Search

This article demonstrates an associative latent transition-based model for analyzing the association between two discrete-time survival analysis (DTSA) models. DTSA allows social and behavioral researchers to investigate the qualitative change in event occurrence. Typically, DTSA is applied to a single survival process, where the hazard of an event occurrence is estimated across time. However, when researchers have hypotheses concerning the relationship between two survival processes, DTSA needs to be extended to associative DTSA. The study describes the latent transition based DTSA and its extension to explore the association between two event occurrences. M + software is used to illustrate the model parameterizations that are applicable to social and behavioral research topics. The illustrations use data related to juvenile delinquent and criminal behaviors. We close with a discussion of the capabilities of latent transition based DTSA. [ABSTRACT FROM AUTHOR]

Objective: Neural network deidentification studies have focused on individual datasets. These studies assume the availability of a sufficient amount of human-annotated data to train models that can generalize to corresponding test data. In real-world situations, however, researchers often have limited or no in-house training data. Existing systems and external data can help jump-start deidentification on in-house data; however, the most efficient way of utilizing existing systems and external data is unclear. This article investigates the transferability of a state-of-the-art neural clinical deidentification system, NeuroNER, across a variety of datasets, when it is modified architecturally for domain generalization and when it is trained strategically for domain transfer.Materials and Methods: We conducted a comparative study of the transferability of NeuroNER using 4 clinical note corpora with multiple note types from 2 institutions. We modified NeuroNER architecturally to integrate 2 types of domain generalization approaches. We evaluated each architecture using 3 training strategies. We measured transferability from external sources; transferability across note types; the contribution of external source data when in-domain training data are available; and transferability across institutions.Results and Conclusions: Transferability from a single external source gave inconsistent results. Using additional external sources consistently yielded an F1-score of approximately 80%. Fine-tuning emerged as a dominant transfer strategy, with or without domain generalization. We also found that external sources were useful even in cases where in-domain training data were available. Transferability across institutions differed by note type and annotation label but resulted in improved performance. [ABSTRACT FROM AUTHOR]

Ventilation systems depend on periodic preventive maintenance. Unfortunately, mission critical systems, exposed to varying levels of particulates and other contaminants, may require unpredictable and highly variable maintenance intervals. Direct sensing for pressure drops or other condition metrics can provide prognostic indicators, but at the expense of the installation and maintenance burden of the sensors themselves. Rotor slot harmonics from motors have been observed for decades to track motor speed. This paper presents hardware instrumentation and a signal processing algorithm that can nonintrusively track the rotor slot harmonics of multiple fan motors in an aggregate electrical stream. Combined with a physics-based model of the motors under observation, this nonintrusive data can be used to track the maintenance condition of multiple fan motors on a single electrical service, minimizing the costs of installation and data analysis and reducing the total number of sensors necessary to track fan health. Results are demonstrated with experiments on-board a US Coast Guard Cutter. [ABSTRACT FROM AUTHOR]

This article presents a design for data acquisition hardware that can automatically track and reject utility frequency components of a measured current or voltage waveform. Measurements of subtle harmonic content in the waveform can therefore be made with the full dynamic range of the on-board analog-to-digital converter. This permits the detection of small, higher frequency signals for applications to condition-based maintenance and energy scorekeeping. Experimental demonstrations in this article show the hardware prototype isolating the rotor slot harmonics in a motor’s input current. The high-resolution measurements of slot harmonics permit the estimation of motor speed from the detected slot harmonic frequency and the rejected fundamental frequency. The performance and accuracy of the speed estimator are demonstrated with both steady-state and transient experiments. [ABSTRACT FROM AUTHOR]

This paper proposes a method to boost the terminal voltages using a capacitive network. The boosted voltage can improve the power capability of an interior permanent magnet synchronous motor (IPMSM). In the conventional drive system fed by a voltage-source inverter, the terminal voltages of the motor are limited by the dc-link voltage. Such a voltage limitation reduces the torque capability at high speeds. Thereby, the maximum output power is limited. To overcome this feature, series capacitor network between the inverter and the motor is introduced in the proposed IPMSM drive system. The series capacitors boost the motor terminal voltages higher than the inverter voltages to increase the rated speed, resulting in improved power capability. The rated speed in the proposed drive system can be redefined to the desired value through a capacitive network. The enhanced power capability and design freedom in the rated speed are particularly advantageous in fan-load applications where the load torque is proportional to the square of speed. This article presents the structure, d–q equivalent model, operating characteristics, and operable area of the proposed drive. Experimental results are given to verify the feasibility and effectiveness of the proposed system. [ABSTRACT FROM AUTHOR]

