다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Accurate health estimation and lifetime prediction of lithium-ion batteries are crucial for durable electric vehicles. Early detection of inadequate performance facilitates timely maintenance of battery systems. This reduces operational costs and prevents accidents and malfunctions. Recent advancements in “Big Data” analytics and related statistical/computational tools raised interest in data-driven battery health estimation. Here, we will review these in view of their feasibility and cost-effectiveness in dealing with battery health in real-world applications. We categorise these methods according to their underlying models/algorithms and discuss their advantages and limitations. In the final section we focus on challenges of real-time battery health management and discuss potential next-generation techniques. We are confident that this review will inform commercial technology choices and academic research agendas alike, thus boosting progress in data-driven battery health estimation and prediction on all technology readiness levels.

/pdf/data-driven-health-estimation-and-lifetime-prediction-of-12knsapmli.pdf

Data-driven health estimation and lifetime prediction of lithium-ion batteries: A review

https://www.cell.com/joule/pdf/S2542-4351(19)30585-9.pdf

Battery Lifetime Prognostics

A review of the state of health for lithium-ion batteries: Research status and suggestions

A novel prediction method based on the support vector regression for the remaining useful life of lithium-ion batteries

A novel particle filter approach for indoor positioning by fusing WiFi and inertial sensors

Knowledge distilling based model compression and feature learning in fault diagnosis

Global navigation satellite system (GNSS) signal is vulnerable and easily interfered by spoofing because of its opening signal structure and weak signal power. A signal quality monitoring (SQM) technique has been shown to be viable to detect spoofing attacks on GNSS signals. However, the effectiveness of spoofing detection employing a single SQM metric alone is limited and conditional upon the features extracted of that specific SQM metric. The complementary features among various SQM metrics can be exploited to implement joint detection to overcome the deficiency of the individual SQM metric. Motivated by this idea, this paper investigates the multi-metric joint detection technique, which combines various SQM metrics into a composite SQM metric to detect spoofing attacks. This paper proposes two combination strategies, namely amplitude combination mode and probability of false alarm combination mode (PfaM). The overall performance of different metric combinations was verified using simulations and the Texas Spoofing Test Battery dataset. Results show that the PfaM detector outperforms all single SQM metric detectors under various scenarios.

GNSS Spoofing Detection by Means of Signal Quality Monitoring (SQM) Metric Combinations

Signal quality monitoring (SQM) techniques, originally designed for multipath detection, were recently found to be useful to identify underway spoofing attacks. Conventional SQM-based methods directly employ the values of the SQM metrics to monitor spoofing attacks. They have good feasibility with simple structures but suffer from significant performance loss for frequency unlocked spoofing cases due to the drift of the relative carrier phase. We developed an enhanced SQM technique for detecting an onset of spoofing. It is known that the value of the SQM metric fluctuates significantly during the interaction stage between the counterfeit signal and authentic signal. As the variance of metric can better reflect this fluctuation, we choose the moving variance (MV) of the SQM metric as a new metric to detect the occurrence of spoofing. The basic principle of the proposed method is introduced. Its ability to detect spoofing has been validated using the Texas Spoofing Test Battery dataset and compared with the classic SQM methods and a moving average-based method. The results show that the proposed MV-based SQM method is advantageous in the detection of an onset of a frequency unlocked spoofing attack.

Wenquan Feng

Papers

A novel prediction method based on the support vector regression for the remaining useful life of lithium-ion batteries

A novel particle filter approach for indoor positioning by fusing WiFi and inertial sensors

Knowledge distilling based model compression and feature learning in fault diagnosis

GNSS Spoofing Detection by Means of Signal Quality Monitoring (SQM) Metric Combinations

Moving variance-based signal quality monitoring method for spoofing detection