Extremum-seeking control system for a chilled water plant
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional extremum-seeking control systems in chilled water plants face challenges due to the use of periodic dither signals, which cause noticeable oscillations and require careful frequency selection based on the plant's natural frequency, making it difficult to optimize condenser water temperature setpoints effectively without knowledge of the plant dynamics.
Innovation Solution
An extremum-seeking control system using a stochastic excitation signal, such as a random walk signal, to perturb the condenser water temperature setpoint, estimate the gradient of power consumption, and modulate the setpoint to drive the gradient to zero, eliminating the need for frequency selection and reducing noticeable oscillations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a periodic dither signal is used to perturb the control input, then the extremum-seeking control can optimize system performance, but large variations and noticeable oscillations occur in both the control input and plant output
Solution Approach 1:
The patent changes the fundamental parameter of the dither signal from periodic to aperiodic. By using an aperiodic signal instead of a periodic one, the system maintains the exploration capability needed for extremum-seeking control while eliminating the regular oscillations that are noticeable to operators. This parameter change transforms the signal characteristics to resolve the contradiction between optimization effectiveness and operational smoothness.
2Measurement precision
If a periodic dither signal is used, then gradient estimation can be performed, but careful frequency selection is required based on plant dynamics which is difficult without knowledge of plant behavior
Solution Approach 1:
The patent extracts the frequency selection requirement from the control system design. By using an aperiodic dither signal, the system removes the need to match the dither frequency with plant natural frequencies. This extraction eliminates the complexity of frequency tuning while preserving the gradient estimation capability, as the aperiodic signal still provides sufficient excitation for accurate gradient calculation without requiring frequency matching.
Data Source
AI summary
An extremum-seeking control system for a chilled water plant includes a feedback controller and an extremum-seeking controller. The feedback controller is configured to operate equipment of the chilled water plant to achieve a condenser water temperature setpoint. The equipment include at least one of a chiller compressor, a condenser water pump, and a cooling tower fan. The extremum-seeking controller is configured to provide the condenser water temperature setpoint to the feedback controller. The extremum-seeking controller is configured to determine an optimal value for the setpoint condenser water temperature setpoint by perturbing the condenser water temperature setpoint with a stochastic excitation signal, monitoring a power consumption of the equipment resulting from the perturbed condenser water temperature setpoint, estimating a gradient of the power consumption with respect to the condenser water temperature setpoint, and modulating the condenser water temperature setpoint to drive the estimated gradient to zero.