Determination of the transfer function between the injected flow-rate and high-pressure time histories for improved control of common rail diesel engines

2016 ◽  
Vol 18 (3) ◽  
pp. 212-225 ◽  
Author(s):  
Alessandro Ferrari ◽  
Emanuele Salvo

Theoretical and experimental methodologies have been proposed and illustrated to determine the transfer function between the injected flow-rate and the rail pressure for common rail injection systems. An analytical transfer function has been calculated in the frequency domain, utilizing a previously developed lumped parameter model of the overall hydraulic layout of a common rail system. The predicted transfer function has been compared, in a Bode diagram, with an experimental estimation of the transfer function, based on the measured rail pressure and injected flow-rate time histories that were acquired at the hydraulic rig for different working conditions. The experimental estimation of the transfer function has been worked out by applying a selective spectral technique in order to reduce the effects of measurement noise on the rail pressure and injected flow-rate time histories. The accuracy of the model-derived transfer function has been improved significantly by integrating a pressure control system sub-model, which includes the action of the electronic control unit on the rail pressure time history through the pressure regulator, in the hydraulic model of the common rail circuit. Finally, the time histories of the rail pressure, predicted by means of the complete injection apparatus model, have been compared with the corresponding experimental traces at different working conditions and a very satisfactory agreement has in general been found. The methodologies proposed for the accurate evaluation of the transfer function between the injected flow-rate and the rail pressure time histories can be applied to diesel engines in order to implement innovative closed-loop strategies for the injected mass control.

2016 ◽  
Vol 18 (3) ◽  
pp. 273-290 ◽  
Author(s):  
Alessandro Ferrari ◽  
Pietro Pizzo

A fully predictive model of a Common Rail fuel injection apparatus, which includes a detailed simulation of rail, pump, piping system, injectors and rail pressure control system, is presented and discussed. The high-pressure pump and injector sub-models have been validated separately and then coupled to the rail and pressure control system sub-models. The complete predictive model has been validated and applied to investigate the effects of the dynamics of each component of the injection apparatus on the rail pressure time history. Variable timing of the high-pressure pump delivery phases has also been considered, and the influence of this parameter on the injection performance has been analysed for both single- and multiple-injection events. Furthermore, the injection system dynamics during the transients between steady-state working conditions has been investigated in order to highlight the role played by the dynamic response of the pressure control system on the rail pressure time history.


Author(s):  
Mateusz Graczyk ◽  
Kjetil Berget ◽  
Joachim Allers

Sloshing, a violent fluid motion in tanks is of current interest for many branches of the industry, among them gas shipping. Although different methods are commonly combined for analyzing sloshing in LNG carriers, time histories of the pressure in the tanks are most reliably obtained by experiments. Very localized pressures may be important for the structural response of the tank containment system. Moreover, the typical pressure time history duration is similar to the structural natural frequency. Therefore, pressure measurements need to be performed with due account for temporal and spatial distribution. This requires a high sampling resolution both in time and space. Fine spatial resolution becomes especially important when local pressure effects are of interest, such as pressure profile passing a membrane corrugation of Mark III containment or Invar edge of No.96 containment. In this paper experimental approach applied by MARIN-TEK for analyzing sloshing phenomenon is presented. The focus is put on investigating effects of Invar edges. A transverse 2D model of a typical LNG carrier is used. Local pressure effects are investigated based on low filling level tests with different wall surfaces: smooth and with horizontal protrusions representing the surface similar to the No.96 containment system.


