Building Automation Parallel Computation Engine for Faster HVAC Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current HVAC system control algorithms require iterative rebuilding and testing, leading to inefficient development and deployment processes, as changes often necessitate rebuilding the entire algorithm, which is time-consuming and inefficient.
Innovation Solution
A method for executing computations in parallel within a building management system, utilizing computing threads to determine property values based on dependencies, allowing for efficient generation and deployment of control algorithms, and optimizing resource allocation across multiple nodes for cost-effective and reliable operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If iterative rebuilding and testing of control algorithms is performed, then algorithm functionality can be improved, but development time and efficiency deteriorate
Solution Approach 1:
The patent segments the control algorithm into independent computing threads that can be executed in parallel. Each thread handles specific computational tasks (e.g., reading sensor data, processing control logic, actuating devices) independently, allowing simultaneous execution without requiring complete algorithm rebuilds for testing individual components.
Solution Approach 2:
The system performs preliminary actions by pre-compiling and storing computing threads in a thread pool before actual control execution. This allows algorithms to be prepared and validated in advance, with individual threads that can be independently modified and tested without affecting the entire system, thus reducing iterative development time.
2Productivity
If parallel computing threads are executed, then computational speed is improved, but system complexity increases
Solution Approach 1:
The patent introduces an intermediary controller that manages the parallel computing threads. This controller coordinates thread execution, handles synchronization, and manages the thread pool, thereby abstracting the complexity of parallel computation from the individual control algorithms while maintaining high computational speed.
Solution Approach 2:
The system creates multiple copies of computing threads that can be executed simultaneously in parallel. Each thread is an independent copy that performs the same or different computational tasks, enabling faster processing through parallel execution while the thread management system handles the complexity of coordination.
3Adaptability or versatility
If entire algorithm rebuilds are performed for changes, then algorithm updates can be implemented, but deployment efficiency deteriorates
Solution Approach 1:
The patent extracts individual computing threads from the complete control algorithm, allowing specific threads to be modified, updated, or replaced independently. This enables targeted algorithm updates without requiring rebuilds of the entire algorithm, significantly improving deployment efficiency while maintaining adaptability.
Solution Approach 2:
The system implements dynamic algorithm updates by allowing computing threads to be added, removed, or modified in the thread pool during runtime. This dynamic structure enables flexible algorithm adaptation without complete rebuilds, as individual threads can be updated independently and activated when needed.
Data Source
AI summary
A method for executing computations in parallel for a building management system of a building includes receiving a computing job request to determine values for one or more particular properties, receiving a property model indicating dependencies between a plurality of properties, the plurality of properties including the one or more particular properties, wherein the plurality of properties include building data for the building, and generating one or more computing threads based on the property model, wherein each computing thread includes a sequence of computations for determining values for the plurality of properties. The method further includes executing the computing threads in parallel to determine the values for the particular properties.