Method and apparatus for measuring and improving efficiency in refrigeration systems
An apparatus for optimizing an efficiency of a refrigeration system, comprising means for measuring a refrigeration efficiency of an operating refrigeration system; means for altering a process variable of the refrigeration system during efficiency measurement; and a processor for calculating a process variable level which achieves an optimum efficiency. The process variables may include refrigerant charge and refrigerant oil concentration in evaporator.
1. A method for control, comprising:
providing a control having an automated processor receiving at least one sensor input and producing at least one control output, and having at least one memory;
segmenting an operational space of the control into a plurality of regions selected based on at least the plurality of sensor inputs, each region being associated with data stored in the memory representing a respective model of that region corresponding to at least a sensitivity to temporally related variations in the at least one sensor input, to the at least one output, and a respective statistical variability of that region;
adaptively improving the model of a respective region in dependence on at least one of:
(a) a statistical variability of the respective region during operation of the control, wherein regions with high statistical variability are subject to increased searching through variation of the at least one output to define different portions of the respective region, as compared to regions of low statistical variability, and concurrently monitoring the at least one input as a at least a basis for adaptively improving the model of the respective region, and
(b) a filling density of input-output relationships of the respective region, wherein a region with a low filling density of the operational space is explored to increase a respective filling density and determine an optimum state of the at least one output, and a region with a high filling density is explored less than a region of low filling density, and correspondingly transitioned more directly to a most efficient condition of the at least one output; and
storing data defining the adaptively improved model in the at least one memory.
2. The method according to claim 1 , wherein the at least one output controls a refrigerant charge level of a refrigeration system.
3. The method according to claim 1 , wherein the at least one output controls a compressor speed of a refrigeration system.
4. The method according to claim 1 , wherein the at least one output controls a refrigerant oil concentration in an evaporator of a refrigeration system.
5. The method according to claim 1 , wherein the model is adaptively improved in dependence on a filling density of the respective region.
6. The method according to claim 1 , wherein the model is adaptively improved in dependence on a statistical variability of the respective region.
7. The method according to claim 1 , wherein the automated processor analyzes the at least one input to determine a correlation between a state of the at least one input with a system efficiency.
8. The method according to claim 1 , wherein a model for a respective region comprises a linear model relating the at least one input to the at least one output.
9. The method according to claim 1 , wherein a model for a respective region comprises a non-linear model relating the at least one input to the at least one output.
10. The method according to claim 1 , wherein the automated control has a plurality of different modes, comprising at least a system startup mode, a system shutdown mode, a system operation mode, and a system failure mode.
11. An automated control system, comprising:
an automated processor configured to receive at least one sensor input and produce at least one control output;
at least one memory;
said automated processor being adapted to:
segment an operational space into a plurality of regions based on at least the plurality of sensor inputs, each region being associated with data stored in the memory representing a respective model of that region corresponding to at least a sensitivity to temporally related variations in the at least one sensor input, to the at least one output, and a respective statistical variability of that region;
adaptively improve the model of a respective region in dependence on at least one of:
a. a statistical variability of the respective region during operation of the control, wherein regions with high statistical variability are subject to increased searching, through variation of the at least one output to define different portions of the region, as compared to regions of low statistical variability, and concurrently monitoring the at least one input as at least a basis for the adaptive improvement of the model of the respective region, and
b. a filling density of the respective region, wherein regions with a low filling density of the operational space are explored to increase a respective filling density and to determine an optimum state of the at least one output, and regions with a high filling density are explored less than regions with a high filling density and correspondingly transitioned more directly to a most efficient condition of the at least one output; and
store data defining the adaptively improved model in the at least one memory.
12. The system according to claim 11 , wherein the at least one output controls a refrigerant charge level of a refrigeration system.
13. The system according to claim 11 , wherein the at least one output controls a compressor speed of a refrigeration system.
14. The system according to claim 11 , wherein the at least one output controls a refrigerant oil concentration in an evaporator of a refrigeration system.
15. The system according to claim 11 , wherein the memory stores data representing a filling density of the operational space of a region.
16. The system according to claim 11 , wherein the automated processor adaptively improves the model in dependence on a filling density of the respective region.
17. The system according to claim 11 , wherein the automated processor adaptively improves the model in dependence on a statistical variability of the respective region.
18. The system according to claim 11 , wherein the automated processor analyzes the at least one input to determine a correlation between a state of the at least one input with a system efficiency.
19. The system according to claim 11 , wherein the automated processor defines a model for a respective region comprising a non-linear model relating the at least one input to the at least one output.
20. The system according to claim 11 , wherein the automated processor is configured to control operation according to a plurality of different modes, comprising at least a startup mode, a shutdown mode, an operation mode, and a failure mode.