Search Results (76)

Non-intrusive load monitoring (NILM) enables appliance-level disaggregation from a single household meter, yet existing Seq2point-based methods are plagued by inadequate noise robustness, unsatisfactory state recognition, and limited cross-dataset generalization. This paper proposes a denoising multi-task convolutional neural network that fundamentally differs from prior approaches by coupling a denoising autoencoder with task learning through a shared reconstruction head—rather than treating denoising as an isolated preprocessor or simply stacking independent loss branches. This design forces the shared feature extractor to preserve fine-grained temporal signal fidelity while jointly optimizing power regression and state classification, thereby imposing an implicit regularization that suppresses noise interference and enhances transferable representation. The model is evaluated on UK-DALE and REDD datasets, achieving MAE/F1 scores of 12.91 W/84.75% and 5.02 W/96.99%, respectively. Ablation studies confirm the synergistic gains from the joint reconstruction–regression–classification paradigm. Furthermore, statistical analysis across five typical appliance types (e.g., kettle, washing machine, and refrigerator) against three state-of-the-art Seq2Point variants reveals that the proposed method yields consistently superior MAE and F1 improvements with statistical significance (paired Wilcoxon test, p < 0.05) and markedly lower performance variance, demonstrating robust efficacy across diverse load profiles. These results substantiate the proposed model as a reliable and statistically validated solution for fine-grained residential energy management in smart grids. Full article

In submitting conference proceedings to Engineering Proceedings, the Volume Editors of the proceedings would like to certify to the publisher that all papers published in this volume have been subjected to peer review by the designated expert referees and were administered by the Volume Editors in a double-blind peer review [...] Full article

Background/Objectives: Irritable bowel syndrome (IBS) is among the most prevalent disorders of gut–brain interaction, yet its implications for spine surgery remain poorly characterized. This narrative review examines how IBS influences symptom presentation and postoperative outcomes in spine surgery patients. Methods: We synthesized the neurobiologic, epidemiologic, and perioperative literature linking IBS with musculoskeletal pain, spine-related symptomatology, and surgical outcomes, drawing on spine-specific data where available and on related surgical and chronic-pain populations where it was not. Results: IBS is characterized by central sensitization, impaired descending inhibition, increased temporal summation, autonomic dysregulation, and a high prevalence of psychiatric comorbidity, which manifest as widespread hyperalgesia and symptom amplification that overlap with pain mechanisms common in spine surgery patients. Epidemiologic studies indicate that patients with IBS undergo musculoskeletal and spinal procedures at disproportionately high rates, reflecting both symptom burden and diagnostic uncertainty from viscerosomatic overlap. These same factors have been associated with greater postoperative pain, elevated opioid requirements, slower functional recovery, and reduced satisfaction after spine surgery, although direct IBS-specific spine data remain limited. IBS may also confound preoperative assessment by mimicking radicular, discogenic, or sacroiliac pain. Conclusions: IBS represents an under-recognized potential modifier of symptom localization, perioperative pain trajectories, and functional recovery in spine surgery. Greater awareness of IBS-related nociplastic and psychosocial mechanisms may improve preoperative evaluation, risk stratification, perioperative management, and the design of future outcome studies. Full article

Non-Intrusive Load Monitoring (NILM) enables estimation of the energy use of individual appliances in smart buildings from a single aggregate meter. In practice, however, this task is not straightforward. Signals from different appliances can overlap, and the measured data may also include distortions such as spikes, drops, and noise. To address these issues, this study presents a multi-task triple-hybrid deep learning framework that handles appliance classification and anomaly detection together. The model brings together 1D-CNN, BiLSTM, and Transformer Attention so that local patterns, temporal dependencies, and wider contextual information can be learned within the same structure. It also uses a dual-output design to classify appliance categories and detect anomaly types simultaneously. Experiments were carried out on Building 1 of the UK-DALE dataset with four appliances: kettle, microwave, washer dryer, and fridge freezer. For the anomaly task, synthetic disturbances were added to segmented signal windows and grouped as normal, spike, drop, and noise. To check how well the proposed framework handled different scenarios, it was tested on both the UK-DALE and REDD datasets. Looking at the main UK-DALE results, the model correctly identified appliances 99.48% of the time and spotted anomalies with 98.80% accuracy. A secondary test on the REDD dataset yielded an 86.44% classification score. This proves the architecture can adjust to completely new power grid environments without losing its edge. On top of that, when pitted against standard benchmark models like Seq2Point, this triple-hybrid design clearly does a better job of mapping out complex signal changes. As a result, it yields much stronger anomaly detection metrics. Full article

