No / The Job Shop Scheduling Problem (JSSP) is a hard combinatorial optimisation problem. This paper presents a heuristic-based Genetic Algorithm (GA) or Hybrid Genetic Algorithm (HGA) with the aim of overcoming the GA deficiency of fine tuning of solution around the optimum, and to achieve optimal or near optimal solutions for benchmark JSSP. The paper also presents a detail GA parameter analysis (also called sensitivity analysis) for a wide range of benchmark problems from JSSP. The findings from the sensitivity analysis or best possible parameter combination are then used in the proposed HGA for optimal or near optimal solutions. The experimental results of the HGA for several benchmark problems are encouraging and show that HGA has achieved optimal solutions for more than 90% of the benchmark problems considered in this paper. The presented results will provide a reference for selection of GA parameters for heuristic-based GAs for JSSP.

Energy-efficient powertrain components and advanced vehicle control strategies are two effective methods to promote the potential of hybrid electric vehicles (HEVs). Aiming at hybrid system efficiency improvement, this thesis presents a comprehensive review of energy-efficient hybrid powertrain specific engines and proposes three improved energy management strategies (EMSs), from a basic non-adaptive real-time approach to a state-of-the-art learning-based intelligent approach. To evaluate the potential of energy-efficient powertrain components in HEV efficiency improvement, a detailed discussion of hybrid powertrain specific engines is presented. Four technological solutions, i.e., over-expansion cycle, low temperature combustion mode, alternative fuels, and waste heat recovery techniques, are reviewed thoroughly and explicitly. Benefits and challenges of each application are identified, followed by specific recommendations for future work. Opportunities to simplify hybrid-optimized engines based on cost-effective trade-offs are also investigated. To improve the practicality of HEV EMS, a real-time equivalent consumption minimization strategy (ECMS)-based HEV control scheme is proposed by incorporating powertrain inertial dynamics. Compared to the baseline ECMS without such considerations, the proposed control strategy improves the vehicle drivability and provides a more accurate prediction of fuel economy. As an improvement of the baseline ECMS, the proposed dynamic ECMS offers a more convincing and better optimal solution for practical HEV control. To address the online implementation difficulty faced by ECMS due to the equivalence factor (EF) tuning, a predictive adaptive ECMS (A-ECMS) with online EF calculation and instantaneous power distribution is proposed. With a real-time self-updating EF profile, control dependency on drive cycles is reduced, and the requirement for manual tuning is also eliminated. The proposed A-ECMS exhibits great charge sustaining capabilities on all studied drive cycles with only slight increases in fuel consumption compared to the basic non-adaptive ECMS, presenting great improvement in real-time applicability and adaptability. To take advantage of machine learning techniques for HEV EMS improvement, a deep reinforcement learning (DRL)-based intelligent EMS featuring the state-of-the-art asynchronous advantage actor-critic (A3C) algorithm is proposed. After introducing the fundamentals of reinforcement learning, formulation of the A3C-based EMS is explained in detail. The proposed algorithm is trained successfully with reasonable convergence. Training results indicate the great learning ability of the proposed strategy with excellent charge sustenance and good fuel optimality. A generalization test is also conducted to test its adaptability, and results are compared with an A-ECMS. By showing better charge sustaining performance and fuel economy, the proposed A3C-based EMS proves its potential in real-time HEV control. / Thesis / Doctor of Philosophy (PhD)

No description available.

No description available.

No description available.

