Central Plant Control Using Graph-Based Thermodynamic State Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for predicting thermodynamic states in complex HVAC systems are inefficient in terms of computational resources, making it exhaustive to determine optimal operating parameters for minimizing power consumption.
Innovation Solution
A system and method that utilize a processing circuit to identify a reduced subset of thermodynamic states based on HVAC device connections, predicting these states using a non-linear solver, and determining optimal operating parameters to operate HVAC devices efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional non-linear solver is used to predict thermodynamic states for all HVAC devices, then complete thermodynamic analysis is achieved, but computational resource consumption increases significantly
Solution Approach 1:
The HVAC system is segmented into multiple subplants based on device connections and thermodynamic relationships. The non-linear solver is applied selectively to each subplant rather than the entire system, reducing the computational scope while maintaining prediction accuracy for each segment. This segmentation allows the system to process thermodynamic states in manageable portions, significantly lowering overall computational resource consumption.
Solution Approach 2:
Different computational approaches are applied to different parts of the HVAC system based on their specific characteristics. The non-linear solver is used locally only where necessary (in specific subplants requiring detailed thermodynamic analysis), while other areas use simplified methods. This local quality approach ensures accurate predictions where needed while avoiding unnecessary computational overhead in other regions of the system.
2Productivity
If multiple candidate sets of operating parameters are evaluated to minimize power consumption, then optimal operating parameters are determined, but computational time and resources increase
Solution Approach 1:
The evaluation of operating parameters is segmented across multiple subplants. Each subplant's candidate parameter sets are evaluated independently using the non-linear solver, allowing parallel processing and reducing overall computational time. This segmented approach enables the system to determine optimal parameters for each subplant separately, then combine results to achieve system-wide optimization without the exhaustive computational burden of evaluating all parameters simultaneously across the entire system.
3Reliability
If full thermodynamic states are computed for complex HVAC arrangements, then complete system analysis is achieved, but device complexity and computational overhead increase
Solution Approach 1:
The complex HVAC system is divided into multiple simpler subplants based on device connections and thermodynamic relationships. Each subplant is analyzed separately using the non-linear solver, which reduces the complexity of individual computational tasks while maintaining overall system analysis completeness. This segmentation transforms a single complex computational problem into multiple manageable subproblems that can be solved more efficiently.
Solution Approach 2:
After individual subplant analyses are completed, the results are merged to form the complete system analysis. The thermodynamic states and operating parameters from each subplant are combined to determine overall system performance and optimal operation. This merging approach maintains analytical completeness while avoiding the complexity of analyzing the entire system as a single unit.
Data Source
AI summary
A controller for an energy plant includes a processing circuit having a processor and memory which stores instructions executed by the processor. The processing circuit is configured to identify, from a plurality of thermodynamic states affected by a plurality of heat, ventilation, and air conditioning (HVAC) devices, a reduced subset of the plurality of thermodynamic states to be predicted based on connections between the plurality of HVAC devices. The processing circuit is configured to predict values of the reduced subset of the plurality of thermodynamic states and operate the plurality of HVAC devices based on the predicted values of the reduced subset of the plurality of thermodynamic states.