Additive manufacturing is transforming high-frequency electronics prototyping by offering a sustainable and cost-effective alternative to traditional methods. This work addresses and demonstrates two areas: the use of 3D printing for millimeter-wave (mmWave) antennas, and chip-to-chip or chip-to-PCB interconnects. Both approaches facilitate reduced material waste. A 47 GHz series-fed microstrip patch array was printed on flexible Kapton using aerosol jet technology, showing performance comparable to etched arrays on Roger’s substrates. Crucially, the Kapton film can be peeled off after testing, allowing the reuse of expensive low-loss substrates. Therefore, this method supports rapid, low-waste prototyping. To address future chip-to-chip and chip-to-PCB mmWave interconnect limitations, XTPL’s Ultra-Precise Dispensing (UPD) was used to fabricate 3D-printed micro-interconnects. At 73 GHz, these interconnect structures achieved return loss better than 10 dB and insertion loss under 1 dB—outperforming traditional bondwires. Together, these results show 3D printing’s potential to enable sustainable, high-performance mmWave RF systems. Full article

Appliance recognition from aggregate household measurements is challenging under real deployment conditions, where multiple devices operate concurrently and sensor data are affected by imperfections such as noise, missing samples, and nonlinear meter response. In contrast to many studies that rely on curated or idealized datasets, this work investigates appliance recognition using real multimodal utility data (electricity, water, gas) collected at the building entry point, in the presence of substantial uninstrumented background activity. We present a case study evaluating transparent, rule-based detectors designed to exploit characteristic temporal dependencies between modalities while remaining interpretable and robust to sensing imperfections. Four household appliances—washing machine, dishwasher, tumble dryer, and kettle—are analyzed over six weeks of data. The proposed approach achieves reliable detection for structured, water-related appliances (22/30 washing cycles, 19/21 dishwashing cycles, and 23/27 drying cycles), while highlighting the limitations encountered for short, high-power events such as kettle usage. The results illustrate both the potential and the limitations of conservative rule-based detection under realistic conditions and provide a well-documented baseline for future hybrid systems combining interpretable rules with data-driven adaptation. Full article

Powder addition onto a molten-lead surface followed by stirring is widely used for desilvering during lead bullion refining operations. We model submerged zinc particle injection by coupling (i) a transient particle–metal reaction following Ohguchi with a time-dependent reaction efficiency E, (ii) a Stefan-type estimate of the zinc melting time Tf, and (iii) hydrodynamic descriptors of residence (τ_res) and mixing (τ_mix) times. The model is validated against experiments under a benchmark condition (gas velocity U = 3.32 m/s, 70% submergence), achieving a mean absolute percentage error of 1.13% for the experimental desilvering curve. A parametric study over lance submergence (30–90% of bath depth), injection velocity (3.32–9.79 m/s), and geometric scalings of lance and kettle identifies conditions where the hydrodynamic residence time τres approaches the Stefan melting time, maximizing liquid-Zn contact with molten Pb. Specifically, the proposed optimum balances the competing effects of plume buoyancy at high velocities—which tends to reduce residence time—against the deeper injection depth, ensuring that particles remain submerged long enough to fully melt and react. Within 16 simulated scenarios, the pair “90% submergence + U = 9.79 m/s” provides the best multi-criteria performance (desilvering fraction, E, and residence time) under realistic constraints. A parametric sensitivity analysis ranks injection velocity and submergence as the dominant levers, with geometry playing a secondary role over the tested ranges. The coupled hydrodynamic–kinetic framework provides quantitative guidance for optimizing industrial desilvering by particle injection and is extensible to other powder-injection refining operations. Full article