With the introduction of electronic control units to automotive vehicles, system complexity has increased. With this change in complexity, new standards have been created to ensure safety at the system level for these vehicles. Furthermore, vehicles have become increasingly complex with the push for electrification of automotive vehicles, which has resulted in the creation of hybrid electric and battery electric vehicles. The goal of this thesis is to provide an example of a hazard and operability analysis as well as a hazard and risk analysis for a hybrid electric vehicle. Additionally, the safety standards developed do not align well with educational prototype vehicles because the standards are designed for corporations. The hybrid vehicle supervisory controller example within this thesis demonstrates how to define a system and then perform system-level analytical techniques to identify potential failures and associated requirements. Ultimately, through this analysis suggestions are made on how best to reduce system complexity and improve system safety of a student built prototype vehicle. / Master of Science / With the introduction of electronic control units to automotive vehicles, system complexity has increased. With this change in complexity, new standards have been created to ensure safety at the system level for these vehicles. Furthermore, vehicles have become increasingly complex with the push for electrification of automotive vehicles, which has resulted in the creation of hybrid electric and battery electric vehicles. There are different ways for corporations to demonstrate adherence to these standards, however it is more difficult for student design projects to follow the same standards. Through the application of hazard and operability analysis and hazard and risk analysis on the hybrid vehicle supervisory controller, an example is provided for future students to follow the guidelines established by the safety standards. The end result is to develop system requirements to improve the safety of the prototype vehicle with the added benefit of making design changes to reduce the complexity of the student project.

The negative effects from rising carbon levels have created the need to find alternative energy sources that are more carbon neutral. One such alternative energy source is to use the biomass derived from forest trees to fulfill the need for a renewable alternative fuel. Through increased understanding and optimization of regulatory mechanisms that control wood development the potential exists to increase biomass yield. Transcription factors (TFs) are DNA-binding regulatory proteins capable of either activation or repression by binding to a specific region of DNA, normally located in the 5-prime upstream promoter region of the gene. In the first section of this work, six DNA promoters from wood formation-related genes were screened by the Yeast One-Hybrid (Y1H) assay in efforts to identify novel interacting TFs involved in wood formation. The promoters tested belong to genes involved in lignin biosynthesis, programmed cell death, and cambial zone associated TFs. The promoters were screened against a mini-library composed of TFs expressed 4-fold or higher in differentiating xylem vs phloem-cambium. The Y1H results identified PtrRAD1 with interactions involving several of the promoters screened. Further testing of PtrRAD1 by Yeast Two-Hybrid (Y2H) assay identified a protein-protein interaction (PPI) with poplar DIVARACATA RADIALIS INTERACTING FACTOR (DRIF1). PtrDRIF1 was then used in the Y2H assay and formed PPIs with MYB/SANT domain proteins, homeodomain family (HD) TFs, and cytoskeletal-related proteins. In the second section of this work, PPIs involving PtrDRIF1s' interaction partners were further characterized. PtrDRIF1 is composed of two separate domains, an N-terminal MYB/SANT domain that interacted with the MYB/SANT domain containing PtrRAD1 and PtrDIVARICATA-like proteins, and a C-terminal region containing a Domain of Unknown Function 3755 (DUF3755). The DUF3755 domain interacted with HD family members belonging to the ancient WOX clade and Class II KNOX domain TFs. In addition, PtrDRIF1 was able to form a complex between PtrRAD1 and PtrWOX13c in a Y2H bridge assay. PtrDRIF1 may function as a regulatory module linking cambial cell proliferation, lignification, and cell expansion during growth. Combined, these findings support a role for PtrDRIF1 in regulating aspects of wood formation that may contribute to altering biomass yield. / Ph. D. / Trees are unique among plants since they have extremely long life spans and the ability to generate large quantities of woody biomass. The woody biomass derived from forest trees can function to provide renewable energy in the form of biofuels. The process of wood formation is complex and requires coordinated activation of genes involved in multiple metabolic pathways. Transcription factors (TFs) are DNA-binding regulatory proteins capable of either activation or repression by binding to a specific region of DNA. These protein-DNA interactions regulate gene expression during plant growth and development. In this study, new regulators of genes known to be involved in wood formation were identified using the Yeast One-Hybrid (Y1H) assay. One of the proteins identified, PtrRAD1 had not been previously linked to wood formation and was a candidate for further characterization. Further testing of PtrRAD1 by the Yeast Two-Hybrid (Y2H) assay resulted in identification of a protein-protein interaction with Populus trichocarpa DIVARICATA RADIALIS INTERACTING FACTOR (DRIF1). PtrDRIF1 was then used in the Y2H assay to identify numerous interacting proteins, in addition to those reported previously in other species. Further characterization of PtrDRIF1, identified an N-terminal region capable of forming interactions with MYB/SANT domain proteins, and C-terminal region that interacted with homeodomain proteins. PtrRAD1, PtrDRIF1, and the homeodomain containing PtrWOX13c were able to form a complex in an Y2H-bridge assay. Combined, these findings support a potential role for PtrDRIF1 in regulating wood polarity, wood formation, and stem cell proliferation.

