Implementing an automatic calibration system for an engine control unit (ECU) on a test bench that focuses on fuel-film dynamics in a port fuel injection (PFI) internal combustion engine.
Using NI LabVIEW and NI CompactRIO for flexible, real-time engine parameter calibration.
L. Tenze - ESTECO
E. Corti - ALMA AUTOMOTIVE
F. Pignaton - ESTECO
L. Onesti - ESTECO
G. Mancini - ALMA AUTOMOTIVE
Car manufacturers can reduce costs and development time by applying optimization techniques to automatic ECU calibration. Alma Automotive and ESTECO collaborated to demonstrate the value of combining National Instruments devices with a modeFRONTIER framework while incorporating specific applications developed by Alma Automotive for the engine setup. Typical optimization processes are slow, but we aimed for a short calibration time for optimal engine-under-test management. Therefore, our main challenge was to achieve optimum parameter results as quickly as possible while maximizing system accuracy.
The Alma Automotive engine test bench instrumented by LabVIEW provides complete control of the system under test, and the multidisciplinary optimization framework modeFRONTIER (an ESTECO product) completes the infrastructure optimization capabilities. We implemented the test bench control system using CompactRIO, which acquires data and drives the actuators. The signals obtained by test-bed sensors are routed to modeFRONTIER as optimization targets. The optimization algorithm, bound by the user target, automatically chooses the parameter values, driving the transducers toward the optimal solution.
The engine and test bench controllers (connected to the engine and to bench transducers/actuators, respectively) are implemented using CompactRIO devices that are properly programmed and equipped with a field-programmable gate array (FPGA). The system sends signals from CompactRIO through an Ethernet connection to two host PCs running Windows 7. The PCs acquire and store the data, and set up the operating parameters. The interfaces developed by Alma Automotive provide a simple way to interact with the system.
We installed modeFRONTIER on the host PC that is connected to the engine controller. Under typical working conditions, the engine runs on a steady operating point (identified by load and revolution speed). The measurements from the sensors are sent via CompactRIO to the destination PC, where modeFRONTIER interacts with the ECU to optimize the chosen target.
The system we created uses the real-time processing characteristics of CompactRIO to test the optimization algorithms provided by modeFRONTIER . The LabVIEW integration node, provided by modeFRONTIER, guarantees communication between the systems, which makes connection fast and simple.
During the first test, we controlled the spark advance to achieve the maximum brake torque value. Our goal was to test the effectiveness of the chosen hardware and software environments and check the efficiency of the interaction between LabVIEW and modeFRONTIER. modeFRONTIER ran on the test-bed PC, receiving signals from in-cylinder pressure sensors, and properly processed them to obtain the target to maximize indicated mean effective pressure.
Before the test, we set the range of spark advance degrees on modeFRONTIER. We chose the simplex (Nelder-Mead ) algorithm for maximum torque. We reached the goal in approximately 10 optimization steps.
A more complex and meaningful experiment to test the effectiveness of the system is the fuel-film automatic compensation. We used modeFRONTIER algorithms to identify the value of the X and tau constants in real time to compensate for fuel-film dynamics.
On a spark-ignition engine with the injection on the intake manifold, a fraction of the injected gasoline (identified by the X value) falls on the intake runner wall. The deposited fuel goes in the cylinder after the evaporation time (the tau value).
To highlight the effect related to fuel-film dynamics, and to extract concise and meaningful information, we properly set the engine operating point (controlled in an open loop) and drove the injectors with step waveform . In ideal conditions, the system reproduces the injection profile from the lambda signal on the exhaust collector. With the compensation algorithm off, the variations of the lambda signal do not follow the imposed steps due to lowpass behaviour. Instead, with the compensation on, the closer the signal waveform is to ideal, the better the calibration. The evaluated signal error is strictly correlated to the difference between the ideal transitions and the measured air/fuel ratio.
The tight integration between the Alma Automotive test environment (based on CompactRIO and LabVIEW) and the ESTECO optimization framework helped us quickly create a demonstrator to automatically tune two ECU subsystems. In particular, test environment flexibility gave us complete control over the engine and test bed, while the advanced optimization algorithms included in modeFRONTIER helped us quickly and efficiently reach the targets. We plan to apply the optimization at different engine operating points to completely map the compensation system.
In the future, we can use the proposed approach with other ECU subsystems to build a complete automatic calibration environment for the engine, avoiding direct user interaction while reducing cost and time.
For further information and technical support, please contact ESTECO, email@example.com
Please contact Esteco for purchasing information.
Provider Homepage: http://esteco.com/
Product Homepage: http://esteco.com/home/mode_frontier/Integration/mF4LV.html