Advanced AC Motor Control S/W Development Sang-Ho Yoon Senior Application Engineer The MathWorks

Transcription

1 Advanced AC Motor Control S/W Development Sang-Ho Yoon Senior Application Engineer The MathWorks 2012 The MathWorks, Inc. 1

2 Agenda Develop ECUs with Model-Based Design Generate Application Code for Prototyping Generate Algorithm Code for Production Case Study: Brushless Motor Control Application Questions 2

3 ECU Development Process with Model-Based Design System Requirements Sim Design with Simulation Executable Specifications Models Continuous Test and Verification System Integration and Tuning HIL System Design RP Automatic Code Generation Hardware/Software Integration PIL Sim: Simulation RP: Rapid Prototyping PCG: Production Code Generation SIL: Software-in-the-Loop Testing PIL: Processor-in-the-Loop Testing HIL: Hardware-in-the-Loop Testing Software Design RP (Embedded) Coding PCG Software Integration SIL 3

4 Simulation with Simulink Simulating plant and controller in one environment allows engineers to optimize system-level performance. Controls engineers and domain specialists can work together to detect integration issues in simulation. Embedded algorithm Simulink Model of physical system 4

5 Rapid Prototyping Generate, deploy, and tune code for a component (algorithm or controller) on a real-time simulator connected to system hardware Host Embedded algorithm Simulink Model of physical system Code Generation Tuning and logging Harness Rapid prototyping hardware (real-time simulator) Physical system 5

6 Rapid Prototyping on Embedded Processors Run the generated code in real time, tune parameters, and monitor realtime data on the same processor you plan to use in mass production, or a close equivalent to it. Host Embedded algorithm Simulink Model of physical system Code Generation Tuning and logging Harness Embedded microprocessor Physical system 6

7 IDE Link Processor-in-the-Loop Testing Use processor-in-the-loop PIL to evaluate the behavior of a candidate algorithm on the target processor. Host Embedded algorithm Simulink Model of physical system Code Generation PIL Testing on Embedded Hardware Embedded microprocessor 7

8 Hardware-in-the-Loop Testing Final test before integration using simulated plant executing in real time. Host Embedded algorithm Simulink Model of physical system Code Generation Code Generation Harness Embedded microprocessor Real-time simulator 8

9 Agenda Develop ECUs with Model-Based Design Generate Application Code for Prototyping Generate Algorithm Code for Production Case Study: Brushless Motor Control Application Questions 9

10 xpc Target Focusing Areas System Requirements Sim Design with Simulation Executable Specifications Models Continuous Test and Verification System Integration and Tuning HIL System Design RP Automatic Code Generation Hardware/Software Integration PIL Sim: Simulation RP: Rapid Prototyping PCG: Production Code Generation SIL: Software-in-the-Loop Testing PIL: Processor-in-the-Loop Testing HIL: Hardware-in-the-Loop Testing Software Design RP (Embedded) Coding PCG Software Integration SIL 10

11 What is xpc Target? 1 2 Simulink Coder Host computer with MATLAB xpc Target Turnkey computer 3 Ethernet or RS 232 An environment that allows the real-time execution of Simulink models on a separate xpc Target Kernel compatible PC. 11

12 What is xpc Target? Host computer with MATLAB Host Computer with MATLAB Ethernet or RS 232 xpc Target Turnkey computer xpc Target Turnkey Computer Ethernet or RS 232 An Environment that provides interactive access between the real-time application and the host computer Allows live parameter tuning, control from the original Simulink model and offline analysis support in MATLAB. 12

13 What is xpc Target? 2 Host computer with MATLAB xpc Target Turnkey computer 4 Ethernet or RS An Environment that provides interactive access between the real-time application and the host computer Allows live parameter tuning, control from the original Simulink model and offline analysis support in MATLAB. 13

14 What is xpc Target? Host computer with MATLAB xpc Target Turnkey PC computer Ethernet or RS 232 An Environment that provides numerous I/O device driver blocks Blocks are easily configurable within the Simulink model and communicate with actual hardware in real-time. 14

15 What is xpc Target? xpc Target Turnkey computer An Environment that provides numerous I/O device driver blocks Blocks are easily configurable within the Simulink model and communicate with actual hardware in real-time. 15

16 Mobile Real-Time Target Machine Performance : 2GHz Dual Core Intel processor 2GB RAM, 100GB HD Gigabit Ethernet Expandable : 5 PCI Slots Digital and Analog I/O, CAN, Serial, etc FPGA Expansion Card : High-speed PWM and Quadrature generation and decoding High-frequency control loops Programmable with Mathworks HDL Coder 16

17 xpc Target Turnkey Target Hardware partnership with Speedgoat Hardware customized to fit your performance and I/O needs Saves you time and hassle of acquiring, installing, configuring, and testing Commercial Version : Academia Version : 17

18 Agenda Develop ECUs with Model-Based Design Generate Application Code for Prototyping Generate Algorithm Code for Production Case Study: Brushless Motor Control Application Questions 18

19 Considerations for Embedded S/W Development 19

20 Code Generation Products: Simulink Coder and Embedded Coder Simulink Coder Generates code for use in simulation and prototyping applications Comes with Generic Real-Time (GRT) based targets Embedded Coder Generates efficient code that can be customized to look like hand code for production Comes with Embedded Real-Time (ERT) based targets 20

21 Agenda Develop ECUs with Model-Based Design Generate Application Code for Prototyping Generate Algorithm Code for Production Case Study: Brushless Motor Control Application Questions 21

22 Brushed DC vs. AC Motor Control Brushed DC Motor Control : Mechanical Commutation Brushless DC Motor: Electrical Commutation Battery Embedded Controller Inverter Software 22

23 Example PMSM Controller Controller Outputs: Three PWM Duty Cycles Controller Algorithm Attributes: Outer Loop Velocity Inner Loop Current Coordinate transforms (ABC to/from dq) Multiple PI Controllers Rotor Phase Currents (AC Waveforms) 23

24 Optimize Performance with Simulation u + PMSM Control Algorithm Inverter and Motor y Controller Plant Simulating plant and controller in one environment allows engineers to optimize system-level performance. Controls engineers and domain specialists can work together to detect integration issues in simulation. 24

25 Rapid Prototyping System Model Controller Model Motor Model Code Generation Controller C Code Motor Hardware Real Time Simulator Rapid Prototyping 25

26 Hardware High-Level Block Diagram Simulink Host Real-Time Target Application Hardware Power Electronics xpc Target Turnkey Mobile Machine PMSM Motor Encoder 26

27 Embedded Code Generation System Model Controller Model Motor Model Code Generation Controller C Code Motor Hardware Embedded Software Embedded System 27

28 Integrating Generated Controller Code with an Embedded Software Project Embedded Software Project Execute at 25kHz { Command ADC Controller PWM Encoder 28

29 Integrating Generated Controller Code with an Embedded Software Project Embedded Software Project Main() { Commands_Init PWM_Init ADC_Init Encoder_Init } Controller_Init while(1) { } ADC_InterruptServiceRountine_25kHz() { Commands_Read ADC_Read Encoder_Read } Controller_Step PWM_Write 29

30 Serial / TCPIP Assessing behavior and execution time using Processor-In-the-Loop (PIL) Simulink Test Signals Controller Model Verifications Embedded Processor Serial / TCPIP Code Generation Controller C Code PIL Implementation 30

31 Questions? 2012 The MathWorks, Inc. 31