.