Machine learning algorithm selection for windage alteration fault diagnosis of mine ventilation system

Machine learning algorithms have been widely used in mine fault diagnosis. The correct selection of the suitable algorithms is the key factor that affects the fault diagnosis. However, the impact of machine learning algorithms on the prediction performance of mine fault diagnosis models has not been fully evaluated. In this study, the windage alteration faults (WAFs) diagnosis models, which are based on K-nearest neighbor algorithm (KNN), multi-layer perceptron (MLP), support vector machine (SVM), and decision tree (DT), are constructed. Furthermore, the applicability of these four algorithms in the WAFs diagnosis is explored by a T-type ventilation network simulation experiment and the field empirical application research of Jinchuan No. 2 mine. The accuracy of the fault location diagnosis for the four models in both networks was 100%. In the simulation experiment, the mean absolute percentage error (MAPE) between the predicted values and the real values of the fault volume of the four models was 0.59%, 97.26%, 123.61%, and 8.78%, respectively. The MAPE for the field empirical application was 3.94%, 52.40%, 25.25%, and 7.15%, respectively. The results of the comprehensive evaluation of the fault location and fault volume diagnosis tests showed that the KNN model is the most suitable algorithm for the WAFs diagnosis, whereas the prediction performance of the DT model was the second-best. This study realizes the intelligent diagnosis of WAFs, and provides technical support for the realization of intelligent ventilation.     

Synergistic effect of high-intensity ultrasound and β-cyclodextrin treatments on browning control in apple juice

This work evaluated the synergistic effects of combined high-intensity ultrasound (HIU) with β-cyclodextrin (β-CD) treatments on inhibiting browning of apple juice and explored the mechanism through simulation system. The combined treatment of 300 W HIU with 0.006 g mL⁻¹ β-CD had a synergistic impact on maintaining juice colour, resulting in a 39.06% reduction in browning degree, only a 36.64% decrease in total phenolic content, and a 17.82% reduction in PPO activity. The inhibition of enzymatic browning in simulated system revealed that HIU suppressed the enzyme (Polyphenol oxidase, PPO) and β-CD inhibited enzyme (PPO) and embedded substrate (polyphenol). The results of spectroscopic analysis showed that the particle-size distribution of PPO narrowed, the content of α-helix in the secondary structure increased, the fluorescence intensity increased, and the maximum wavelength was red-shifted after HIU and β-CD treatment. Changes in structure could further result in PPO activity loss. Hence, the combined treatment could synthetically alleviate the browning of apple juice.     

Reusable Polyacrylonitrile-Sulfur Extractor of Heavy Metal Ions from Wastewater

Mercury, lead, and cadmium are among the most toxic and carcinogenic heavy metal ions (HMIs), posing serious threats to the sustainability of aquatic ecosystems and public health. There is an urgent need to remove these ions from water by a cheap but green process. Traditional methods have insufficient removal efficiency and reusability. Structurally robust, large surface-area adsorbents functionalized with high-selectivity affinity to HMIs are attractive filter materials. Here, an adsorbent prepared by vulcanization of polyacrylonitrile (PAN), a nitrogen-rich polymer, is reported, giving rise to PAN-S nanoparticles with cyclic π-conjugated backbone and electronic conductivity. PAN-S can be coated on ultra-robust melamine (ML) foam by simple dipping and drying. In agreement with hard/soft acid/base theory, N- and S-containing soft Lewis bases have strong binding to Hg²⁺, Pb²⁺, Cu²⁺, and Cd²⁺, with extraordinary capture efficiency and performance stability. Furthermore, the used filters, when collected and electrochemically biased in a recycling bath, can release the HMIs into the bath and electrodeposit on the counter-electrode as metallic Hg⁰, Pb⁰, Cu⁰, and Cd⁰, and the PAN-S@ML filter can then be reused at least 6 times as new. The electronically conductive PAN-S@ML filter can be fabricated cheaply and holds promise for scale-up applications.     

Effect of interface types on the heat-resistance of Cansas-II SiCf /SiC composites

The heat-resistance of the Cansas-II SiC/CVI-SiC mini-composites with a PyC and BN interface was studied in detail. The interfacial shear strength of the SiC/PyC/SiC mini-composites decreased from 15 MPa to 3 MPa after the heat treatment at 1500 °C for 50 h, while that of the SiC/BN/SiC mini-composites decreased from 248 MPa to 1 MPa, which could be mainly attributed to the improvement of the crystallization degree of the interface and the decomposition of the matrix. Aside from the above reasons, the larger declined fraction of the interfacial shear strength of the SiC/BN/SiC mini-composites might also be related to the gaps in the BN interface induced by the volatilization of B₂O₃·SiO₂ phase, leading to a significant larger declined fraction of the tensile strength of the SiC/BN/SiC mini-composites due to the obvious expansion of the critical flaws on the fiber surface. Therefore, compared with the CVI BN interface, the CVI PyC interface has better heat-resistance at high temperatures up to 1500 °C due to the fewer impurities in PyC.     

