HVAC distribution system identification

• US 5,579,993 A
• Filed: 01/06/1995
• Issued: 12/03/1996
• Est. Priority Date: 01/06/1995
• Status: Expired due to Term

• Current Assignee: Siemens Industry Incorporated (Siemens AG)
• Original Assignee: Landis & Gyr Powers, Inc. (Landis+Gyr Group AG)
• Inventors: Ahmed, Osman, Klein, Sanford A., Mitchell, John W.
• Primary Examiner(s): Wayner, William E.
• Application Number:
  US08/369,781
• Time in Patent Office: 697 Days
• Field of Search: 236/49.3, 236/78 D, 165/22, 395/61, 395/21, 364/165
• US Class Current: 236/49.3
• CPC Class Codes: F24F 11/30 for purposes related to the operation of the system, e.g. for safety or monitoring
• CPC Class Codes: F24F 11/46 Improving electric energy efficiency or saving
• CPC Class Codes: F24F 11/54 using one central controller connected to several sub-controllers
• CPC Class Codes: F24F 11/63 Electronic processing
• CPC Class Codes: F24F 11/64 using pre-stored data
• CPC Class Codes: F24F 11/77 by controlling the speed of ventilators
• CPC Class Codes: G05B 13/027 using neural networks only
• CPC Class Codes: Y10S 706/914 Process plant

Abstract

A controller implemented in a heating, ventilation and air-conditioning (HVAC) distribution system provides improved control by implementing a general regression neural network to generate a control signal based on identified characteristics of components utilized within the HVAC system. The general regression neural network utilizes past characteristics and desired (calculated) characteristics to generate an output control signal. The general regression neural network is implemented in a feedforward process, which is combined with a feedback process to generate an improved control signal to control components within the HVAC distribution system. Implementation of the general regression neural network is simple and accurate, requires no input from an operator supervising the controller, and provides adaptive, real-time control of components within the HVAC distribution system.

Claims

1. A method of controlling a component implemented in a heating, ventilation and air-conditioning (HVAC) distribution system, the method comprising the steps of:

identifying past characteristics of the component of the HVAC distribution system;

calculating desired characteristics of the component of the HVAC distribution system; and

generating, based on the past and desired characteristics of the component, a control signal to control the component of the HVAC distribution system utilizing a general regression neural network.

2. The method of claim 1 wherein said step of identifying past characteristics further comprises storing the past characteristics in a look-up table.

3. The method of claim 2 wherein said step of identifying past characteristics further comprises identifying past characteristics based on values of measured system variables.

4. The method of claim 1 wherein said component further comprises one of the group of a fan, a pump, a damper, a fume hood, a supply box, or a general exhaust box.

5. A method of controlling a component implemented in a heating, ventilation and air-conditioning (HVAC) distribution system, the method comprising the steps of:

storing values of past characteristics in a look-up table;

generating, by a general regression neural network, an output signal based on certain values of the past characteristics stored in the look-up table;

controlling the component of the HVAC distribution system utilizing the output signal generated by the general regression neural network.

6. The method of claim 5 wherein said component further comprises one of the group of a fan, a pump, a damper, a fume hood, a supply box, or a general exhaust box.

7. An apparatus for controlling a component implemented in a heating, ventilation and air-conditioning (HVAC) distribution system, the apparatus comprising:

means for identifying past characteristics of the component of the HVAC distribution system;

means for calculating desired characteristics of the component of the HVAC distribution system; and

means, based on the past and desired characteristics of the component, for generating a control signal to control the component of the HVAC distribution system utilizing a general regression neural network.

8. The apparatus of claim 7, wherein said means for identifying a control signal further comprises means for storing the values of the past characteristics in a look-up table for use by the general regression neural network.