Machine Learning on Field Test Data

Thesis background

Volvo Penta has a vested interest in utilizing and understanding vast amounts of data. As Martin Lundstedt (Volvo Group President and CEO) recounts, “It’s important to have good data management and ownership [of data], because then on top, you can create a lot of smart solutions.”
To that end, we at Volvo Penta Field Test (VPFT) have applied data engineering principles to build a large-scale data lake. This has enabled us to create a data-driven and cloud-native workflow for advanced analytics using Microsoft Azure and Databricks. Along with our data-driven workflow, data engineering and analytics have enabled us to reach valuable strategic conclusions and optimizations.
During this endeavor, technologies such as Artificial Neural Networks (ANN), evolutionary algorithms, semantic segmentation, and various clustering techniques have become core components of our current workspace. The most recent thesis project at VPFT created a deep learning process for detection and semantic segmentation of engine drive cycles.
We at VPFT are now committed to taking the next step in our machine learning journey.

Thesis Scope of Work

Our purpose with this thesis is to study real world engine usage data to further our understanding of engine behavior and applications. For this purpose, we propose two approaches using machine learning and advanced analytics.
The first approach is to examine fault code data to find clusters of engines that encounter similar fault codes and are used alike. Our intended approach is inspired by Natural Language Processing (NLP). By embedding the different fault codes as vectors, that can be used in a vector space model, engine behavior could be studied in an analogous fashion to synonym detection in NLP.
The second approach is to detect divergent behavior of sensor readings. By creating machine learning models on time resolute signals, from multiple different sensors on the engine, a regression model can be attained. The idea is to select several correlated signals and train the model to predict any one signal given the others. Thus, any input that deviates from the norm could be detected by comparing the model output to the actual signal.
If either of these approaches sound interesting, we would like to have a meeting with you where we can further discuss your thesis. We are also open to other avenues of investigation that you may be thinking of.

Workspace Subset

A selection of tools and technologies we currently use in day-to-day operations.
  • Python 3
    • Pandas
    • NumPy
    • SciPy
  • Azure
    • DevOps
    • Data Factory
    • Functions
    • CosmosDB
    • SQL Server
  • SQL
  • Databricks
  • Apache Spark
Qualifications & Required Documents

You should possess these qualifications:
  • Master student in any of the following fields
  • Artificial Intelligence
  • Computer Science
  • Data Science
  • Machine Learning
  • Physics
  • Knowledge of machine learning algorithms and big data
  • Some programming proficiency
Your application should include the following:
  • CV
  • Cover Letter
  • Transcript of grades
Contact
Andreas Nyman, Manager Field Test and Data Management, andreas.nyman@volvo.com

About Field Test

Volvo Penta Field Test is divided into two primary areas, field test activities and Volvo Penta Data Management.
Field test activities are divided into Marine- and Industrial product segments. The field test engineers are responsible for test objects connected to selected OEMs (Original Equipment Manufacturers) or customers and for a specific engine range depending on the projects. The responsibility covers planning and start-up of field tests objects as well as support during the test execution, and completion of test.
Volvo Penta Data Management are responsible for structuring data and building data science tools. We are principally working on data infrastructure, setting up data storage structures from raw data, building advanced analytics tools, and producing a data-driven culture and collaborative environment.
You will be part of a global and diverse team with highly skilled and passionate professionals who trust each other, and embraces change to stay ahead. We make our customers win.

Practical Information

Thesis Level: Master (30 ECTS points)
Language: English
Starting date: January 2022
Number of students: 2 students
Last application date: 1st of December 2021
Examiner proposal:
Entity: AB Volvo Penta
State/Province: Västra Götaland
City/Town: Göteborg
Employment/Assignment Type: Thesis
Functional Area: Technology

The Volvo Group drives prosperity through transport solutions, offering trucks, buses, construction equipment, power solutions for marine and industrial applications, financing and services that increase our customers’ uptime and productivity. Founded in 1927, the Volvo Group is committed to shaping the future landscape of sustainable transport and infrastructure solutions. Countless career opportunities are offered across the group’s leading brands and entities that share a culture of Trust, Passion, High Performance, Change and Customer Success. 
www.volvogroup.com/career. 

With Volvo Penta, a world-leading supplier of engines and complete drive systems for marine and industrial applications, you will be part of a global and diverse team of highly skilled professionals who works with passion, trust each other and embraces change to stay ahead. We make our customers win.

欢迎加入我们

申请流程

申请

您提交申请之后,即会收到一封确认邮件。点击确认之后,您仍然可以更新您的个人资料。

Testimonials

类似职位

Engine synchronous sensor processing in the transmission control unit Technology Göteborg, Sweden 发布日期: 
Feature Leader Reliability & Durability Technology Göteborg, Sweden 发布日期: 
Enterprise Architect Technology Göteborg, Sweden 发布日期: