Implementing a computational fluid dynamics model using a plurality of computation units

1. An apparatus comprising:

a plurality of computation units configured to implement a one-dimensional computational fluid dynamics model for controlling a physical system, wherein each computation unit is associated with a respective node of the one-dimensional computational fluid dynamics model, wherein the plurality of computation units are configured to operate in parallel;

wherein the plurality of computation units are configured to implement the one-dimensional computational fluid dynamics model to receive a boundary condition signal for the one-dimensional computational fluid dynamics model at a time t 1 , wherein the boundary condition signal represents a first physical variable sampled at a first location in the physical system; and

wherein the plurality of computation units are configured to implement the one-dimensional computational fluid dynamics model to generate an output signal representing a second physical variable at a second location in the physical system, wherein the first physical variable and the second physical variable are different physical variables, wherein the second physical variable is not sampled in the physical system, and wherein the output signal is usable for controlling the physical system;

wherein the plurality of computation units are configured to implement the one-dimensional computational fluid dynamics model to generate the output signal before a second boundary condition signal is received at a time t 2 .

2. The apparatus of claim 1 , wherein the difference between the time t 2 and the time t 1 is between ten microseconds and ten milliseconds.

3. The apparatus of claim 1 , wherein the apparatus is implemented as a field programmable gate array that implements the plurality of computation units.

4. The apparatus of claim 1 , wherein the physical system is an engine, wherein the apparatus is implemented as at least a portion of an engine control unit that controls the engine.

5. The apparatus of claim 1 , wherein each parallel computation unit comprises:

a logic unit;

a memory unit; or

a math unit.

6. The apparatus of claim 1 , wherein the plurality of computation units are configured to perform hardware-in-the-loop testing of a controller, wherein the boundary condition signal is received from the controller, and wherein the output signal is provided to the controller.

7. The apparatus of claim 1 , further comprising:

a controller for controlling the physical system, wherein the controller is configured to provide the boundary condition signal to the plurality of computation units and receive the output signal from the plurality of computational units to perform hardware-in-the-loop testing.

8. The apparatus of claim 1 , wherein the plurality of computation units are also configured to implement a model to process the boundary condition signal prior to input to the computational fluid dynamics model.

9. The apparatus of claim 1 , wherein the plurality of computation units are configured to implement a plurality of physical system models, wherein the plurality of computation units are configured to process the boundary condition signal to generate the output signal using the plurality of physical system models.

10. The apparatus of claim 1 , wherein the boundary condition signal is sampled by a sensor coupled to the physical system.

11. The apparatus of claim 10 , wherein the sensor is sampled at a rate of between 100 Hz and 100 kHz.

12. A method, comprising:

receiving an input signal at time t 1 , wherein the input signal represents a first physical variable sampled at a first location in a physical system;

processing the input signal by a plurality of computation units to generate an output signal, wherein the plurality of computation units implement a one-dimensional computational fluid dynamics model for controlling the physical system, wherein each computation unit is associated with a respective node of the one-dimensional computational fluid dynamics model, wherein said processing the input signal to generate the output signal is based on the one-dimensional computational fluid dynamics model, wherein said processing comprises the plurality of computation units operating in parallel, wherein the output signal represents a second physical variable at a second location in the physical system, wherein the first physical variable and the second physical variable are different physical variables, wherein the second physical variable is not sampled in the physical system, wherein the output signal is usable for controlling the physical system, and wherein said processing comprises generating the output signal before receiving a second input signal representing the first physical variable sampled at the first location in the physical system at time t 2 .

13. The method of claim 12 , wherein the difference between the time t 2 and the time t 1 is between ten microseconds and ten milliseconds.

14. The method of claim 12 , wherein the plurality of computation units are implemented on a field programmable gate array.

15. The method of claim 12 , wherein the physical system is an engine, wherein said receiving and said processing is performed to control the engine.

16. The method of claim 12 , wherein each parallel computation unit comprises:

a logic unit;

a memory unit; or

a math unit.

17. The method of claim 12 , wherein said receiving the input signal and said providing the output signal is performed to implement hardware-in-the-loop testing of a controller, wherein the input signal is received from the controller, and wherein the output signal is provided to the controller.

18. The method of claim 12 , wherein said processing the input signal comprises utilizing one or more physical models to process the input signal prior to input to the computational fluid dynamics model.

19. The method of claim 12 , wherein the input signal is sampled by a sensor coupled to the physical system.

20. The method of claim 19 , wherein the sensor is sampled at a rate of between 100 Hz and 100 kHz.