Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
Defining, Analyzing, and Clustering Drive Cycles for Engine Applications Through Feature Engineering
No full textAn ordered sequence of measurements, or time series, is the raw representation from
which many analytics and data-driven decisions are made in the automotive industry.
Volvo Penta is a company determined to become more data-driven to stay
ahead of the competition. In this endeavor, signals from various manufactured engines
in the field are logged as time series data. Through analysis of this data lies
the information from which to obtain the repeated behavioral pattern of an engine
in use, otherwise known as drive cycles. This thesis aims to segment time series
data from an engine placed in a log stacker into drive cycles and engineer, extract,
and select relevant features descriptive of engine usage from these. From the cycles
with accompanying features, clustering techniques will be applied to group the
drive cycles into behaviorally distinct categories. Drive cycles were defined by segmenting
consecutive engine signals whenever the gap between recorded signal values
exceeded ten minutes. Filtering was then applied to remove uninformative and outlier
cycles. From the remaining cycles, features were derived, which, after selection,
yielded eight temporal and statistical features that formed the basis for clustering.
Using these features, the clustering algorithms k-means, agglomerative, and meanshift
were employed, resulting in clusters that depicted two main engine behaviors.
These were high- and low-load behaviors. The exception was the resulting clusters
produced by mean-shift, which exhibited a single dominant behavior and a few deviating
cycles. A consensus partition integrated all three methods and agreed with the
solution provided by k-means. This thesis demonstrates that feature-based, unsupervised
clustering can group drive cycles segmented from engine field test data into
behaviorally distinct categories. These findings further the goals of Volvo Penta’s
mission to become more data-driven and demonstrate the potential of clustering as
a first step in areas such as engine simulation and predictive maintenance
Side chain engineering of PTQ-derivatives for improved performances and processability in photocatalysis and organic photovoltaics
No full textImplementation and Evaluation of Last-Level Cache Partitioning
No full textWith parallel applications executing on a multiprocessor comes the problem of interference due to resource sharing between processor cores. The contention of resources can lead to a lack of fairness between the cores. Furthermore, having a shared last-level cache introduces destructive interference in the form of cores evicting other cores’ cache blocks. Cache partitioning is a method to solve these problems, which has both been extensively researched in academia and implemented in end-products in industry. In this thesis we will analyze the effects of cache partitioning in terms of performance and overhead. We implement six replacement algorithms and a cache design in RTL, all supporting way-based cache partitioning. These are then used in a test bench where execution is simulated by feeding the test bench memory traces extracted from the L2-accesses of benchmark programs from the SPEC-CPU-2006 benchmark suite.
From our results we conclude that way-based partitioning can easily be implemented with little overhead. We also show that in some cases, the increase of shared ways available to an application does not offset the effect of interference, which leads to worse performance of the application. From the aforementioned memory traces we generate reuse histograms to analyze the expected cache behaviour of the simulated benchmark programs. This gives us the ability to predict a good way to partition the cache, based on the running applications
Optimization of estimating intrinsic parameters of a camera using a robot arm
No full textEye-tracking technology is applied across industries such as automotive, aerospace,
and human-computer interaction, where precise gaze estimation relies on the accurate
calibration of eye-tracking cameras. In this work, a robot arm is used to perform
the intrinsic calibration of eye tracking cameras. The motivation behind using a
robotic arm for the intrinsic calibration of an eye-tracking camera instead of a manual
procedure stems from several key advantages related to precision, repeatability,
and efficiency. Control over the camera poses in front of the checkerboard leads to a
more accurate and efficient calibration process. This eliminates the challenges and
inconsistencies associated with manual camera movements and opens the door for
automating and scaling the calibration process with minimal human involvement.
