Central plant control system with dynamic computation reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for predicting thermodynamic states and determining operating parameters in complex HVAC systems are inefficient in terms of computational resources, making it exhaustive to predict states for multiple sets of operating parameters and compare power consumptions.
Innovation Solution
A central plant controller that dynamically reduces computation by identifying and excluding inoperable HVAC devices from the subset based on schematic relationships, allowing for efficient determination of reduced subsets of HVAC devices for which operating parameters are needed, thereby optimizing system performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional non-linear solver is used to predict thermodynamic states for multiple candidate sets of operating parameters, then complete thermodynamic analysis is achieved, but computational resources (processor usage and memory) are excessively consumed
Solution Approach 1:
The system segments the HVAC system into independent subplants (e.g., chiller subplant, boiler subplant, cooling tower subplant). Each subplant can be analyzed separately using non-linear solvers only when necessary, rather than analyzing the entire system as one large coupled system. This segmentation reduces the computational burden while maintaining thermodynamic prediction accuracy for each subplant.
Solution Approach 2:
The system applies partial action by using non-linear solvers only for specific subplants or specific operating conditions rather than for all HVAC devices at all times. For example, non-linear thermodynamic analysis is applied only to subplants where it is necessary to determine operating parameters, while other subplants use simplified models or predefined relationships, reducing overall computational resource consumption.
2Reliability
If full thermodynamic states are predicted for all HVAC devices in a complex arrangement, then complete system analysis is obtained, but the process becomes computationally exhaustive and inefficient
Solution Approach 1:
The HVAC system is divided into multiple independent subplants, each with its own control system that can determine operating parameters independently. This segmentation allows the control system to focus computational resources on individual subplants rather than analyzing the entire complex system simultaneously, improving computation efficiency while maintaining reliable analysis for each subplant.
Solution Approach 2:
The system uses preliminary simplified models or predefined thermodynamic relationships to quickly assess subplant operating conditions before applying full non-linear thermodynamic analysis only when necessary. This preliminary action filters out cases where complete analysis is not needed, improving overall computational efficiency while ensuring reliability when full analysis is applied.
3Ease of operation
If operating parameters are determined for all HVAC devices, then complete control optimization is achieved, but computational time and resources are wasted on inoperable or already-determined devices
Solution Approach 1:
The control system applies partial action by determining operating parameters only for subplants or devices where it is necessary to do so. Inoperable devices or devices with already-determined parameters are excluded from the optimization process. This approach maintains complete control optimization for the operational portion of the system while avoiding wasted computational time on devices that do not require parameter determination.
4Measurement precision
If non-linear solver is applied to complex HVAC system arrangements, then accurate thermodynamic predictions are obtained, but processor usage and memory consumption increase significantly
Solution Approach 1:
The complex HVAC system is segmented into smaller, independent subplants. Each subplant is analyzed separately using non-linear solvers, which reduces the complexity of the computational system compared to analyzing the entire HVAC system as one large coupled system. This segmentation maintains thermodynamic prediction accuracy for each subplant while reducing processor usage and memory consumption.
Data Source
AI summary
A controller for a plurality of heating, ventilation, or air conditioning (HVAC) devices includes a processing circuit that includes one or more processors and memory. The controller detects a change in condition that affects an operating status of a first HVAC device of the plurality of HVAC devices. The controller uses schematic relationships between the plurality of HVAC devices to determine a reduced subset of the plurality of HVAC devices for which operating parameters are to be generated based on the operating status of the first HVAC device. The controller generates operating parameters for the reduced subset of the plurality of HVAC devices and operates the plurality of HVAC devices using the operating parameters.