Understanding and quantify human performance is an essential component to guarantee and control the safety of critical installations where human intervention can represent the ultimate safety defence. Human reliabilit...
详细信息
The automated insulin delivery (AID) system is a powerful approach for regulating the blood glucose concentration (BGC) of people with type 1 diabetes (T1D) by delivering optimal doses of insulin. While rapid progress...
详细信息
ISBN:
(数字)9781665473385
ISBN:
(纸本)9781665473385
The automated insulin delivery (AID) system is a powerful approach for regulating the blood glucose concentration (BGC) of people with type 1 diabetes (T1D) by delivering optimal doses of insulin. While rapid progress has been made in developing hybrid closed-loop AID system in recent decades and several commercial AID systems are available, meal consumption, physical activity, inter- and intra-subject variations, and significant time delay in diffusion of the infused insulin are still major challenges that are preventing the development of fully-automated AID systems. In this work, a fully-automated personalized adaptive model predictive control (paMPC) approach based on a recursively updated latent variable model is proposed for BGC regulation. In the glucose prediction model, stability of the glycemic dynamics that serves as prior knowledge is incorporated into the modelingprocess to improve its numerical properties and prediction power. Based on the model, a fully personalized MPC (pMPC) with a personalized model and personalized hyperparameters is proposed. Then, adaptive rules in response to unannounced meals and physical activity are proposed and integrated into the pMPC to improve the control effectiveness and reduce the risk of exercise-induced hypoglycemia. The results of a simulation study demonstrate the advantages of the proposed approach in BGC regulation where manual input of meals and physical activity are not needed, achieving full automation of the AID system. The pMPC kept BGC in range for 75.8% of the simulation period and adaptive pMPC for 83.45%.
Professional human drivers usually have more than one driving strategy to handle incoming traffic situations. These different strategies activate different performance characteristics of the vehicle, enabling the driv...
详细信息
ISBN:
(数字)9781665479271
ISBN:
(纸本)9781665479271
Professional human drivers usually have more than one driving strategy to handle incoming traffic situations. These different strategies activate different performance characteristics of the vehicle, enabling the driver to minimize the risk in a variety of situations by optimizing the strategy selection. In the same spirit, we define a novel concept of strategy-wise performance metric and creatively combine this performance metric with reachability analysis to evaluate candidate control strategies. Such a performance evaluation produces solid guarantees on which strategies will not qualify for the given traffic scenario. Then we automate the strategy selection process by weighing and minimizing the overall risk of each strategy candidate.
The metal industry has a critical need for materials to be produced with a more economical method that minimizes or is free of any environmental impact, making the conventional quality control techniques that are heav...
详细信息
ISBN:
(数字)9798331537555
ISBN:
(纸本)9798331537562
The metal industry has a critical need for materials to be produced with a more economical method that minimizes or is free of any environmental impact, making the conventional quality control techniques that are heavily reliant on human inspection laborious, subjective, and prone to human error. This paper discusses how machine learning and data-driven next-generation quality control systems may represent a paradigm shift in the process of metal production. Using real-time data from several production processes, such as temperature, processing conditions, and material composition, the machine learning model can make very accurate forecasts and monitor the quality of the end product. It also adjusts the production process to make the best quality possible in real time, with the aid of pattern detection and anomalies.
TEM automation is dedicated to providing high-volume, fast and precise TEM data to enable semiconductor manufacturers to develop and control fabrication processes. It automates TEM operation and measurement procedures...
详细信息
Incorporating prior knowledge into a data-driven modeling problem can drastically improve performance, relia-bility, and generalization outside of the training sample. The stronger the structural properties, the more ...
详细信息
ISBN:
(数字)9798350382655
ISBN:
(纸本)9798350382662
Incorporating prior knowledge into a data-driven modeling problem can drastically improve performance, relia-bility, and generalization outside of the training sample. The stronger the structural properties, the more effective these improvements become. Manifolds are a powerful nonlinear generalization of Euclidean space for modeling finite dimensions. When additionally assuming that the manifold carries (Lie) group structure, this imposes a drastically stricter global constraint. The range of their applications is very wide and includes the important case of robotic tasks. We apply this idea to Canonical Correlation analysis (CCA). In traditional CCA one constructs a hierarchical sequence of maximal correlations of up to two paired data sets in Euclidean spaces. We here generalize the CCA concept to respect the structure of Lie groups and demonstrate its efficacy through the substantial improvements it achieves in making structure-consistent pre-dictions about changes in the state of a robotic hand.