The robotic arm ensures a level of consistency and reliability that would be hard to
achieve manually, making it a compelling choice for eye-tracking camera calibration,
especially in high-precision applications
Temperature Controlled Thermal Jacket
No full textSammanfattning
Detta kandidatarbete undersöker design och implementering av en temperaturreglerad prototyp för en elektriskt uppvärmd jacka. Värme genereras i en nikromtråd då den belastas med effekt från ett litium-polymerbatteri. Effektleveransen hanteras genom en designad synkron buck-omvandlare. Ett proportionell-integrerande reglersystem har implementerats för att reglera utspänningen och stabilisera temperaturen i värmeelementet. Säkerhetsaspekter såsom isolering, sensorers noggrannhet och begränsningar i användargränssnittet har också tagits i beaktning för att göra jackan praktisk och säker för vardagligt bruk. En 1,5 meter lång nikromtråd valdes som värmeelement då den gav tillräcklig värmeeffekt. Termiska tester genomfördes på rena, isolerade och integrerade uppsättningar för att analysera värmefördelningen. Ett specialdesignat kraftsteg med en synkron buck-omvandlare utvecklades och implementerades. En Arduino mikrokontroller, tillsammans med skärm och lämpligt gränssnitt mot temperatursensor, styr och reglerar kraftsteget. Ett grafiskt användargränssnitt gör det möjligt för användaren att se och ställa in önskad temperatur. Jackan är utrustad med ett fodrande lager av bomull som hjälper till att fördela värmen och skydda användaren från direktkontakt med värmeelementen. Säkerhetsfunktioner såsom en nödbrytare och en värmetröskel har integrerats för att förhindra överhettning eller personskador. Prototypen uppnådde inte de angivna värdena för temperatur med godtagbar noggrannhet och stabilitet. Tester reslutera i tempraturer på ±5◦C på den bestämda tempraturen 45◦C efter 10 minuter. Trots vissa begränsningar i batteritid och värmefördelning utgör projektet en god grund för vidareutveckling och förfining av elektriskt uppvärmda kläder
Improving Thunks in the Verified PureCake Compiler
No full textCompiler construction is a research area that is constantly being developed, as the need for faster and more correct programming languages and implementations is steadily growing. While much work has been done to ensure compilers output efficient and semantically intact code, the complexity and size of the current optimising compilers lead to bugs that can be hard to find and diagnose. Situations where correctness is critical has therefore led to the development of formally verified compilers, that guarantee the correctness of the generated code. The price paid for developing verified compilers is that every step in the compilation process must be carefully implemented and formalised in order for the verification system to accept it, which makes optimisation pipelines much harder to implement.
This project’s work is based on the PureCake verified compiler for pure, lazy functional languages. PureCake is an extension of the CakeML project, which itself is a verified compiler for strict functional languages, both developed using the HOL4 formal verification system. The PureCake compiler is already a complete verified compiler, but it lacks the optimisations that would make its generated code as efficient as the generated code of optimising non-verified compilers for lazy languages such as GHC. This project’s goal is to enhance PureCake’s and CakeML’s language with features that will enable critical optimisations to be implemented in order to close the gap between verified and non-verified compilers for lazy languages in terms of runtime efficiency.
In this thesis, we take on the task of improving the quality of PureCake’s generated code, by optimising the handling of thunks by the PureCake and CakeML compilers. We first target the representation of thunks inside PureCake, by enriching the semantics of the operations responsible for handling thunks and by proving these changes correct. Then, we implement thunks in CakeML and lay the foundations for future optimisations in the runtime. This work will help bring PureCake up to par with unverified compilers in terms of the generated executable’s performance, while also making sure that the soundness of the compilation process remains intact
Interactive Buckling and Design Evaluation of Circular Hollow Sections - Assessment of Design Guidelines and Buckling Mode Interaction Through Numerical Analysis and Parametric Study of Buckling Resistance of Circular Hollow Steel Sections
No full textThis master’s thesis investigates the behaviour of circular hollow section (CHS) in
steel under axial loading. Particular emphasis is placed on how the interaction
between global and local buckling modes affects member strength. The objective
was to complement and compare how current design guidelines assess these sections
against the obtained results.
