Building Data Hierarchy Batching for Faster BMS Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large building management systems (BMS) with complex hierarchies, the computation of metrics across multiple spaces and time aggregations becomes cumbersome, leading to slow user interfaces, delayed key performance indicator calculations, and difficulty in applying business rules.
Innovation Solution
A BMS that includes a space hierarchy database to store sibling relationships among points, a batch metrics engine to aggregate data samples from sibling points, and a controller to adjust building equipment operations based on calculated aggregate metrics, facilitating efficient metric calculation and user interface updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional metric computation methods are used in large BMS with complex hierarchies, then comprehensive metrics can be calculated across all spaces and time aggregations, but the computation becomes cumbersome and slow, resulting in poor user interface performance and delayed key performance indicator calculations
Solution Approach 1:
The patent segments the building management system into a hierarchical structure with multiple levels (building, floor, zone, space) and introduces specialized components at each level including space hierarchy databases, batch metrics engines, and real-time metrics engines. This segmentation allows metric calculations to be distributed and performed at appropriate levels, improving overall calculation speed while managing system complexity through modular organization.
Solution Approach 2:
The patent implements preliminary action by pre-calculating and storing aggregated metrics in space hierarchy databases before they are needed. The batch metrics engine performs preliminary aggregation of data samples from multiple points, and these pre-computed metrics are stored for rapid retrieval. This eliminates the need to recalculate metrics from raw data each time they are requested, significantly improving response time.
2Loss of time
If metrics are calculated in real-time for every user interface update, then the most current data is available, but the computation becomes cumbersome and slow, affecting user interface performance
Solution Approach 1:
The system performs preliminary aggregation of metric data in batch processes and stores these pre-calculated metrics in space hierarchy databases. When the user interface needs metric data, it retrieves pre-computed values rather than calculating them in real-time, dramatically reducing retrieval time and improving interface responsiveness.
Solution Approach 2:
The patent introduces space hierarchy databases as intermediary storage between the raw data collection layer and the user interface layer. These databases store pre-aggregated metrics that serve as intermediaries, allowing the user interface to query consolidated metric data without triggering complex real-time calculations across the entire system hierarchy.
3Measurement precision
If comprehensive data aggregation is performed across all sibling points and spaces, then accurate aggregate metrics are obtained, but the computation becomes cumbersome and slow
Solution Approach 1:
The patent segments data aggregation into hierarchical levels where sibling points are aggregated to space-level metrics, spaces are aggregated to zone-level metrics, zones to floor-level metrics, and floors to building-level metrics. The space hierarchy database stores sibling relationships and enables segmented aggregation at each level, maintaining accuracy while reducing overall computational complexity through distributed processing.
Solution Approach 2:
The system performs partial aggregation by calculating metrics at appropriate hierarchical levels rather than aggregating all data from every sibling point at the highest level. The batch metrics engine aggregates data samples from sibling points to generate space-level metrics, and only performs additional aggregation when needed for higher levels, avoiding unnecessary computation while maintaining accuracy.
Data Source
AI summary
A building management system includes a plurality of meters configured to provide data samples of a plurality of points relating to a building and a space hierarchy database configured to store a sibling relationship for each of the points. Each sibling relationship identifies two or more of the points as sibling points. The building management system includes a batch metrics engine configured to receive a first data sample of a first point, access the sibling relationship for the first point to identify one or more sibling points of the first point, aggregate the first data sample with data samples of the sibling points to generate a batch, and calculate an aggregate metric using the first data sample and the other data samples in the batch. The building management system also includes a controller configured to adjust an operation of building equipment based on the aggregate metric.