An incremental machine learning approach is proposed to quantify the influence of occupants' gender, age, and nationality on predictive models of occupant indoor comfort, perceived productivity, and perceived happiness. A three-step methodology is proposed, including: (1) data collection through sensors and a questionnaire; (2) development and comparison of machine learning models; and (3) incremental introduction of demographic variables while observing its influence in the the model's performance. Results confirm that the inclusion of demographical variables in the models can improve prediction accuracy (F1 scores) by as much as 10%. However, reductions in the models' performances were observed in some instances, highlighting the need to strike a balance between increasing model complexity and the subsequent rise in computational costs.
Producing precise and complex functional parts for industrial application with different types of material has been an enormous challenge in the additive manufacturing industry. Several attempts have been made to optimize the quality of the fabricated parts by understanding the relationship between the input and output process parameters. The main objective of this paper is to construct a reliable data set utilizing literature data and to obtain a linear regression model that explains the interaction between the laser power, scan speed, hatch distance, and layer thickness with the relative density of fabricated steel parts. The R2 value of the linear model revealed that 77.36% of the data fit the regression model and can be utilized in future optimization processes.
This research proposes an optimization model for the economic statistical design of the np chart for detecting shifts in fraction nonconforming. In this model, the charting parameters of the np chart are optimized to minimize the Expected Total Cost (ETC) and in the meantime, ensure that the false alarm rate and the inspection rate do not exceed the allowable levels. It is found that the proposed optimal np chart considerably outperforms the traditional np chart.
This paper proposes a new combined scheme (termed as Mnp-EWMA scheme) for monitoring multi-attribute characteristics. The proposed Mnp-EWMA scheme combines both the multi-attribute np (Mnp) and multi-attribute Exponentially Weighted Moving Average (MEWMA) charts for effective monitoring of fraction nonconforming. In this study, the Average Number of Defectives (AND) is used as an objective function in an optimization model to identify the best design variables of Mnp-EWMA scheme. The performance of the proposed Mnp-EWMA scheme is compared with that of the Mnp and MEWMA charts. The results show that the Mnp-EWMA chart outperforms the Mnp and MEWMA charts under different scenarios.
Fresnel lenses have many solar energy-based applications and focus light from x-rays to extreme ultraviolet (UV) radiation. Typically, solar concentrators are based on large-scale Fresnel lenses. The Fresnel lens novel design for a targeted application is very prominent among researchers. In this work, using a novel micro-Fresnel lens in healthcare applications for pressure sensing is proposed. Such lenses might be used separately or combined with commercial contact lenses for optical sensing purposes. These lenses have a significant role in ophthalmologic disease detection, such as glaucoma. For this purpose, a flat Fresnel lens is designed and simulated to observe focusing ability and investigate focal length variation with geometric lens parameters. A flat Fresnel lens is transformed into a curved one, in analogy to a contact lens uses for vision correction, therapeutics, and cosmetics. This work also opens the possibilities of additively manufactured patient-specific contact lenses for sensing applications.
Diffusive reflection spectroscopy is employed to study the influence of probe pressure on human skin and nails. A non-invasive study is conducted which covers the important body parts for in-vivo measurements. Reflection spectra were measured for fingertips, forearm, and nails. A set of probe pressure ranging from 0 to 258 kPa for the finger and 0 to 385 kPa for the nail is applied. It is found that all the measured spectra revealed the increase in the reflection intensity under increasing pressure in the specific wavelength region (525-575 nm - oxyhemoglobin region). Since every tissue has its own composition and morphology so the shape, size, intensity, and position of the measured peaks are affected. Time-based reflection spectroscopy was also performed on the forearm under blood occlusion for 5 mins with an interval of 30 sec.
Anatase TiO2 nanosheets with high {001} facet exposition were synthesized via a one-pot hydrothermal method. Commercial WO3 nanopowders were then mechanically mixed with the as-synthesized TiO2 to prepare WO3/TiO2 hybrid samples and build the heterojunctions. The physicochemical properties of the samples were characterized by numerous methods, including XRD, Raman, BET, PL, SEM, and TEM. The photocatalytic performance was tested by the oxidation of CO under UV light. The effects of annealing temperature and the mass ratio of WO3 to TiO2 on the photocatalytic activity were also investigated. The results illustrated that pure TiO2 without annealing processing showed the highest oxidation rate compared to all other samples, suggesting that the charge separation by facets is more efficient than WO3-TiO2 heterojunctions.
Synthesis of Fe3O4-Ag nanoparticles (NPs) was implemented via utilizing the method of coprecipitation for synthesizing the Fe3O4 nanoparticles, and reducing silver nitrate (AgNO3) to Ag-nanoparticles, and impregnating the Ag NPs into the Fe3O4 NPs. The synthesized composite material was characterized using the FT-IR, XRD, and the contact angle. The results showed that the Ag NPs were attached to the Fe3O4, and that the composite nanoparticles exhibit hydrophilic behavior. Such properties make these materials a great candidate for oil-water separation related applications.
Three-dimensional (3D) structures fabricated from a 2D material such as graphene have recently attracted huge attention owing to the outstanding electro-mechanical and thermal properties of various types of 2D materials. The interest has expanded to integrate the 2D material into 3D printing structures to form an architected foam. Several studies have focused on fabricating 3D structures from 2D materials using different techniques. In this work, the self-assembly hydrothermal-assisted dip-coating technique has been employed to fabricate graphene foams. The graphene is coated on a 3D printed polymeric scaffold that takes the topology of the mathematically-known triply periodic minimal surfaces (TPMS). A series of tests were performed to measure the multifunctional (electrical, thermal, mechanical) properties of the TPMS-based graphene foam. The results show that these lightweight 3D graphene foams (3DGF) can be used in various applications.
The use of iron-based metal-organic frameworks (MOFs) coated with PEG-folated liposomes is investigated as novel hybrid anticancer nanocarriers. Specifically, using the folate moiety for targeting cancerous cells and ultrasound as an external triggering mechanism is studied. The experimental work examines the encapsulation efficiencies and release profiles of the prepared liposome-coated MOF. The experiments were carried out using an anticancer drug (namely Doxorubicin, DOX) at two different pH values (5.3 and 7.4) with ultrasound. Preliminary results show that the encapsulation efficiency was in the range of 96-98% and the release efficiency was between 96-98% at the investigated pH values within 180 mins.