To fulfil the aim, the work began with a literature review to deepen understanding
of the subject. This was followed by the development and validation of an FE-model
capable of replicating buckling behaviour observed in previous studies. A subsequent
parametric study was conducted, covering geometries and parameters relevant for a
comprehensive perspective on CHS buckling behaviour.
The parametric study results, compared with design guidelines, show that current
methods create issues for CHS classified as CSC 4. These are treated as shell structures,
although their behaviour resembles columns, leading to unsafe and inefficient
design. To improve accuracy, an effective area method and a local buckling reduction
factor from the British steel design standard (BS5950-1) and a study by
Toffolon and Taras (2017) were investigated. These methods aligned more closely
with results and provided safer, more efficient CSC 4 designs.
In addition, the interaction between global and local buckling was studied. In
columns showing interactive failure, ultimate load was reduced by 8 % compared to
columns with only local modes, despite nearly identical geometry. Two parameters
were identified for determining the transition from interactive to fully local buckling:
L′ which is defined as L′ =
q
l2t
r3 and Ncr,global
Ncr,local
.
TThis master’s thesis investigates the behaviour of circular hollow section (CHS) in
steel under axial loading. Particular emphasis is placed on how the interaction
between global and local buckling modes affects member strength. The objective
was to complement and compare how current design guidelines assess these sections
against the obtained results.
To fulfil the aim, the work began with a literature review to deepen understanding
of the subject. This was followed by the development and validation of an FE-model
capable of replicating buckling behaviour observed in previous studies. A subsequent
parametric study was conducted, covering geometries and parameters relevant for a
comprehensive perspective on CHS buckling behaviour.
The parametric study results, compared with design guidelines, show that current
methods create issues for CHS classified as CSC 4. These are treated as shell structures,
although their behaviour resembles columns, leading to unsafe and inefficient
design. To improve accuracy, an effective area method and a local buckling reduction
factor from the British steel design standard (BS5950-1) and a study by
Toffolon and Taras (2017) were investigated. These methods aligned more closely
with results and provided safer, more efficient CSC 4 designs.
In addition, the interaction between global and local buckling was studied. In
columns showing interactive failure, ultimate load was reduced by 8 % compared to
columns with only local modes, despite nearly identical geometry. Two parameters
were identified for determining the transition from interactive to fully local buckling:
L′ which is defined as L′ =
q
l2t
r3 and Ncr,global
Ncr,local
.
The major conclusions were: (i) current guidelines result in unsafe and inefficient
design for CHS in CSC 4, and an effective area method, similar to the one established
in the British standard, eliminate unsafe design but show conservative results for
the reduction factor stated; and (ii) interactive buckling reduces buckling resistance
of CHS of up to 8 % and thusly should be considered in design
Planning and Follow-Up Processes in Production and Its Influence on Productivity - A Case Study of Two Projects in the Swedish Construction Sector
No full textThe construction industry is behind other sectors in terms of efficiency improvements
and has low productivity levels. Swedish construction projects are often delayed,
which is usually related to low productivity and inefficient working methods. The
current methodologies used for the planning and follow-up processes in production
have been shown to be difficult to apply in practice, with limited support to examine
productivity. Since there are multiple different definitions of productivity, the overall
development of productivity has slowed down and created difficulties in measuring
and using the term in the construction sector. This study has therefore narrowed its
focus down to efficiency and effectiveness when referring to productivity. The aim of
this study was to examine how planning and follow-up processes were conducted,
whether there are standardised approaches or if each project has its own approach.