Public perceptions on artificial intelligence driven disaster management: Evidence from Sydney,Melbourne and Brisbane

In recent years, artificial intelligence (AI) is being increasingly utilised in disaster management activities. The public is engaged with AI in various ways in these activities. For instance, crowdsourcing applications developed for disaster management to handle the tasks of collecting data through social media platforms, and increasing disaster awareness through serious gaming applications. Nonetheless, there are limited empirical investigations and understanding on public perceptions concerning AI for disaster management. Bridging this knowledge gap is the justification for this paper. The methodological approach adopted involved: Initially, collecting data through an online survey from residents (n = 605) of three major Australian cities; Then, analysis of the data using statistical modelling. The analysis results revealed that: (a) Younger generations have a greater appreciation of opportunities created by AI-driven applications for disaster management; (b) People with tertiary education have a greater understanding of the benefits of AI in managing the pre- and post-disaster phases, and; (c) Public sector administrative and safety workers, who play a vital role in managing disasters, place a greater value on the contributions by AI in disaster management. The study advocates relevant authorities to consider public perceptions in their efforts in integrating AI in disaster management.     

The role of microbial electrogenesis in regulating methane and nitrous oxide emissions from constructed wetland-microbial fuel cell

This study assesses a sustainable solution to greenhouse gases (GHGs) mitigation using constructed wetland-microbial fuel cells (CW-MFC). Roots of wetland plant Acorus Calamus L. are placed in biological anode to better enable anode microorganisms to obtain rhizosphere secretion for power improvement. Three selected cathode materials have a large difference in GHG emissions, and among them, carbon fiber felt (CFF) shows the lowest emissions of methane and nitrous oxide, which are 0.77 ± 0.04 mg/(m²·h) and 130.78 ± 13.08 μg/(m²·h), respectively. The CFF CW-MFC achieves the maximum power density of 2.99 W/m³. As the influent pH value is adjusted from acidic to alkaline, the GHGs emissions are reduced. The addition of Ni inhibits GHGs emission but decreases the electricity, the power density is reduced to 1.09 W/m³, and the methane and nitrous oxide emission fluxes decline to 0.20 ± 0.04 mg/(m²·h) and 15.49 ± 1.86 μg/(m²·h), respectively. Low C/N ratio reduces methane emission, while high C/N ratio effectively inhibits nitrous oxide emission. At the influent pH 8 and C/N = 5:1, the methane emission flux is approximately 10.60 ± 0.27 mg/(m²·h), and the nitrous oxide emission flux is only 10.90 ± 1.10 μg/(m²·h). Based on the above experimental results by controlling variable factors, it is proposed that CW-MFC offers an environment-friendly solution to regulate GHG emissions.     

A novel method of shrimp blanching by CO2 heat pump: Quality,energy, and economy analysis

To reduce the energy consumption of the shrimp blanching process and improve the economic value of the blanched product, a transcritical CO₂ heat pump blanching system (THPB system) was designed in this paper. The trends of astaxanthin were investigated at atmospheric pressure near boiling temperature, combined with the color and structural properties of shrimp samples, and the optimal blanching times of 270 s and 240 s were obtained at 90°C and 95°C, respectively. In contrast to the fuel blanching system (FB system) at 100°C, the annual standard coal consumption of the THPB system with 90°C blanching is decreased by 79%, and the annual operating cost can be saved by CNY 63,800, with a payback period of about 3.13 years.Industrial relevanceBlanching is one of the effective ways to prolong the shelf life of shrimp. However, the research on the blanching time and temperature of shrimp is not comprehensive. In addition, the traditional fuel blanching process has high energy consumption and pollution, and can no longer meet the quality requirements of the modern food processing industry. Heat pump has been shown to have better performance in food drying, but it is less used in blanching. The information presented in this study may provide other insights into food processing.     

Effect of temperature on raw whole milk density and its potential impact on milk payment in the dairy industry

The objective of this study was to determine the effect of temperature on whole milk density measured at four different temperatures: 5, 10, 15, and 20 °C. A total of ninety-three individual milk samples were collected from morning milking of thirty-two Holstein Friesian dairy cows, of national average genetic merit, once every two weeks over a period of 4 weeks and were assessed by Fourier transform infrared spectroscopy for milk composition analysis. Density of the milk was evaluated using two different analytical methods: a portable density meter DMA35 and a standard desktop model DMA4500M (Anton Paar GmbH, UK). Milk density was analysed with a linear mixed model with the fixed effects of sampling period, temperature and analysis method; triple interaction of sampling period x analysis method x temperature; and the random effect of cow to account for repeated measures. The effect of temperature on milk density (ρ) was also evaluated including temperature (t) as covariate with linear and quadratic effects within each analytic method. The regression equation describing the curvature and density–temperature relationship for the DMA35 instrument was ρ = 1.0338−0.00017T−0.0000122T² (R² = 0.64), while it was ρ = 1.0334 + 0.000057T−0.00001T² (R² = 0.61) for DMA4500 instrument. The mean density determined with DMA4500 at 5 °C was 1.0334 g cm⁻³, with corresponding figures of 1.0330, 1.0320 and 1.0305 g cm⁻³ at 10, 15 and 20 °C, respectively. The milk density values obtained in this study at specific temperatures will help to address any bias in weight–volume calculations and thus may also improve the financial and operational control for the dairy processors in Ireland and internationally.