Adaptive control strategy and method for optimizing hybrid electric vehicles

• US 7,954,579 B2
• Filed: 02/04/2008
• Issued: 06/07/2011
• Est. Priority Date: 02/04/2008
• Status: Active Grant

• Current Assignee: Turntide Technologies, Inc.
• Original Assignee: Illinois Institute of Technology
• Inventors: Emadi, Ali, Lukic, Srdjan M., Rodriguez, Fernando
• Primary Examiner(s): Shriver, II; J. Allen
• Assistant Examiner(s): WALTERS, JOHN DANIEL
• Application Number:
  US12/025,376
• Publication Number:
  US 20090198396A1
• Time in Patent Office: 1,219 Days
• Field of Search: 180/65.265, 180/65.275, 180/65.28, 180/65.285, 903/930, 701/22
• US Class Current: 180/65.265
• CPC Class Codes: B60W 10/06 including control of combustion engines
• CPC Class Codes: B60W 10/08 including control of electric propulsion units, e.g. motors or generators
• CPC Class Codes: B60W 10/26 for electrical energy, e.g. batteries or capacitors
• CPC Class Codes: B60W 20/00 Control systems specially adapted for hybrid vehicles hybrid vehicle design, B60K6/00; electric vehicles B60L
• CPC Class Codes: B60W 20/11 using model predictive control [MPC] strategies, i.e. control methods based on models predicting performance utilising navigation and traffic information in the control strategy B60W20/12
• CPC Class Codes: B60W 2050/0013 Optimal controllers
• CPC Class Codes: B60W 2050/0014 Adaptive controllers
• CPC Class Codes: B60W 2050/0016 State machine analysis
• CPC Class Codes: B60W 2510/244 Charge state
• CPC Class Codes: B60W 30/1882 characterised by the working point of the engine, e.g. by using engine output chart
• CPC Class Codes: Y02T 10/40 Engine management systems
• CPC Class Codes: Y02T 10/62 Hybrid vehicles
• CPC Class Codes: Y02T 10/84 Data processing systems or methods, management, administration
• CPC Class Codes: Y10S 903/93 Conjoint control of different elements

Abstract

This invention relates a control strategy for a hybrid electric vehicle having an electric motor, a battery and an internal combustion engine. The control strategy improves fuel economy and reduces emissions while providing sufficient acceleration over a varying set of driving conditions through an adaptive control unit with an artificial neural network. The artificial neural network is trained on a pre-processed training set based on the highest fuel economies of multiple control strategies and multiple driving profiles. Training the artificial neural network includes a training algorithm and a learning algorithm. The invention also includes a method of operating a hybrid electric vehicle with an adaptive control strategy using an artificial neural network.

Claims

1. A hybrid electric vehicle including an electric motor, a battery and an internal combustion engine, the vehicle comprising:

a power control module having an adaptive control unit for interfacing with sensors and actuators of the electric motor, the battery and the internal combustion engine;

the adaptive control unit including an artificial neural network that is adapted to changing driving conditions or patterns, the artificial neural network comprising an input layer, a hidden layer and an output layer, wherein the input layer comprises between five and fifty neurons, the hidden layer comprises between five and fifty neurons and the output layer comprises between one and ten neuron; and

a combined control strategy for an entire operating range of the hybrid electric vehicle including a plurality of control strategies for a plurality of driving profiles with optimized operation points of each of the plurality of driving profiles, wherein the artificial neural network combines the optimized operation points of each of the plurality of driving profiles into the combined control strategy.

2. The vehicle of claim 1 wherein the adaptive control unit includes at least one of the group consisting of a digital signal processor, a micro-processor, a field programmable gate array and combinations thereof.

3. The vehicle of claim 1 wherein the vehicle includes at least one of the group consisting of a car, a bus, a truck, and a sport utility vehicle.

4. The vehicle of claim 1 wherein the adaptive control unit signals to at least one of the actuators to maximize fuel economy, minimize emissions, meet peak power demands and to respond to changing driving conditions.

5. The vehicle of claim 4 wherein the changing driving conditions include urban profiles, highway profiles and combinations thereof.

6. The vehicle of claim 1 wherein the fuel economy improves by at least between about 1 percent and about 7 percent above a constant causal based control scheme.

7. The vehicle of claim 1 further comprising a training set based on changing driving conditions or profiles for training the artificial neural network.

8. The vehicle of claim 7 further comprising a pre-processing algorithm to operate on the training set and reduce repetitive row vectors.

9. The vehicle of claim 7 wherein the training set includes input and output values based on highest fuel economies for various driving conditions and control schemes.

10. The vehicle of claim 1 further comprising a plug for plug-in hybrid functionality.

11. A hybrid electric vehicle including an electric motor, a battery and an internal combustion engine, the vehicle comprising:

a power control module having an adaptive control unit for interfacing with sensors and actuators of the electric motor, the battery and the internal combustion engine;

the adaptive control unit including an artificial neural network that is adapted to changing driving conditions or patterns, wherein the artificial neural network includes an input layer having about 20 neurons, a hidden layer having about 20 neurons and an output layer having about 1 neuron; and

a combined control strategy for an entire operating range of the hybrid electric vehicle including a plurality of control strategies for a plurality of driving profiles with optimized operation points of each of the plurality of driving profiles, wherein the artificial neural network combines the optimized operation points of each of the plurality of driving profiles into the combined control strategy.

12. The vehicle of claim 11 wherein the adaptive control unit includes at least one of the group consisting of a digital signal processor, a micro-processor, a field programmable gate array and combinations thereof.

13. The vehicle of claim 11 wherein the adaptive control unit signals to at least one of the actuators to maximize fuel economy, minimize emissions, meet peak power demands and to respond to changing driving conditions.

14. The vehicle of claim 13 wherein the changing driving conditions include urban profiles, highway profiles and combinations thereof.

15. The vehicle of claim 11 further comprising a training set based on changing driving conditions or profiles for training the artificial neural network.

16. The vehicle of claim 15 further comprising a pre-processing algorithm to operate on the training set and reduce repetitive row vectors.

17. A method of operating a hybrid electric vehicle comprising:

providing a hybrid electric vehicle having an adaptive control unit, wherein the adaptive control unit includes an artificial neural network;

receiving input signals from sensors of the hybrid electric vehicle;

calculating output signals in the adaptive control unit based on maximizing fuel economy, minimizing emissions, meeting peak power demands and providing acceptable transient response during changing driving conditions;

sending output signals to an internal combustion engine or an electric motor of the hybrid electric vehicle; and

training the artificial neural network with a plurality of control strategies for various driving conditions or profiles for a sufficient number of iterations and combining optimized operation points of each of the various driving conditions or profiles into a combined control strategy, wherein training includes a Levenberg-Marquardt training algorithm and a gradient decent learning algorithm with momentum weight/bias.

18. The method of claim 17 further comprising applying a pre-processing algorithm to the plurality of control strategies for reducing repetitive vectors.

19. The method of claim 17 further comprising periodically updating the step of calculating based on age or wear of drivertrain components.

20. The method of claim 17 wherein the artificial neural network signals to at least one of the actuators to maximize fuel economy, minimize emissions, meet peak power demands and to respond to changing driving conditions or profiles including urban and highway use.