Chalmers Research
Not a member yet
88095 research outputs found
Sort by
LGEM+: A\ua0First-Order Logic Framework for\ua0Automated Improvement of\ua0Metabolic Network Models Through Abduction
No full textScientific discovery in biology is difficult due to the complexity of the systems involved and the expense of obtaining high quality experimental data. Automated techniques are a promising way to make scientific discoveries at the scale and pace required to model large biological systems. A key problem for 21st century biology is to build a computational model of the eukaryotic cell. The yeast Saccharomyces cerevisiae is the best understood eukaryote, and genome-scale metabolic models (GEMs) are rich sources of background knowledge that we can use as a basis for automated inference and investigation. We present LGEM+, a system for automated abductive improvement of GEMs consisting of: a compartmentalised first-order logic framework for describing biochemical pathways (using curated GEMs as the expert knowledge source); and a two-stage hypothesis abduction procedure. We demonstrate that deductive inference on logical theories created using LGEM+, using the automated theorem prover iProver, can predict growth/no-growth of S. cerevisiae strains in minimal media. LGEM+ proposed 2094 unique candidate hypotheses for model improvement. We assess the value of the generated hypotheses using two criteria: (a) genome-wide single-gene essentiality prediction, and (b) constraint of flux-balance analysis (FBA) simulations. For (b) we developed an algorithm to integrate FBA with the logic model. We rank and filter the hypotheses using these assessments. We intend to test these hypotheses using the robot scientist Genesis, which is based around chemostat cultivation and high-throughput metabolomics
Physics-based model predictive control for power capability estimation of lithium-ion batteries
Get PDFThe power capability of a lithium-ion battery signifies its capacity to continuously supply or absorb energy within a given time period. For an electrified vehicle, knowing this information is critical to determining control strategies such as acceleration, power split, and regenerative braking. Unfortunately, such an indicator cannot be directly measured and is usually challenging to be inferred for today\u27s high-energy type of batteries with thicker electrodes. In this work, we propose a novel physics-based battery power capability estimation method to prevent the battery from moving into harmful situations during its operation for its health and safety. The method incorporates a high-fidelity electrochemical-thermal battery model, with which not only the external limitations on current, voltage, and power, but also the internal constraints on lithium plating and thermal runaway, can be readily taken into account. The online estimation of maximum power is accomplished by formulating and solving a constrained nonlinear optimization problem. Due to the relatively high system order, high model nonlinearity, and long prediction horizon, a scheme based on multistep nonlinear model predictive control is found to be computationally affordable and accurate
Spare Parts Demand Prediction by Using a Random Forest Approach
No full textSpare parts forecasting, in general, is a complex task, due to its intermittent and erratic demand patterns. Furthermore, the underlying reason for the demand is usually not considered since most forecasting methods are based on demand history, merely. Spare parts demand is dependent on the need for replacement of components due to repair or maintenance reasons, which varies due to, e.g., life cycle, utilization of the finished product, and the number of finished products. Although machine learning methods have been more prevalent for spare parts forecasting in recent years, the prediction of initial demand, i.e., what to be stocked before the breakdowns and maintenance needs occur is understudied. So, the purpose of this paper is to investigate if the spare part need can be predicted before the demand occurs and, by that, increase the availability to the customers and decrease the cost of unavailability, e.g., expediting costs and lost sales. By adopting a machine learning model based on decision trees, called Random Forest, we predicted the probability of initial sales based on the installed base, and categorical variables such as product group, vital code, function, and weight, for spare parts at an automotive company. The analysis was made for three different markets in the Asia-pacific region and predictions were made for three different time horizons, 1, 3, and 6\ua0months and the model performance shows potentially good results with an accuracy of around 70%. We also analyzed the business impact concerning availability and supply chain-related costs where, for example, we obtained a substantially lower total supply chain cost
The need for research and innovation to facilitate upscaling of low-carbon concrete
Get PDFFor decades, research has been carried out with a focus on concrete structures during curing to mitigate the risk of thermal cracking. Computer programs and aids/tools have also been developed to assess stress and cracking risk analysis of concrete structures during curing. However, today with the recent introduction of low-carbon concretes to reduce the environmental impact of constructions, the reliability of the tools and working procedures, i.e. concrete characterization, is questioned, and a roadmap for research and innovation is called for. The project\u27s primary purpose is to investigate the need for research and innovation regarding upscaling the usage of low-carbon concrete. The nature of the study is based on an industry-focused workshop with specialists from Scandinavia. Increased knowledge of hardening concrete\u27s cracking risk-related properties is of the utmost importance for the construction industry as the need for its understanding has recently increased
Real-life demonstration of flexibility provision by smart charging of EVs and stationary battery storage
Get PDFThe main aim of the work is to demonstrate in real-life the possibility to manage the flexible demand-side resources, including DC fast EV charger, V2G EV charger and batteries, in a smart manner to mitigate potential grid congestion problems. For this purpose, an intelligent “EV management platform” has been developed and used in the demonstration. Through this platform, the DSO can determine a threshold level for loading of the transformer based on historical loading data of the transformer and can determine the charge/discharge profiles for the equipment for the designated time, considering the charging preferences of electric vehicle users. The results from the demonstration showed that the demand-side resources can be dispatched automatically to satisfy the users’ needs while effectively preventing the local grid congestion problems