This article proposes a method to extend the operating area of the surface permanent magnet motor by inserting the passive power networks between the motor and the voltage source inverter (VSI). In the conventional motor drive system, since the operating area is limited by the maximum voltage, which is synthesized by the inverter, the output torque capability is reduced in the high-speed region. In order to extend the operating area, two types of passive power networks are proposed in this article. The power networks supply the reactive power to the motor for amplifying the motor terminal voltage. Due to the actions, the proposed motor drive system can utilize a higher maximum motor terminal voltage than that in the conventional drive systems. Depending on the network types, the operating point at which the maximum power or torque is generated is shifted to a high speed. The design, control, and analysis methods of the proposed system are presented in this article. The effectiveness of the proposed system is verified through the experimental results. [ABSTRACT FROM AUTHOR]

This paper proposes a d–q cross-coupling current controller for permanent magnet synchronous motor (PMSM) drives. Slotting and cross-coupling effects in PMSMs cause the effective air-gap length to vary with the rotor angle and distort the distribution of the air-gap flux. Such distortions have a significant influence on the current control by changing the relationship between the stator currents and voltages. To investigate their effects, a d–q model is established in this paper, which is suitable for describing the voltage characteristics of the actual PMSMs. The model contains the angle-dependent parameters, the cross-coupling inductance between d–q axes, and the q-axis permanent magnet flux linkage, unlike the ideal d–q model. The transfer function matrix from the d–q currents to d–q voltages of PMSM is newly derived from the model. Based on that, the current controller with the feedback and feedforward terms of 2-by-2 matrix is proposed. It is designed so that the independent gains are in low-pass filter forms and the coupling control gains are zero with eliminating interference between d–q axes. The proposed controller contributes to providing the stable control performance by canceling the slotting and cross-coupling effects. The performance of the proposed current controller is verified by experimental results. [ABSTRACT FROM AUTHOR]

De-identification aims to remove 18 categories of protected health information from electronic health records. Ideally, deidentification systems should be reliable and generalizable. Previous research has focused on improving performance but has not examined generalizability. This paper investigates both performance and generalizability. To improve current state-ofthe- art performance based on long short-term memory (LSTM) units, we introduce a system that uses gated recurrent units (GRUs) and deep contextualized word representations, both of which have never been applied to de-identification. We measure performance and generalizability of each system using the 2014 i2b2/UTHealth and 2016 CEGS N-GRID deidentification datasets. We show that deep contextualized word representations improve state-of-the-art performance, while the benefit of switching LSTM units with GRUs is not significant. The generalizability of de-identification system significantly improved with deep contextualized word representations; in addition, LSTM units-based system is more generalizable than the GRUs-based system. [ABSTRACT FROM AUTHOR]

This paper proposes a novel drive method for interior permanent magnet machines (IPMMs). In the proposed system, a two-phase IPMM is fed from a single-phase H-bridge inverter for low-cost applications. Its main winding in d -axis is connected to the inverter and the auxiliary winding in q-axis is inherently short circuited. This configuration enables the motor to operate in a wider speed range with higher energy efficiency than most other single-phase drive systems. Also, its torque capability in the high-speed range is as high as that of the three-phase IPMM drive despite the reduced power devices, drivers, and current sensors. Accordingly, the proposed system has benefits such as simple structure, low cost, high-power density, and improved efficiency. In this paper, the system is modeled with an equivalent three-phase machine. Based on the model, its torque characteristics and operating areas are analyzed and compared to those of the conventional IPMM drives. A speed control algorithm is also developed. It includes a maximum torque per ampere control to improve drive efficiency and a flux-weakening control to maximize the feasible inverter output voltage. The proposed system and control algorithm are verified by experimental results. [ABSTRACT FROM PUBLISHER]

This paper proposes a single-phase grid-connected wound rotor machine that has the single-phase stator and three-phase rotor windings. The machine has no auxiliary winding or capacitor unlike conventional single-phase machines. Nevertheless, it can operate in all four quadrants of the torque–speed plane with a three-phase inverter. For adjustable speed drives, an isolated three-phase inverter is applied to the rotor windings while the stator winding is directly connected to a single-phase source. The grid filter and rectifier of the conventional system are eliminated and the rotor-side slip rings can be also removed by the inverter integration. So the overall structure of the proposed drive system is simple and cost effective. In this paper, the proposed machine is modeled into a modified d–q model considering the absence of q-axis stator coil. Its characteristics are analyzed and vector control methods of the grid power factor, dc-link voltage, and speed are proposed. For more efficient control, the optimal rotor current set is calculated from the minimum copper loss condition. The system has wider operating areas than other single-phase drive systems. The feasibility of the proposed system is verified by experiments. [ABSTRACT FROM AUTHOR]