Dual control with dual sensor averaging and substitution

• US 5,436,826 A
• Filed: 10/20/1993
• Issued: 07/25/1995
• Est. Priority Date: 10/20/1993
• Status: Expired due to Term

• Current Assignee: United Technologies Corporation (Raytheon Technologies Corporation d/b/a RTX)
• Original Assignee: United Technologies Corporation (Raytheon Technologies Corporation d/b/a RTX)
• Inventors: O'Flarity, Linda
• Primary Examiner(s): Envall, Jr., Roy N.
• Assistant Examiner(s): BROWN, THOMAS
• Application Number:
  US08/139,485
• Time in Patent Office: 643 Days
• Field of Search: 364/184-187, 371/8.1, 371/8.2, 371/9.1, 371/14, 371/24
• US Class Current: 700/79
• CPC Class Codes: G05D 1/0077 using redundant signals or controls

Abstract

A gas turbine engine with dual fuel controls is disclosed. The controls have individual sensors that monitor N2. Each control uses the average value of N2 from both sensors for providing fuel to the engine and performing other operations. The fuel controls contain signal processors capable of determining if their respective sensors are operating out of range, in which case the fuel control uses the value of N2 from sensor associated with the other fuel control. If both sensors are out of range, the signal processors use a stored value for N2 or a value synthesized from another engine parameter. The signal processors are also capable of retesting the sensor over successive cycles to see if the sensor has corrected and is within a tighter range. If the difference between sensor values exceeds a range, the sensor with the best value for N2 is used.

Claims

1. A control system comprising first and second controllers, a sensor connected to the first controller, a sensor connected to the second controller, and the sensors each producing a signal representing the magnitude of a common parameter used by each of the controllers to produce an output, the output from each controller being summed to provide a control signal, characterized by:

each controller comprising a signal processor, and a communication bus between each controller, each signal processor comprising means for producing a controller output signal based on the average value of the signals produced by the sensors, for producing a fault signal indicating that a sensor connected to the controller produces a signal outside of an acceptable range, for removing the fault signal if a sensor that causes the fault single produces a signal with a value not greater than an acceptable value, less than the maximum value for said acceptable range, during a plurality of successive tests of the sensor by the signal processor and for using a signal produced by a sensor connected to the other controller in response to said fault signal; and

means for providing data communication between signal processors in each controller.

2. The control system described in claim 1, further characterized in that the fault signal is produced when the local sensor signal exceeds a maximum value by a first percentage and the signal processor comprise means for removing the fault signal if a sensor produces a signal that does not exceed a said maximum value by a second percentage less than the first percentage over a plurality of successive sample of the sensor signal.

3. The control system described in claim 2, further characterized in that each signal processor comprises means for selecting a sensor connected to the other controller when a difference between the values for both sensors exceeds a stored level.

4. The control system described in claim 3, further characterized in that each signal processor comprises means for using a stored value in place of the local sensor signal when both signal processors produce a fault signal.

5. A gas turbine control system comprising two fuel controls each connected to an associated engine sensor providing a signal with a value indicating the magnitude of the same engine operating parameter, characterized in that:

each fuel control contains a signal processor and signals processors are in communication with each other;

each signal processor comprises means for providing a fault signal indicating that a local sensor in communication with the signal processor is providing a single value at an out of range value exceeding an acceptable value, for removing the fault signal if the value for the local sensor signal is within an acceptable range for a plurality of successive samples, the range being less than a range defined by the difference between the acceptable value and the out of range value for using the local sensor in communication with the other signal processor in response to the fault signal and for using the average value from both sensors in the absence of the fault signal.

6. The gas turbine engine described in claim 5, further characterized in that each signal processor comprises means for selecting the local sensor with a signal value closest to a preset value for other engine operating conditions when the difference between the signal values between the local sensors exceeds a stored level.

7. The gas turbine engine described in claim 6, further characterized in that the signal processors comprise means for using a stored value for the value of the parameter when both signal processors produce the fault signal for the respective local sensor.

8. A method for using two controls, each producing a component in a summed output to provide system control based on the output of two sensors that provide signals representing the magnitude of a common parameter, each sensor supplying the output to one of the two controls, characterized by the steps of:

each control providing a fault signal if the sensor connected to it is providing a signal exceeding an acceptable level and removing the fault signal after successive tests of the sensor producing the fault signal, if each test demonstrating that the output from the sensor meets a more stringent signal level test that the level that caused the production of the fault signal;

communicating the fault signal to the other control to cause the other control to provide, to the control producing the fault signal, the signal value for the sensor connected to said other control which uses said signal value to produce an output; and

in the absence of the fault signal, each control using the average of the sensor outputs to produce an output.

9. The method described in claim 8, further characterized by the step of using default value for the sensor outputs is used in both controls produce a fault signal.

10. The method described in claim 8, further characterized by the step of comparing the levels of the outputs from the sensors and using the output form the sensor with a value closest to a stored level, if the outputs differ by more than a certain amount.