Corona-pandemin blev en stor bidragande faktor för flertal organisationer att anpassa sitt arbetssätt till en hybrid arbetsmiljö. Oberoende om hybrida arbetsförhållanden blivit påtvingat under pandemin eller om det funnits inom organisationen tidigare finns det många möjligheter där ibland flexibilitet som framtida organisationer vill och kommer att nyttja. Därför är det numer vanligt att organisationer vill och försöker utnyttja de fördelar som finns med både traditionell arbetsmiljö och hybrid arbetsmiljö vilket är ett nytt sätt att bygga och forma en arbetsplats på.  När organisationsförändringar och förändringar i arbetssätt inträffar medför det förändringar för både medarbetare och formella ledare. Ledarskap är något som ständigt förändras och det mer moderna synsättet på ledarskap skiljer sig markant mot det traditionella som oftast förknippas med hård styrning och auktoritet. Det moderna ledarskapet handlar mer om att vara lyhörd, flexibel och fokus ligger på mjuka värden. Precis som många andra faktorer i en organisation påverkas av ett nytt arbetssätt påverkas så även ledarskapet. Studien har besvarat forskningsfrågan, det vill säga hur ledarskapet kan anpassas till den hybrida arbetsmiljön för att bibehålla och främja teamsammanhållning. Forskningsfrågan har besvarats med hjälp av 5 intervjuer med formella ledare som arbetat inom organisationer som både verkat i traditionell arbetsmiljö och hybrid arbetsmiljö. Resultatet av studien visar att formella ledare bör anpassa sitt ledarskap på en generell nivå för att främja sammanhållning i den hybrida arbetsmiljön men att behovet av anpassningen kan skilja sig från organisation till organisation sett utifrån bransch.

