Virtual sensor - AI model training using VOLVO Brake temperatures

Sammanfattning

Accurate prediction of brake disc temperatures in heavy-duty vehicles is essentialfor ensuring safety, reducing wear and improving braking performance. Excessiveheat buildup in the disc can lead to brake fade, accelerated material degradationand increased emissions of harmful wear particles. This thesis focuses on predictingbrake disc temperatures using time-series data collected from controlled dynamometertests. The dataset includes braking signals such as torque, pressure and speed,recorded at high frequency under a wide range of operating conditions. Various machinelearning models, including neural networks, were developed to predict brakedisc temperatures during individual braking events. This work serves as a foundationfor future efforts to extend temperature prediction models to real-world fielddata and ultimately support the development of intelligent thermal monitoring systemsthat can reduce brake wear, improve safety and help meet upcoming Euro 7regulations on particle emissions from braking systems.