Study of Software System Integration for Transient Simulation of Future Cooling System for Heavy Truck Applications
[Examensarbete på avancerad nivå]
The work investigates the integration between tools for analysis and simulation of cooling system at Volvo Group Trucks Technology. At the same time it is a consequent step in evaluating GT-SUITE as a tool for analysis and simulation of cooling systems. The interaction between the truck's cooling system and the engine has so far not been modeled in its entirety in a completely integrated, detailed model capable of simulating transient runs. This work marks the first steps in this direction. A number of different methods for modeling cooling system performance exist. This project focuses on 1D simulation tools, which are generally preferred in the context of transient simulations of power train and engine installation systems. The Cooling Analysis and Simulations group at Volvo Group Trucks Technology uses KULI as 1D simulation tool for analysis of cooling performance. Volvo Powetrain, on the other hand, uses GT-SUITE for engine simulations. It is expected to improve the quality of the simulation, (i.e the accuracy of the results) and improve system integration by using one tool for both areas of simulation. GT-SUITE is a powerful tool for modeling the uid-dynamics and heat-transfer phenomena, which occur in the cooling system. The work includes a detailed model of the main coolant circuit since this is vital for the authenticity of the transient simulation. This thesis is a natural continuation of a project performed by the author during the summer of 2012, which evaluated GT-SUITE as a tool for steady-state simulation of cooling systems. The basic models of the cooler package were developed and calibrated during this project. This work delivers two transient models of FH 13liter Euro 6 cooling system integrated with a predictive engine model, provided by Volvo Powertrain. As a first step, all necessary models and control algorithms were obtained from different technical units within the organization and were further reworked and refined to fit the purpose of this work. Such are the engine model obtained from BF66360 System Analysis and Simulation at Volvo Powertrain, the fan control model provided by BF72362 Cooling Systems at Powertrain Installation and the coolant pump control model from BF69317 Vehicle Functions. Component performance data was as well acquired for different components within the coolant circuit: thermostat valve, coolant pump, engine oil cooler, etc. GEM3D was used to ease the process of creating authentic models of the coolant logistic components (pipes, hoses, owsplits, etc) The first model produced in this work, delivers a basic representation of the physics in the system and its main aim is to prove the technical feasibility of the concept. It includes all necessary functional features : working fan control, coolant pump mode control, function for temperature compensation for the effect of air recirculation. The model was used to obtain critical cooling system-related parameters from Hamburg-Kassel drive cycle and critical parameters were compared to measurements from tests. The results, acquired from the first model have satisfactory consistency with the data from test. Average coefficient of determination achieved by the model is R2 >0.85, which for most of the parameters is higher than the results given by the currently available program for transient simulations. The rate of execution of the model was successfully increased to 1 x Real Time. The second model implements two-way communication between the engine model and the cooling system model: the temperature on the outlet of the CAC and torque consumed by coolant pump and fan are fed back to the engine model and therefore the interaction between the performance of the two separate systems is partially accounted for. Average coefficients of determination achieved from this model are similar to the ones from the model with single connectivity. Execution time rate did not alter significantly in comparison to the previous model. The implemented interaction between the subsystems allows for a predictive model of the boost temperature and investigations on fuel economy. The work has proven the feasibility and the integrability of an engine model and a model of a cooling system in GT being controlled in Simulink environment. This paper can be seen as a comprehensive manual to building, tuning and executing such a model.
Nyckelord: 1D cooling simulations, transient simulations, GT-SUITE, Matlab, Simulink, engine models, trucks, calibration
Det verkar som att din webbläsare saknar möjlighet att visa PDF. Klicka Öppna i nytt fönster för att läsa.