Data Mining deals with extracting valid, novel, easily understood by humans, potentially useful and general abstractions from large data. A data is large when number of patterns, number of features per pattern or both are large. Largeness of data is characterized by its size which is beyond the capacity of main memory of a computer. Data Mining is an interdisciplinary field involving database systems, statistics, machine learning, visualization and computational aspects. The focus of data mining algorithms is scalability and efficiency. Large data clustering and classification is an important activity in Data Mining. The clustering algorithms are predominantly iterative requiring multiple scans of dataset, which is very expensive when data is stored on the disk. In the current work we propose different schemes that have both theoretical validity and practical utility in dealing with such a large data. The schemes broadly encompass data compaction, classification, prototype selection, use of domain knowledge and hybrid intelligent systems. The proposed approaches can be broadly classified as (a) compressing the data by some means in a non-lossy manner; cluster as well as classify the patterns in their compressed form directly through a novel algorithm, (b) compressing the data in a lossy fashion such that a very high degree of compression and abstraction is obtained in terms of 'distinct subsequences'; classify the data in such compressed form to improve the prediction accuracy, (c) with the help of incremental clustering, a lossy compression scheme and rough set approach, obtain simultaneous prototype and feature selection, (d) demonstrate that prototype selection and data-dependent techniques can reduce number of comparisons in multiclass classification scenario using SVMs, and (e) by making use of domain knowledge of the problem and data under consideration, we show that we obtaina very high classification accuracy with less number of iterations with AdaBoost. The schemes have pragmatic utility. The prototype selection algorithm is incremental, requiring a single dataset scan and has linear time and space requirements. We provide results obtained with a large, high dimensional handwritten(hw) digit data. The compression algorithm is based on simple concepts, where we demonstrate that classification of the compressed data improves computation time required by a factor 5 with prediction accuracy with both compressed and original data being exactly the same as 92.47%. With the proposed lossy compression scheme and pruning methods, we demonstrate that even with a reduction of distinct sequences by a factor of 6 (690 to 106), the prediction accuracy improves. Specifically, with original data containing 690 distinct subsequences, the classification accuracy is 92.47% and with appropriate choice of parameters for pruning, the number of distinct subsequences reduces to 106 with corresponding classification accuracy as 92.92%. The best classification accuracy of 93.3% is obtained with 452 distinct subsequences. With the scheme of simultaneous feature and prototype selection, we improved classification accuracy to better than that obtained with kNNC, viz., 93.58%, while significantly reducing the number of features and prototypes, achieving a compaction of 45.1%. In case of hybrid schemes based on SVM, prototypes and domain knowledge based tree(KB-Tree), we demonstrated reduction in SVM training time by 50% and testing time by about 30% as compared to complete data and improvement of classification accuracy to 94.75%. In case of AdaBoost the classification accuracy is 94.48%, which is better than those obtained with NNC and kNNC on the entire data; the training timing is reduced because of use of prototypes instead of the complete data. Another important aspect of the work is to devise a KB-Tree (with maximum depth of 4), that classifies a 10-category data in just 4 comparisons. In addition to hw data, we applied the schemes to Network Intrusion Detection Data (10% dataset of KDDCUP99) and demonstrated that the proposed schemes provided less overall cost than the reported values.

Im Vortrag werden die komplexen und teilweise konträren Herausforderungen an zukünftige automobile Antriebsstränge und Getriebesysteme vorgestellt. Die klassische Suche nach neuen Antriebssträngen, basierend auf Intuition und Expertenwissen, kann diesen vielseitigen Zielstellungen kaum noch gerecht werden. Aus diesem Grund wird ein neuer methodischer Ansatz vorgestellt, mit dem konventionelle und hybride Getriebekonzepte rechnergestützt generiert werden können. Die Voraussetzung dafür bildet die eindeutige und kompakte mathematische Abbildung von beliebigen Getriebesystemen oder ganzen Antriebssträngen im Computer auf Basis eines Wellenmodells. Diese Codierung lässt sich effizient hinsichtlich diverser kinematischer und kinetischer Belastungen für frei wählbare Freiheitsgrade analysieren. Die rechnergestützte Synthese und Bewertung neuer Getriebe wird am Beispiel von Planetenautomatikgetrieben vorgestellt, wobei alle kombinatorisch möglichen konventionellen und hybriden Systeme Beachtung finden. Beispielsweise werden für eine aktuell typische Anzahl an Planetenradsätzen und Schaltelementen mehrere hundert Milliarden Getriebevarianten untersucht. Am Ende eines Syntheselaufs werden dem Entwicklungsingenieur die besten Systeme in einem Ranking vorgeschlagen. Der Vortrag schließt mit der Vorstellung exemplarischer Ergebnisse dieser neuen Entwicklungsmethode. Das neue Achtgang-Hybrid-Automatikgetriebe und das neue Neungang-Hybrid-Doppelkupplungsgetriebe verdeutlichen die Entwicklungspotentiale zukünftiger Getriebesysteme, bei denen hochfunktionale und effiziente Eigenschaften mit kleinstmöglichem mechanischem Aufwand, geringen Kosten und Bauraum kombiniert werden. Die IAV GmbH nutzt dieses Entwicklungstool erfolgreich zur Erzeugung von verschiedenen Getriebearten, wie Planetenautomatikgetrieben, Doppelkupplungsgetrieben, Handschaltgetrieben und stufenlos leistungsverzweigten Getrieben mit mechanischem oder elektrischem Variator.