Directional freezing of multicomponent fruit juice solutions is governed by coupled heat transfer, solute rejection, and interface instability. However, quantitative comparisons of freezing evolution under static confined conditions remain limited. This study investigated artificial orange, strawberry, and lemon juice solutions prepared at concentrations of 10%, 15%, and 25% in 100 mL and frozen in a horizontal two-dimensional Hele-Shaw cell at surface temperatures of -10 °C and -15 °C. Time-resolved images were analyzed to quantify freezing-front distance, projected frozen area, and macroscopic boundary morphology. The results were obtained from three independent repetitions and statistically evaluated using analysis of variance. The interval freezing rate declined from 0.1464 to 0.0167 mm/min for lemon, from 0.1388 to 0.0178 mm/min for orange, and from 0.1471 to 0.0163 mm/min for strawberry. Lower cooling temperatures enhanced early-stage freezing propagation. Higher juice concentration generally reduced early interface advancement, with increasing thermodynamic inhibition associated with solute loading. The freezing front and projected frozen area also converged to similar values under most conditions, indicating that the freezing pathway rather than the end state was primarily influenced by concentration and temperature. These results show that the kinetics and the shape of the confined solidification are influenced by the juice composition, concentration, cooling intensity and freezing stage.
With the continuous improvement of research and development levels and manufacturing capabilities in the air-conditioner industry, user requirements for air-conditioner noise performance have gradually shifted from simple improvement of A-weighted sound pressure level to higher comprehensive sound quality performance. At present, air-conditioner sound quality evaluation predominantly relies on professional subjective evaluation by experienced engineers. This evaluation mode is highly dependent on personal experience and suffers from low testing efficiency, making it unable to adapt to the rapid iteration requirements of new product development. In addition, traditional linear models fail to accurately characterize the complex nonlinear mapping relationship between objective noise parameters and human subjective auditory perception, limiting their suitability for sound quality evaluation. To solve the above problems, this study proposes a subjective-objective integrated sound quality evaluation model for air-conditioner noise based on support vector regression (SVR) optimized by particle swarm optimization (PSO). It effectively captures the nonlinear correlation between objective psychoacoustic indices and human subjective perception, overcoming the inherent limitations of traditional linear models and single subjective or objective evaluation systems. Compared with the conventional SVR model and the SVR model optimized by genetic algorithm (GA), the proposed model achieves better prediction accuracy and fitting performance.
This study aims to address critical technical challenges in the field of red blood cells (RBCs) cryopreservation, with a focus on overcoming the limitations of clinically conventional glycerol-based cryoprotective agents (CPAs), including high toxicity, complex removal procedures, and functional impairment of RBCs caused by residual glycerol. We have innovatively developed a glycerol-free cryopreservation protocol based on a composite system of N,N-diglycine and α-lactose . Experimental data demonstrated a post-cryopreservation recovery rate of 89.66 ± 0.79% for RBCs treated with the novel formulation. Furthermore, the efficacy of this new CPA was comprehensively validated by assessing post-thaw hemoglobin release, conducting morphological observation of thawed cells via fluorescence microscopy, evaluating post-thaw cell membrane integrity by measuring phosphatidylserine externalization using flow cytometry, and performing detailed morphological analysis through both fluorescence microscopy and cryo-scanning electron microscopy.At the molecular level, molecular dynamics simulations revealed a non-destructive interfacial stabilization mechanism of N,N-diglycine, while α-lactose effectively inhibited ice crystal formation. Their synergistic interaction significantly mitigated cryopreservation-induced damage. Compared with existing technologies, this protocol not only completely avoids the adverse effects associated with glycerol but also offers significant advantages, such as simplified operational procedures and notable cost-effectiveness. These groundbreaking findings provide an important theoretical foundation and technical support for the development of a safer and more efficient long-term RBCs preservation system. The protocol demonstrates promising potential for future clinical translation (Blasi et al., 2012) and considerable industrial value, pending further validation in human-based studies.
To address temperature-control challenges in HVAC systems subject to model uncertainty and unknown external disturbances, a predefined-time adaptive control strategy using a fuzzy-neural-network-based disturbance observer is proposed. A state-space model capturing the coupled dynamics of indoor zone temperature and cooling coil temperature is first established, based on which the temperature tracking problem is recast as a predefined-time state convergence problem. A fuzzy neural network (FNN) is employed to identify the unknown nonlinear system dynamics online, and a predefined-time adaptive disturbance observer (PTADO) is constructed to provide feedforward compensation for the lumped disturbance without requiring prior knowledge of the disturbance bound. Building upon the PTADO, a predefined-time adaptive tracking controller (PTAC) is further developed, which guarantees that the closed-loop system converges within a user-assignable settling time regulated by a single design parameter. Comparative simulations against PI and finite-time sliding mode control demonstrate that the proposed scheme achieves faster convergence, higher steady-state accuracy, and chattering-free control signals under time-varying internal heat gain disturbances.
Thermoelectric cooling has attracted increasing interest for wearable electronics and Internet-of-Things (IoT) devices because these applications require compact and localized temperature control. In such applications, thermal loads often vary with operating states and environmental conditions rather than remain constant. Although transient thermoelectric cooler (TEC) behavior has been widely investigated, systematic characterization of TEC responses to different externally imposed heat-flux waveforms under fixed-current operation remains limited. In this study, a three-dimensional transient numerical model was used to investigate a TEC subjected to fluctuating heat flux imposed on the cold side while the input current was maintained constant. Square, sinusoidal, and triangular heat-flux inputs were examined, together with the effects of period, peak heat flux, and duty cycle on temperature difference and coefficient of performance (COP). The results show that different waveforms produce distinct transient and time-averaged responses even at the same mean heat flux, with square-wave inputs causing the most pronounced fluctuations. Under the investigated conditions, the maximum time-averaged COP was 24.53% higher than the corresponding steady-state value at the same mean heat flux. A dimensionless characterization framework based on the waveform factor, dimensionless time parameter, and heat-flux centroid was then developed to correlate the time-averaged temperature difference and COP. The resulting relations were further assessed against transient numerical simulations for four random heat-flux cases, with maximum relative errors of 6.30% and 8.72%, respectively. These results provide a quantitative basis for characterizing TEC performance under fluctuating cold-side thermal loads within the investigated operating range.