Current solar analysis and simulation tools are either too optimistic because they ignore neighboring obstructions (e.g. NREL’s PVWatts) or too computationally intensive (e.g. EnergyPlus or other highly-detailed simu...
详细信息
We consider the problem of finding control-oriented models for the electrode resistance of submerged arc furnaces (SAF) to aid the metallurgical processcontrol of ferrosilicon production. To accomplish this goal, we ...
详细信息
ISBN:
(数字)9798350360868
ISBN:
(纸本)9798350360875
We consider the problem of finding control-oriented models for the electrode resistance of submerged arc furnaces (SAF) to aid the metallurgical processcontrol of ferrosilicon production. To accomplish this goal, we analyze the field data gathered from the Norwegian metal producer Wacker AS, which are the most important input parameters for accurately predicting electrode resistances. This is done by investigating the predictive capabilities of different linear and non-linear model structures in different furnace operating conditions, and discussing which type of non-linearity induces the best-performing models both in terms of prediction fit and modeling error in opportune test sets. We finally provide interpretations of why the presence of this non-linearity results in the best performance by connecting their structure with domain expertise about the electrical dynamics within SAF circuits.
As the penetration of inverter-based resources (IBRs) in power grids increases, the need for accurate models becomes important to ensure reliable grid planning and operation. Electromagnetic transient (EMT) studies us...
详细信息
ISBN:
(纸本)9781837242689
As the penetration of inverter-based resources (IBRs) in power grids increases, the need for accurate models becomes important to ensure reliable grid planning and operation. Electromagnetic transient (EMT) studies using detailed models can capture important system dynamics, but these models often lack availability or are computationally demanding. To address these challenges, standardized and flexible generic models have been proposed, though their calibration remains a key issue. This paper presents a parameter estimation framework designed to improve the calibration of generic IBR models, employing Particle Swarm Optimization (PSO). The focus is on integrating data from both Time and Frequency Domain testing to capture the system dynamics, and use these data to calibrate key control parameters. While the framework is mainly conceptualized to utilize data collected from Hardware-in-the-Loop (HIL) setups, it is flexible enough to be adapted to different cases, e.g., including other model parameters, and leveraging different types of data, such as offline simulation or measurement data. The proposed framework is demonstrated through a Type-4 generic wind turbine model, estimating six proportional-integral (PI) control parameters across active power, reactive power, and RMS voltage control loops. The study explores the use of different test cases and sensitivity analysis to optimize the calibration process. Results from proof-of-concept experiments showed that Time Domain data alone could effectively calibrate these parameters, providing a close match to reference values. Due to the overlap between control bandwidths of this specific subset, Frequency Domain testing was not advantageous in this case. The findings suggest that although Frequency Domain data might offer benefits, its application is context-dependent, especially when overlapping control dynamics are present. Future work will expand the framework to integrate all key grid-following control parameters and
As modern software system is growing in size and complexity, the customer expectations for software quality have become higher. In the past, many software reliability growth models (SRGMs) were proposed and they are h...
详细信息
As modern software system is growing in size and complexity, the customer expectations for software quality have become higher. In the past, many software reliability growth models (SRGMs) were proposed and they are helped to evaluate the quality of developed software. It is worth noting that some of SRGMs can be used to model the fault detection process (FDP) and the fault correction process (FCP) through an infinite server queueing (ISQ) system or a finite server queueing (FSQ) system. However, it can also be found that most ISQ and FSQ models were developed on a first come first served basis. In this paper, we propose to use the queueing-based simulations to describe the behavior of FCP and assess the software reliability instead of using model-based approaches. Our proposed queueing-based simulation techniques and simulation procedures will be able to thoroughly investigate the FCP and easily provide system performance information estimated based on the staffing level, the average response time, and the average waiting time. Numerical examples based on three real failure data are given and discussed. Our experiments show that the proposed simulation procedures obtain a good prediction capability for software reliability. We expect that the proposed methods can provide effective information for software developing management and help decision makers in resource allocation and cost control.
暂无评论