Accurate prediction of hydrocarbon vapor temperature is crucial to the operation of the Alberta Taciuk Processor (ATP) hydrocarbon vapor handling system. However, the problem of hydrocarbon vapor temperature prediction in ATP systems remains insufficiently studied. Existing models rarely distinguish variable-specific dynamic changes from system-level interactions. To address this limitation, this study proposes a novel deep learning model, termed DOCoR, which achieves effective disentanglement and fusion of individual characteristics and system-level coordination in multivariate time-series data through the synergistic integration of a private encoder, a shared encoder, and an orthogonality constraint mechanism. Compared with the traditional LSTM model, the proposed method improves RMSE, MAE, (R2), and MAPE by 41.60%, 46.93%, 8.72%, and 46.15%, respectively; compared with the GRU model, the corresponding improvements are 29.42%, 44.23%, 4.12%, and 43.24%. These results demonstrate the superior predictive accuracy of the proposed model. This study provides a reliable solution to the challenging problem of hydrocarbon vapor temperature prediction in ATP systems.
The accurate determination of the internal liquid moisture freezing temperature (ILMFT) is a prerequisite for investigating the coupled heat and moisture transfer in walls under freezing conditions. The current model overlooks key differences in material pore structure and hygroscopic properties, adversely affecting the accuracy of internal moisture freezing temperature predictions. To address this gap, an equivalent pore size calculation model was developed based on the pore size distributions (PSDs) and water vapor sorption isotherms (WVSIs) of building wall materials. Using principles of engineering thermodynamics, a calculation model was established to predict the ILMFT of wall materials via the equivalent pore size. To validate the model, three composition materials used in a thermal insulation system of exterior wall free from the demolition template were selected. Their PSDs were tested using nuclear magnetic resonance (NMR) and three-dimensional X-ray microscopy (3D-XRM). The freezing temperatures of internal liquid moisture in these materials were experimentally determined at equilibrium relative humidity (ERH) of 0.965 and 0.829 using a thermometric method. The ILMFT of those three materials were also calculated via the proposed model and compared with the experimental results. The predicted results showed good agreement with the experimental values, and the mean relative errors (MREs) is 3.39 and 3.35% under the two respective conditions. These results demonstrate the high computational accuracy of the model. The establishment of this model provides solid theoretical and technical support for research on the coupled heat and moisture transfer characteristics in building envelopes during freezing, as well as for the optimization of thermal insulation and moisture protection design.
As a periodic excitation device capable of generating oscillating jets without moving parts, fluidic oscillators (FOs) have been widely used in flow control, heat transfer enhancement, and other fields, and have shown great application potential in engineering fields such as food drying and industrial cleaning. However, due to the limitations of practical piping connections and installation space, a circular-to-rectangular inlet transition structure is often required at the inlet of the FO, which increases the risk of internal flow instability. In this study, we employed the unsteady Reynolds-averaged Navier-Stokes (URANS) method to investigate the effects of restricted-inlet conditions on the internal flow characteristics of the FO and its oscillation stability under thicknesses of H = 5-25 mm and inlet-outlet pressure differences of Delta P = 10-700 kPa. The influence of the inlet-restricted structure was quantitatively evaluated by comparison with an unrestricted-inlet fluidic oscillator (UIFO). The results showed that the separation vortices and low-pressure regions in the mixing chamber of the inlet-restricted fluidic oscillator (IRFO) were key to maintaining the oscillating jet. As the thickness H increased, the oscillation performance of the IRFO decreased significantly, and the influence of the inlet-restricted structure on the oscillation performance became more pronounced. Under Delta P = 10 kPa, when H = 15 mm, the oscillation frequency and outlet sweeping angle of the IRFO decreased to 30.5 Hz and 72 degrees, respectively, which were approximately 63.3% and 25.0% lower than those of the UIFO. When the thickness increased to H >= 18 mm, oscillation failure occurred in the IRFO, and no regular dominant frequency could be identified. Increasing Delta P could restore regular oscillation within a certain range, but it became difficult to recover when H >= 20 mm. For the geometry investigated in this study, 1 <= H/D <= 3.4 can be used as a safe reference range to avoid oscillation failure. These findings deepen the understanding of the flow characteristics and oscillation stability of FOs under restricted-inlet conditions and provide a basis for structural design and reliable application under practical engineering constraints.
