Multi-Site Building Management for Normalized Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
It is challenging to identify and address operational inefficiencies across multiple building sites due to differences in environmental conditions, as existing methods struggle to make apples-to-apples comparisons and isolate inefficiencies not attributed to environmental factors.
Innovation Solution
A building management system that collects and normalizes sensor data from multiple sites, compares it with data from similar sites to identify anomalies, and provides recommendations for improvement, while accounting for outdoor conditions and building characteristics, using a local or remote controller and cloud server to control environmental conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data from multiple buildings is compared directly without normalization, then comparison can be performed quickly, but the comparison is inaccurate due to environmental differences
Solution Approach 1:
The patent applies parameter changes by normalizing sensor data through mathematical transformations that adjust for environmental variables. The system changes the parameters of the data representation by introducing normalized metrics that account for outdoor temperature, building characteristics, and operational conditions, enabling accurate comparisons across diverse building portfolios without requiring complex physical measurement adjustments.
Solution Approach 2:
The patent uses an intermediary approach by introducing a cloud-based platform that acts as a mediator between local building sensors and the analysis system. This intermediary normalizes and standardizes data from multiple buildings before comparison, separating the complexity of normalization from the comparison function and enabling accurate multi-building efficiency analysis.
2Measurement precision
If normalization accounting for outdoor conditions and building characteristics is applied, then accurate anomaly identification is achieved, but data processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing normalization factors for different building types and environmental conditions. The system prepares normalized comparison benchmarks in advance, so when actual sensor data arrives, the comparison can be performed quickly against pre-established norms rather than performing full normalization calculations in real-time.
Solution Approach 2:
The patent segments the normalization process into distinct modules: environmental condition adjustment, building characteristic adjustment, and operational parameter adjustment. This segmentation allows the system to apply only the necessary normalization steps for each specific comparison scenario, reducing overall processing time while maintaining accuracy.
3Reliability
If all sensor data from multiple buildings is collected and analyzed centrally, then comprehensive anomaly detection is achieved, but system complexity and data transmission requirements increase
Solution Approach 1:
The patent segments the building management system into independent local building management systems and a centralized analytics platform. Each building operates autonomously with its own sensors and control, while the centralized platform receives only normalized summary data for comparative analysis. This segmentation reduces data transmission requirements and simplifies the centralized system while maintaining comprehensive anomaly detection capabilities.
Data Source
AI summary
Methods and systems for controlling a building. An illustrative method includes receiving sensor data from one or more sensors of the building, using the received sensor data to control one or more building management components to control one or more environmental conditions within the building, normalizing the sensor data and storing the normalized sensor data, comparing the normalized sensor data with normalized sensor data from one or more other buildings to identify one or more anomalies associated the building, and providing a recommended action to improve at least one of the one or more identified anomalies of the building.