It further investigated what challenges existed related to those processes and in turn
how they affected productivity and what improvements could be done to enhance
it. To achieve the aim of this study, an interview study and a case study were done
where two ongoing building projects in one company in the Swedish construction
industry were examined. The conclusions that can be drawn based on this study
were that the planning and follow-up were conducted with varying approaches of the
Last Planner System (LPS). The chosen approach in the projects varied depending
on the individuals involved and relied on their knowledge, priority and focus on these
processes. Another factor that influenced how planning and follow-up were carried
out was the lack of clear structure and requirements on a project level. A significant
challenge which impacted the outcome was collaboration between those involved in
the planning process. The study showed several ways that the planning and follow-up
processes could be improved to potentially enhance productivity in production. It
highlighted the need of securing a general knowledge level and implementing clear
requirements and guidelines for how to work with planning and follow-up on a project
level to be able to enhance productivity. The study further made it visible that the
culture and priority on planning and follow-up varied between the projects and was
highly dependent on individuals but also affected by how the company relates to
them. Establishing a culture that promotes the importance of planning is essential
to accomplish a project environment that enables proper planning and follow-up
processes, allowing projects to focus on improving productivity
Samlokalisering av preanalytiska processer inom verksamheten Klinisk Genetik och Genomik Utmaningar, möjligheter och kommunikation
No full textDen ökade efterfrågan på genetiska analyser har lett till en organisatorisk omstrukturering
där Klinisk Genetik och Genomik bildats vid Sahlgrenska Universitetssjukhuset. Inom
detta centrum spelar sektionen för Provhantering och Preanalys (PoP) en viktig roll.
PoP:s verksamhet är idag utspridd över tre geografiska tomter, men för att förbereda för
inflyttning i det nya Sahlgrenska Life-huset planeras en tidigare samlokalisering.
Denna studie syftar till att identifiera vilka möjligheter och risker som uppstår vid en
samlokalisering av PoP, med särskilt fokus på känsliga provflöden såsom prenatala prover
och Patologisk Anatomisk Diagnos prover. Genom en kombination av kvalitativa inter-
vjuer och värdeflödesanalyser belyser studien fördelar som standardiserade arbetssätt,
förbättrad informationshantering och stordriftsfördelar. Samtidigt identifieras utmaning-
ar i form av ökade transportbehov, förändrade kommunikationsvägar och organisatorisk
anpassning.
Resultaten visar att en lyckad samlokalisering kräver en tydligt kommunikation samt
anpassade beslutsstrukturer. Studien ger även förslag på hur framtida förändringar bör
genomföras för att minimera störningar och maximera nyttan
Optimizing Pricing Strategies Using Machine Learning: A Prototype Development for Derome
No full textThis project explored the application of machine learning models to analyze and forecast sales performance using internal transaction data from Derome, a Swedish company. The study implemented clustering, classification, and regression techniques to build interpretable, high-performing prototypes aimed at enhancing decision-making and revenue forecasting.
For clustering, k-means was applied to annual aggregates of product groups based on sales volume and outcome. Using the elbow method, three distinct performance tiers were identified, validated by a high silhouette score of 0.89. These tiers were assigned to transactions and used as labels in a classification task. Among the evaluated models, a decision tree classifier achieved the highest accuracy of 85.86% on unseen data for the year 2024 and outperformed a strong baseline classifier. Feature importance analysis revealed that the supplier was the most influential factor in predicting a transaction’s associated performance tier.
Regression modeling focused on predicting monthly sales using two approaches: forecasting based on individual transaction-level predictions, and direct forecasting from aggregated historical data. Extreme gradient boosting regressor demonstrated the best overall performance, achieving a mean absolute percentage error (MAPE) of 8.96% between the predicted and true values for 2024 using the second approach. It was also used to forecast 2025 sales, achieving a MAPE of 8.07% when compared to the company’s predicted revenue for that year, indicating strong alignment with business expectations.
Despite limitations such as a restricted hyperparameter search, exclusive reliance on internal data, and initial gaps in domain knowledge, the project successfully delivered functional prototypes. These systems provide a foundation for future development and demonstrate the practical value of machine learning in improving business forecasting and decision-making.
Overall, the project met its objectives by uncovering customer behavior patterns, optimizing internal workflows through machine learning, and supporting pricing and strategic planning with interpretable, data-driven models