This paper studied the synergistic effects during co-pyrolysis between bituminous coal and pine wood by evaluating and optimizing the pyrolysis kinetic parameters. Thermogravimetric analysis was conducted on individual samples and their mixtures (1:1, 2:1, 1:2 mass ratios). The K-K method, combined with peak-differentiating analysis, was utilized to characterize the sub-reactions of the five samples. Subsequently, kinetic parameters associated with each sub-reaction were calculated via the Coats-Redfern method and optimized with the Shuffled Complex Evolution algorithm. The results indicated that the initial pyrolysis reaction temperature (594-597 K) of coal-biomass mixtures was lower than that of individual samples (600 and 667 K). Throughout the pyrolysis process, coal and biomass exhibited synergistic effects. Coal underwent three sub-reactions, while biomass and its mixtures underwent four sub-reactions. The activation energy (E) of coal ranged from 45.75 to 234.55 kJ/mol, while pine wood and its blends exhibited wider ranges of 18.83-216.96 kJ/mol and 18.41-206.14 kJ/mol, respectively. Among the blends, the 2:1 blend showed the lowest E values for the first two sub-reactions (18.41 and 122.78 kJ/mol), whereas the 1:1 blend had a lower E value for the third sub-reaction (162.42 kJ/mol) than that of pine wood (216.96 kJ/mol). This indicates a synergistic effect for the 2:1 blend during the early stage of co-pyrolysis and for the 1:1 blend during the later stage. All samples exhibited high E values for the fourth sub-reaction, presenting an inhibitory effect. These findings facilitate the design and optimization of pyrolysis processes, providing valuable insights for enhancing the efficient utilization of coal.
Fast pyrolysis is a promising thermochemical conversion technology for converting agricultural residues into renewable liquid fuels. This study investigated the effects of reactor temperature, biomass moisture content, and feedstock characteristics on product distribution and bio-oil quality during the fast pyrolysis of corn stalk and rice husk in a laboratory-scale fluidized-bed reactor. Experiments were conducted at reactor temperatures of 500, 600, and 700 degrees C using biomass particles smaller than 0.6 mm and a feed rate of 200 g h-1. The results demonstrated that reactor temperature significantly affected product distribution. For rice husk, increasing the reactor temperature from 500 to 700 degrees C increased bio-oil yield from 50.45 wt.% to 55.53 wt.% while reducing char production. Biomass moisture content also exerted a strong influence on pyrolysis performance. Increasing moisture content from 4.75 wt.% to 15.13 wt.% reduced bio-oil yield from 52.15 wt.% to 40.98 wt.%, corresponding to a decrease of approximately 21.4%. The physicochemical properties of the produced bio-oils were evaluated using lower heating value (LHV), density, and pH measurements. Although corn stalk and rice husk produced comparable bio-oil yields, rice husk generated bio-oil with a higher LHV (18.11 MJ kg-1) than corn stalk (15.47 MJ kg-1), indicating more favorable fuel properties. Both feedstocks produced bio-oil yields exceeding 55 wt.% under high-temperature operating conditions. The results demonstrate that reactor temperature, biomass moisture content, and feedstock characteristics are critical factors governing bio-oil production and quality. Furthermore, bio-oil yields obtained at 700 degrees C were comparable to values commonly reported in the literature despite the higher operating temperature employed in this study. The findings contribute to the understanding of high-temperature fast pyrolysis of agricultural residues and provide useful guidance for the development of sustainable biomass-to-liquid fuel conversion systems.
