Abstract
Turbochargers are key components of engine air-paths that must be carefully considered during the development process. Current turbocharger models are based on characteristic maps derived from experimental measurements taken under steady conditions on dedicated gas stand facility. Under these conditions heat transfer and gas dynamics are ignored and consequently the predictive performances of the models are compromised, particularly under the part load and dynamic operating conditions that are representative of real powertrain operations. Furthermore, for VGT turbochargers and waste-gated turbochargers, the mechanics of the variable vane and waste-gate actuators should be considered.This project attempts to apply a dynamic mathematical model that uses a polynomial structure, the parametric Volterra Series, for the modelling of the turbocharger system to replace the conventional modelling procedure which is based on turbocharger characteristic maps. The model is calculated directly from measured performance data using an extended least squares regression. In this way, both compressor and turbine are modelled together based on data from dynamic experiments rather than steady flow data from a gas stand. Therefore, a Volterra series model is an empirical based model which could represent the turbocharger performance steadily and dynamically.
The combination of fluid, mechanical and thermal phenomenon make the turbine to be a highly dynamic and non-linear modelling challenge. A better understanding of turbine dynamic performance across the frequency spectrum is necessary. The prototype used in this project is a waste-gated turbine. It was modelled using the dual-orifice approach, a lumped capacitance heat transfer model and novel physics based pneumatic actuator mechanism model. Each sub-model has been validated individually against experimental measurements (by waste-gate actuator test rig and engine gas stand test rig). The turbine inlet pressure and temperature and the waste-gate actuator pressure had been perturbed across the full frequency range both individually and simultaneously in separate numerical investigations. The dynamic responses of turbine housing temperature, turbocharger rotor speed, waste-gate opening, and mass flow and gas temperatures/pressures were all investigated to quantify the dynamic response of the turbocharger system across a frequency spectrum from 0.003Hz to 500Hz, i.e. for exhaust gas pulsation in steady state, load steps and cold start drive cycles, to validate the assumption of quasi-steady assumptions for particular modelling problems.
- The mass flow parameter exhibits significant dynamic behaviour above 100Hz (>100Hz), illustrating that the quasi-steady assumption is invalid in this frequency range. While turbocharger housing temperature and turbocharger shaft speed do not react to the rapid changes of boundary conditions. This means that poor estimates in mechanical and thermal inertia will have little effect on the modelling accuracy and a constant estimation of housing temperature or shaft speed will remain accurate.
- The mechanical inertia of the turbine attenuated fluctuations in shaft speed for frequencies between 0.1-10Hz where the waste-gate actuator system showed quasi-steady behaviours. The actuator behaves dynamically at frequency from 10Hz to 100Hz. Within this frequency range (0.1-100Hz), it is still hard for temperature to react the boundary conditions variations. The mass flow rate responses in much a quasi-steady way. This illustrates the important of actuator mass and shaft inertia estimations.
- The thermal inertia of the turbocharger housing meant that housing temperature variations were supressed at frequencies below 0.01Hz, where gas-dynamics and mechanics play less important roles. The thermal inertias are the critical parameters should been taken care in modelling process.
In order to work out a simple method to describe turbine transient performance on thermal, fluid and mechanical dynamics through a wide range of frequency spectrum on different sizes of turbocharger, this project looks into the non-dimensional analysis to eliminate the dimension influence of turbocharger sizes. Based on the findings from previous research, three frequency regions were defined for thermal, mechanical and fluid dynamics of turbocharger respectively. For fluid-dynamics, the Strouhal number has used to judge the importance of gas dynamics in modelling process. In this project, two more non-dimensional parameters derived based on the analogous to Strouhal number are introduced for thermal and mechanical dynamics.
Firstly, the modelling approach has been applied to dynamic data taken from a physics based model (quasi-steady assumption based), acting as a simulation test platform. Varying frequency sinusoidal signals were applied to the compressor and turbine pressure ratios and turbine inlet temperature to drive the physics-based model. Good results are observed from the implement of Volterra series in compressor modelling with nRMSE values of compressor efficiency, mass flow rate and outlet temperature less than 6%, 3% and 2% respectively as well as R2 larger than 0.96. With frequency range covering from 0.003Hz to 300Hz, The data needed for model identification is large making the data processing to be difficult. An advanced physics based model with turbine conventional model replaced by a dynamic model is needed as the previous simulation test platform is inappropriate to be used to identify the turbine dynamic modelling with quasi-steady assumption applied. The studies of turbine dynamic performance investigation provide a way to split the turbine Volterra Series model to be three parts. Each part represent one critical dynamic factor (thermodynamic, gas-dynamic, mechanics) in corresponding frequency region. The combination of these sub models forms the complete turbine dynamic model.
| Date of Award | 1 Jul 2017 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Richard Burke (Supervisor) |
Cite this
- Standard