Finite Blocklength Performance Bound for the DNA Storage Channel
No full textWe present a finite blocklength performance bound for a DNA storage channel with insertions, deletions, and substitutions. The considered bound - the dependency testing (DT) bound, introduced by Polyanskiy et at. in 2010 - , provides an upper bound on the achievable frame error probability and can be used to benchmark coding schemes in the practical short-to-medium blocklength regime. In particular, we consider a concatenated coding scheme where an inner synchronization code deals with insertions and deletions and the outer code corrects remaining (mostly substitution) errors. The bound depends on the inner synchronization code. Thus, it allows to guide its choice. We then consider low-density parity-check codes for the outer code, which we optimize based on extrinsic information transfer charts. Our optimized coding schemes achieve a normalized rate of 87% to 97% with respect to the DT bound for code lengths up to 2000 DNA symbols for a frame error probability of and code rate 1/2
Detection of ultra-fast radio bursts from FRB 20121102A
No full textFast radio bursts (FRBs) are extragalactic transient flashes of radio waves with typical durations of milliseconds. FRBs have been shown, however, to present a wide range of timescales: some show sub-microsecond sub-bursts while others last up to a few seconds. Probing FRBs on a range of timescales is crucial for understanding their emission physics, how to detect them effectively and how to maximize their utility as astrophysical probes. FRB 20121102A is the first known repeating FRB source. Here we show that FRB 20121102A produces isolated microsecond-duration bursts with durations less than one-tenth the duration of other currently known FRBs. The polarimetric properties of these microsecond-duration bursts resemble those of the longer-lasting bursts, suggesting a common emission mechanism producing FRBs with durations spanning three orders of magnitude. In detecting and characterizing these microsecond-duration bursts, we show that there exists a population of ultra-fast radio bursts that current wide-field FRB searches are missing due to insufficient time resolution. These results indicate that FRBs occur more frequently and with greater diversity than initially thought. This could also influence our understanding of energy, wait time and burst rate distributions
Clarifying the complex chemistry of cobalt(II) thiocyanate-based tests for cocaine using single-crystal X-ray diffraction and spectroscopic techniques
Get PDFCobalt(II) thiocyanate-based tests are routinely used to screen cocaine products, with the formation of a blue species interpreted as a positive response. An array of other organic bases has been identified as false positives – including well-documented cocaine product adulterant lidocaine and its salt. False positives prompt continued test development, though improvements are hindered by unresolved product structures and reaction pathways. Toward greater clarity, cobalt(II) thiocyanate reactions with cocaine hydrochloride, along with lidocaine and its salt, were investigated using multiple analytical techniques. Reactions involving cocaine hydrochloride yielded glassy, amorphous blue material while reactions of lidocaine hydrochloride monohydrate produced larger, needle-like crystals whose structure was determined via single-crystal X-ray diffraction to be an ion pair (Hlidocaine+)2([Co(SCN)4]2−)\ub7H2O. While the blue precipitate isolated from reactions involving cocaine hydrochloride was unsuitable for crystallographic structure determination, comparative ultraviolet–visible, attenuated total reflectance infrared, and Raman spectroscopic analysis – along with elemental analysis – supports that this solid is comprised of a comparable ion pair (Hcocaine+)2[Co(SCN)4]2−. Pink crystals isolated from lidocaine reaction vessels were identified as coordination compounds cis-[CoL2(SCN)2] and trans-[CoL2(SCN)2] where L = lidocaine, while pink crystals from both cocaine hydrochloride and lidocaine hydrochloride monohydrate reaction vessels were the coordination polymer trans-[Co(H2O)2(SCN)2]\ub7H2O. The results presented herein enable reaction optimization to favor a desired product, whether ion pair or coordination species
Structural Coverability for Intelligent Automation Systems
No full textIn order to be flexible and handle complex scenarios, intelligent automation systems might benefit from automated planning techniques which rely on specifications and models describing their behavior. However, due to the presence of message passing, latency, jitter, timeouts, failures, and error handling, the verification of such behavior models using formal methods is often unfeasible. Therefore, testing has emerged as an approach to evaluating the behavior of intelligent automation systems. This paper presents a way to analyze structural coverability of behavior models for intelligent automation systems, which is inspired by the modified condition/decision coverage (MC/DC) criterion. This is paired with a testing procedure that enables each test case to influence both the controller and the simulated environment by injecting some specific state. As a result, the proposed coverability criterion can effectively identify segments of the behavior model that have not been adequately tested and suggest additional test cases to improve coverability. An example use case is presented to demonstrate the effectiveness of this approach
Safe Trajectory Sampling in Model-Based Reinforcement Learning
No full textModel-based reinforcement learning aims to learn a policy to solve a target task by leveraging a learned dynamics model. This approach, paired with principled handling of uncertainty allows for data-efficient policy learning in robotics. However, the physical environment has feasibility and safety constraints that need to be incorporated into the policy before it is safe to execute on a real robot. In this work, we study how to enforce the aforementioned constraints in the context of model-based reinforcement learning with probabilistic dynamics models. In particular, we investigate how trajectories sampled from the learned dynamics model can be used on a real robot, while fulfilling user-specified safety requirements. We present a model-based reinforcement learning approach using Gaussian processes where safety constraints are taken into account without simplifying Gaussian assumptions on the predictive state distributions. We evaluate the proposed approach on different continuous control tasks with varying complexity and demonstrate how our safe trajectory-sampling approach can be directly used on a real robot without violating safety constraints