Publication: Elektrikli araçlarda cruise control sistem tasarımı ve model bazlı parametre optimizasyonu
Abstract
Bu tez çalışmasında modern çağın en önemli otomotiv fenomeni elektrikli araçların sürücü ve trafik güvenliğini arttıran, sürüşü konforlu hale getiren sürücü destek sistemlerinden Cruise Control sisteminin tasarlanması ve bu model ile eş zamanlı çalışan sanal simülasyon platformu oluşturularak, bu sanal platform ile model bazlı parametre optimizasyonu yöntemi ile parametre optimizasyonu yapılmıştır. Artan hava kirliliği ve yenilenebilir enerji kaynaklarına yönelişin bir sonucu olarak elektrikli araçlara olan rağbet son yıllarda büyük bir artış göstermiştir. Bu değişikliğin etkisiyle mühendisler ve bilim insanları için araştırma ve geliştirme çalışmaları yapılabilecek birçok yeni alan ortaya çıkmıştır. Sürüş güvenliği, konforunu arttıran, yakıt tüketimini azaltan Cruise Control sistemlerin geliştirilmesi de bu konulardan birisidir. Araç üzerinde yapılan çalışmaların yüksek maliyeti ve harcanan zaman ve eforun fazlalığı, fonksiyon geliştiricileri sanal simülasyon platformları oluşturmaya teşvik etmiştir. Simülasyon çalışmalarının artışındaki en büyük etkenlerden biri de sürekli geliştirilebilir ve kullanılabilir olmasıdır. Buradan hareketle bu çalışmada ilk olarak Cruise Control sistemi için fonksiyon ve kontrolcü modeli oluşturulmuştur. Daha sonrasında bir çok elektrikli araç incelenerek binek elektrikli araç tipi için, araç dinamiği simülasyon AVL VSM™ ile araç modeli, sürücü modeli ve test manevrası modellenmiştir. Araç modeli ile fonksiyon modeli yine aynı program içerisinde birbirine entegre edilmiştir. Modellerin çalışırlığını doğrulamak için test senaryoları tanımlan ve bu testler aracılığıyla sistemin hatasız çalıştığı doğrulanmıştır. Cruise Control sistemi aktif edilip istenilen hıza sabitlendiği ya da sabit hız değiştirilerek başka bir hıza sabitlendiği geçiş durumlarındaki araç hızı hatasının PID kontrol ile minimize edilmesi için optimize edilecek parametreler belirlenmiştir. Model bazlı parametre optimizasyonu için aktif deney tasarımı ve optimizasyon AVL CAMEO 5™ kullanılarak S-optimal tasarım ile 150 adet deney seti oluşturulup, belirlenen manevra ile simülasyonlar koşulmuştur. Simülasyon sonuçları incelenerek araç hızı hata değerleri ile birlikte oluşturulan deney seti programa beslenerek lokal optimizasyon yöntemi ile optimal sonuçlar elde edilmiştir. Elde edilen sonuçlar ile manevra yeniden simüle edilerek araç hızı hatasının minimize edilmesi doğrulanmıştır.
This thesis represents design of Cruise Control system which is one of the driver support systems that increases safety for driver and traffic, provides more driving comfort and parameter optimization using model based parameter optimization method via generated co-simulation platform which works simultaneously for an electric vehicles, which is one of the most important automotive phenomena of the modern age. As a result of increase of air pollution and renewable energy sources trend, the demand of electric vehicles has been massive increased in recent years. With the effect of this change, many new areas have emerged for engineers and scientists to conduct research and development studies. One of new research area is the development of Cruise Control systems that increase driving safety, comfort and reduce fuel consumption. The excessive cost of working on the vehicle and the excessive amount of time and effort have encouraged function developers to generate virtual simulation platforms. One of the biggest factors in the increase of simulation studies is that it can be continuously developed and used. From this point of view, in this study, a function and controller model was created for the Cruise Control system firstly. Afterwards, many electric vehicles were examined and the vehicle model, driver model and test maneuver were modeled using AVL VSM™ which is the vehicle dynamics simulation tool for the passenger type electric vehicle. The vehicle model and the function model are integrated into each other by same program. In order to verify the operability of the models, test scenarios are defined, and it is verified that the system works without errors through these tests. Optimization parameters have determined in order to minimize the vehicle speed error with PID control in transition situations where the Cruise Control system is activated and set to the desired speed or set to another speed by changing set speed. 150 different parameter sets were generated with S-optimal design using active design of experiment (DoE) and optimization tool which is AVL CAMEO 5™ for model-based parameter optimization and then simulations were run with the determined maneuver. The optimal results have been acquired via local optimization method using parameter sets as an inputs and vehicle speed errors as an output in optimization tool after analysis of simulations results. Obtained optimal values was used for final simulation to verify minimizing of vehicle speed error. Then re-simulation performed and verified.
This thesis represents design of Cruise Control system which is one of the driver support systems that increases safety for driver and traffic, provides more driving comfort and parameter optimization using model based parameter optimization method via generated co-simulation platform which works simultaneously for an electric vehicles, which is one of the most important automotive phenomena of the modern age. As a result of increase of air pollution and renewable energy sources trend, the demand of electric vehicles has been massive increased in recent years. With the effect of this change, many new areas have emerged for engineers and scientists to conduct research and development studies. One of new research area is the development of Cruise Control systems that increase driving safety, comfort and reduce fuel consumption. The excessive cost of working on the vehicle and the excessive amount of time and effort have encouraged function developers to generate virtual simulation platforms. One of the biggest factors in the increase of simulation studies is that it can be continuously developed and used. From this point of view, in this study, a function and controller model was created for the Cruise Control system firstly. Afterwards, many electric vehicles were examined and the vehicle model, driver model and test maneuver were modeled using AVL VSM™ which is the vehicle dynamics simulation tool for the passenger type electric vehicle. The vehicle model and the function model are integrated into each other by same program. In order to verify the operability of the models, test scenarios are defined, and it is verified that the system works without errors through these tests. Optimization parameters have determined in order to minimize the vehicle speed error with PID control in transition situations where the Cruise Control system is activated and set to the desired speed or set to another speed by changing set speed. 150 different parameter sets were generated with S-optimal design using active design of experiment (DoE) and optimization tool which is AVL CAMEO 5™ for model-based parameter optimization and then simulations were run with the determined maneuver. The optimal results have been acquired via local optimization method using parameter sets as an inputs and vehicle speed errors as an output in optimization tool after analysis of simulations results. Obtained optimal values was used for final simulation to verify minimizing of vehicle speed error. Then re-simulation performed and verified.