In China, crude phenols, mixtures commonly produced in the coal industry, are inexpensive and abundant in supply, but their valorization is hindered by high energy consumption in the separation process. It is of great academic and commercial significance to improve the separation process of crude phenols to achieve energy efficiency and cost reduction. In this study, a heat pump-assisted distillation (HPD) system for crude phenol separation was developed. External vapor recompression was adopted due to the strong corrosiveness, high toxicity, heat sensitivity, and easy polymerization of crude phenols. Compared with conventional distillation (CD), HPD showed clear advantages in lowering operating costs. The effects of design variables including pressure, the number of theoretical plates and temperature differences between the condenser and reboiler on reflux ratios, kettle temperature, equipment costs, operating costs, and total annual cost (TAC) were investigated and optimized in detail. The effect of steam prices on process economic feasibility was also studied. It was found that HPD reduced at least 55% of the operational cost compared to CD when the steam price was higher than 10.8 USD/GJ. Carbon emission evaluation indicated that CO₂ generated by the HPD process was 56.3% lower than CD. Full article

Sour beer production is strongly influenced by the choice of lactic acid bacteria (LAB), yet few studies have systematically compared strain-specific contributions under controlled kettle souring conditions. This study evaluated the fermentation performance and flavor-modulating potential of three LAB species—Lacticaseibacillus paracasei, Pediococcus pentosaceus, and Leuconostoc mesenteroides—in sour India Pale Ale (IPA) brewing. Growing assessments showed that P. pentosaceus exhibited the most rapid and stable proliferation, while L. mesenteroides required a longer adaptation period. Acidification trials demonstrated that L. paracasei achieved the lowest pH (3.26–3.43), contributing to intense sourness, whereas P. pentosaceus and L. mesenteroides yielded milder acidity (pH 3.41–3.65). Gas chromatography-mass spectrometry showed that P. pentosaceus and L. mesenteroides produced significantly higher levels of fruity and floral esters, including 2-pentanol propanoate, which was approximately 4-fold higher than in the control. Principal component analysis further distinguished the beers according to their volatile profiles. These findings highlight the strain-specific potential of LAB in sour beer brewing and provide practical guidance for flavor differentiation in craft beer production. Full article

Water reservoirs play a crucial role in the environment in many aspects: hydrology, geochemistry, sediment lithology, geo- and biodiversity, landscape, etc. First of all, it is necessary to have accurate information about the spatial distribution of these objects in a given area to assess their size and functioning. Maps and contemporary spatial databases are often incomplete or outdated, especially in regard to small objects, of variable surface area and condition. This article uses the following approach: high-resolution terrain models derived from airborne laser scanning (ALS) were used for visual interpretation of extensive, flat depressions representing water body basins, thus determining the total number of objects, and classifying them as kettle holes, lakes, ponds, and other types of reservoirs (e.g., overbank basins, oxbow lakes). Using an aerial orthophotomap, the objects were subsequently verified as to how many basins are currently occupied by water bodies. The next step was to determine a number of topographic and morphometric parameters for each object in order to assess their functioning conditions. For selected objects, the assessment was expanded to include a geochemical and lithological analysis of the sediments. The study was conducted in the catchment of the Słubia River (136 km²), located in Central Europe, in northwestern Poland. In the Słubia catchment, a total of 931 water body basins were mapped. The dominant forms are kettle holes (<1 ha), representing nearly 80% of all objects. At present, kettle holes are largely devoid of water bodies and subject to a strong human impact. In addition to those, 118 lake basins were identified (>1 ha, the largest being Lake Morzycko, 360 ha), half of which are occupied by water reservoirs. Ponds and other reservoirs were represented by 37 and 47 objects, respectively. From the perspective of contemporary sediment-forming processes in the documented sedimentary basins, the most favorable conditions for biogenous sediment accumulation exist in the catchments of the upper and medium courses of the Słubia River valley. Although the lithological diversity and thickness of individual sediment types in the Słubia catchment represent local features, they corroborate the results of previous telmatologic research conducted in Myślibórz Lakeland. Full article

