Search for: All records

This paper presents the results of a research that created and analyzed a Multimedia dataset for building energy efficiency estimation. First a new Multimedia Building Energy Efficiency (MMBEE) dataset was created from publicly available data. This work then explored the use of the window-to-wall ratio (WWR) information from building facade images and integrated it with traditional tabular data to create new training data, in order to predict building energy efficiency measures. Finally, we discuss potential applications and future research directions in using the MMBEE dataset for building energy efficiency prediction. Throughout the paper, a number of important processes and analyses were performed, which include feature selection, data correlation analysis, WWR extraction, and comparison of deep network and random forest models in building energy efficiency estimation. From this first attempt at using the Multimedia dataset for building energy efficiency estimation, we found the performances of deep models were better than traditional models such as random forest. We also found that there was an optimal point of what features shall be used for the prediction. Nonetheless, the incorporation of the current WWR estimation results did not yield the anticipated enhancement in estimation performance. Subsequently, a comprehensive investigation was conducted to ascertain potential contributing factors, and several avenues for future research were identified to enhance the predictive utility of the WWR feature. 

This paper presents the results of a research that created and analyzed a Multimedia dataset for building energy efficiency estimation. First a new Multimedia Building Energy Efficiency (MMBEE) dataset was created from publicly available data. This work then explored the use of the window-to-wall ratio (WWR) information from building facade images and integrated it with traditional tabular data to create new training data, in order to predict building energy efficiency measures. Finally, we discuss potential applications and future research directions in using the MMBEE dataset for building energy efficiency prediction. Throughout the paper, a number of important processes and analyses were performed, which include feature selection, data correlation analysis, WWR extraction, and comparison of deep network and random forest models in building energy efficiency estimation. From this first attempt at using the Multimedia dataset for building energy efficiency estimation, we found the performances of deep models were better than traditional models such as random forest. We also found that there was an optimal point of what features shall be used for the prediction. Nonetheless, the incorporation of the current WWR estimation results did not yield the anticipated enhancement in estimation performance. Subsequently, a comprehensive investigation was conducted to ascertain potential contributing factors, and several avenues for future research were identified to enhance the predictive utility of the WWR feature. 

The tools and techniques such as imaging and machine learning used in the measurement of many material and microstructural properties are rapidly evolving. In metals, the grain size is routinely measured to estimate the yield strength. This paper describes some of the algorithms used in processing the microstructures to conduct quantitative measurements. The image processing methods provide the possibility to go beyond calculating the ASTM grain size number and calculate the actual surface area of each grain, grain boundary length, and the shape of the grains. The image analysis methods can be very helpful in conducting detailed quantitative analysis with greater accuracy than many labour-intensive manual methods currently in use. The work describes the complexities in applying the imaging methods and approaches in the metallurgical and materials fields. Successful application of such methods can reduce the time and effort required to characterise microstructures and can provide more precise information. 

Abstract This study used radar observations and a high-resolution numerical simulation to explore the interactions between an mesoscale convective system (MCS), cold pool outflows, and atmospheric bores in a non-uniform baroclinic environment. The bores were generated by a nocturnal MCS that occurred on 2–3 June 2017 over the southern North China Plain. The goal of this investigation is to determine how the structure of bores varied within this non-uniform environment and whether and how the bores would maintain the MCS and alter its structure. To the southwest of the MCS, where there was large CAPE and a well-mixed boundary layer, discrete convection initiation occurred behind a single radar fine line (RFL) maintaining the propagation of the MCS. To the southeast of the MCS, multiple RFLs were found suggesting the generation of an undular bore in an environment containing an intense nocturnal stable boundary layer with dry upper layers and little CAPE. Hydraulic and nonlinear theory were applied to the simulation of the MCS revealing that the differences in the bore evolution depended on both the characteristics of the cold pool and the variations in the ambient environment. Thus, the characteristics of the ambient environment and the associated differences in bore structure impacted the maintenance and organization of the MCS. This study implies the importance of an accurate representation of the low-level ambient environment and the microphysics and kinematics within the MCS to accurately simulate and forecast cold pools, the generation and evolution of bores, and their impact on nocturnal MCSs. 

This study analyzes transcripts of conversations in which mathematics teachers and researchers debrief videotaped lessons by, in part, examining aggregated classroom data from the videotaped lesson. We conclude that aggregating data in debrief conversations can support teachers’ concept development when the aggregation a) demonstrates internal contrasts and b) is underscored by participants’ discursive moves. Consequently, we recommend that facilitators seeking to prompt teacher learning use lesson-level aggregations to identify and press on comparisons and distinctions in teaching practice. This study can inform research on teacher learning by unpacking how a common practice—aggregating data—contributes to teachers’ concept development and has implications both for practitioners and for the emerging field of classroom data visualization.