Author(s):  
A. Ferrari ◽  
F. Paolicelli

A detailed analysis of a common rail (CR) fuel injection system, equipped with solenoid injectors for Euro 6 diesel engine applications, has been performed in the frequency domain. A lumped parameter numerical model of the high-pressure hydraulic circuit, from the pump delivery to the injector nozzle, has been realized. The model outcomes have been validated through a comparison with frequency values that were obtained by applying the peak-picking technique to the experimental pressure time histories acquired from the pipe that connects the injector to the rail. The eigenvectors associated with the different eigenfrequencies have been calculated and physically interpreted, thus providing a methodology for the modal analysis of hydraulic systems. Three main modal motions have been identified in the considered fuel injection apparatus, and the possible resonances with the external forcing terms, i.e., pump delivered flow rate, injected flow rate, and injector dynamic fuel leakage through the pilot valve, have been discussed. The investigation has shown that the rail is mainly involved in the first two vibration modes. In the first mode, the rail performs a decoupling action between the high-pressure pump and the downstream hydraulic circuit. Consequently, the oscillations generated by the pump flow rates mainly remain confined to the pipe between the pump and the rail. The second mode is centered on the rail and involves a large part of the hydraulic circuit, both upstream and downstream of the rail. Finally, the third mode principally affects the injector and its internal hydraulic circuit. It has also been observed that some geometric features of the injection apparatus can have a significant effect on the system dynamics and can induce hydraulic resonance phenomena. Furthermore, the lumped parameter model has been used to determine a simplified transfer function between rail pressure and injected flow rate. The knowledge obtained from this study can help to guide designers draw up an improved design of this kind of apparatus, because the pressure waves, which are triggered by impulsive events and are typical of injector working, can affect the performance of modern injection systems, especially when digital rate shaping strategies or closely coupled multiple injections are implemented.


2013 ◽  
Vol 46 (21) ◽  
pp. 717-722 ◽  
Author(s):  
Seungwoo Hong ◽  
Jaewook Shin ◽  
Inseok Park ◽  
Myoungho Sunwoo ◽  
Jongik Jeon ◽  
...  

1985 ◽  
Vol 107 (4) ◽  
pp. 252-257 ◽  
Author(s):  
J. C. Gilkey ◽  
J. D. Powell

Determining fuel-air ratio quickly over a wide range of engine operating conditions is desirable for better transient engine control. This paper describes a method based on cylinder pressure time history pattern recognition which has potential for providing such a high bandwidth measurement. The fact that fuel-air ratio has an effect on the shape of the cylinder pressure trace is well-known. It should therefore be possible to obtain the fuel-air ratio of an engine by examining the pressure trace if the engine speed, load, and EGR are known. The difficulty lies in separating the effects of unknown engine load, speed, and EGR from the fuel-air ratio effects. An algorithm was developed using a wide range of steady state experimental data from a single cylinder engine. Application of the algorithm requires the calculation of first, second and third moments of the cylinder pressure time history. Verification of the algorithm showed that the root mean square error in estimates were about 5 percent for fuel-air ratio and 3 percent for a combination of fuel-air and EGR. These results were obtained using a single pressure trace which yields a response time of 1.5 engine revolutions. The algorithm was also found to be relatively insensitive to the use of different fuels, errors in spark advance, and variations in relative humidity. Research is continuing to verify the accuracy under transient engine conditions. An operational count shows that this algorithm should be well within the limits of present microprocessor technology.


Author(s):  
Andrea Emilio Catania ◽  
Alessandro Ferrari ◽  
Michele Manno

A rather complete mathematical model for a Common Rail injection-system dynamics numerical simulation was developed to support experimentation, layout and control design, as well as performance optimization. The thermo-fluid dynamics of the hydraulic system components, including rail, connecting pipes and injectors was modeled in conjunction with the solenoid-circuit electromagnetics and the mechanics of mobile elements. Onedimensional flow equations in conservation form were used to simulate wave propagation phenomena throughout the high-pressure connecting pipes, including the feeding pipe of the injector nozzle. In order to simulate the temperature variations due to the fuel compressibility, the energy equation was used in addition to mass conservation and momentum balance equations. Besides, the possible cavitation phenomena effects on the mass flow rate through the injector bleed orifice and the nozzle holes were taken into account. A simple model of the electromagnetic driving circuit was used to predict the temporal distribution of the force acting on the pilot-valve anchor. It was based on the experimental time-histories of the current through the solenoid and of the associated voltage that is provided by the electronic control unit (ECU) to the solenoid valve. The numerical code was validated through the comparison of the prediction results with experimental data, that is, pressure, injected flow rate and needle lift time-histories, taken on a high performance test bench Moehwald-Bosch MEP2000-CA4000. The novel injection-system mathematical model was applied to the analysis of transient flows through the hydraulic circuit of a commercial multijet second-generation Common Rail system, paying specific attention to the wave propagation phenomena, to their dependence on solenoid energizing time and rail pressure, as well as to their effects on system performance. An insight was also given into the model capability of accurately predicting the wave dynamics effects on the rate and mass of fuel injected when the dwell time between two consecutive injections is varied.


Sign in / Sign up

Export Citation Format

Share Document