Load disaggregation has emerged as an effective tool for enabling smarter energy management in residential and commercial buildings. By providing appliance-level energy consumption estimation from aggregate data, it supports energy efficiency initiatives, demand-side management, and user awareness. However, several challenges remain in improving the accuracy of energy disaggregation methods. For instance, the amount of noise in energy consumption datasets can heavily impact the accuracy of disaggregation algorithms, especially for low-power consumption appliances. While disaggregation performance depends on hyperparameter tuning, the influence of data characteristics, such as noise, on hyperparameter selection remains underexplored. This work investigates the hypothesis that appliance-specific noise information can guide the selection of algorithm hyperparameters, like the input sequence length, to maximize disaggregation accuracy. The appliance-to-noise ratio metric is used to quantify the noise level relative to each appliance’s energy consumption. Then, the selection of the input sequence length hyperparameter is investigated for each case by inspecting disaggregation performance. The results indicate that the noise metric provides valuable guidance for selecting the input sequence length, particularly for user-dependent appliances with more unpredictable usage patterns, such as washing machines and electric kettles. Full article

2-Methyl-2-cyclopentenone (MCP) is the main by-product of the newly developed heterogeneous catalysis process for producing crotonaldehyde, which serves as an important intermediate for drug synthesis. However, how to recover and purify MCP from the product mixture is not known. To decipher this, a computer-aided process based on the measured phase-equilibrium data was developed. The improved Rose–Williams equilibrium kettle was used to measure the vapor–liquid equilibrium data for MCP–crotonaldehyde and MCP–water. Surprisingly, MCP and water formed a minimum azeotrope, which aided its own recovery from its dilute solution. The mole fraction of MCP in the azeotrope was 9.1%, the mole fraction of water was 90.9%, and the azeotropic temperature was 96.8 °C. Equilibrium data from the two binary systems were correlated using the Wilson and NRTL activity coefficient models. The NRTL-RK model was selected to simulate the process for recovering and purifying MCP. A two-column process was developed and optimized in this study, and the aim of effectively utilizing the by-product MCP was achieved with this process. Full article

The purpose of this investigation was to utilize pulsed electric field (PEF) technology to make sweet potato kettle chips (SPKC) healthier by lowering the amount of oil absorbed and reducing the amount of acrylamide formed during frying. Sweet potatoes were treated continuously in an Elea PEF Advantage Belt One system and prepared as SPKC, without peeling and sliced to a thickness of 1.7 mm. The specific energy for PEF application was set to either low (1.5 kJ/kg) or high (3.0 kJ/kg) with a field strength of 1.0 kV/cm and a pulse width of 6 μm. Batches of 500 g unrinsed potato slices were fried in canola oil at 130 °C for 360 s. The oil content in 3.0 g of fried SPKC was 1.39 g or 46.3%, whereas the oil content was 37.9% for high and 37.7% for low PEF-treatment conditions. Acrylamide (AA) in the fried SPKC was quantified by mass spectrometry to be 0.668 μg/g in the non-PEF control and 0.498 μg/g for low and 0.370 μg/g for high PEF treatment. The results of this study support the use of PEF in SPKC processing to reduce oil absorbance during frying by up to 9% and lower AA by up to 45%. Full article

The vapor–liquid equilibrium (VLE) data of the ternary system methanol–methyl formate–methylal was measured at atmospheric pressure using a modified Rose equilibrium kettle with vapor–liquid double circulation method. The experiment data were correlated with the NRTL, UNIQUAC, and Wilson activity coefficient model equations. The results shown that the root mean square deviation (RMSD) between the calculated and simulated values of the three models followed the order: UNIQUAC ≈ NRTL < Wilson, and except for the RMSD (T) in the range of 0.4–0.5, the others are less than 0.01. In addition, the NRTL model was selected to link with Aspen Plus software to simulate the separation process of polyformaldehyde (POM) waste liquid. The simulation results show that the methyl formate in POM waste stream can be purified by simple distillation, while the methylal separated from the POM waste liquid, which was affected by factors like the azeotropic behavior of binary components, necessitates a complex distillation process. Under optimal operating conditions, the recovery yield of methyl formate through direct distillation can reach 99.7%, with an economic benefit of 6960.1 CNY per ton of waste liquid. Although the economic benefit of the multi-component distillation reach 7281.2 CNY, the increase in the number of equipment and the complexity of the process have negative impacts. Full article