Journal of Shandong University(Engineering Science)

Research on the detection of train wheelset tread defects under complex conditions

Hou Tao, Ren Yikun, Niu Hongxia

Journal of Shandong University(Engineering Science). 2026, 56(5): 1-12. doi:10.6040/j.issn.1672-3961.0.2025.090

Abstract ( 29 )

PDF (16769KB) ( 8 )

Save

References | Related Articles | Metrics

To address the problems of poor detection performance for train wheelset tread defects under complex conditions and the mismatch among computational load, parameter count and accuracy, a detection method based on improved YOLOv9s was proposed. The sparse network structure Ghost-Inception was used to replace RepNCSPELAN4 in the backbone network of YOLOv9s, so as to reduce the model parameters and computational load, and improve detection accuracy and efficiency. The iEMA and SEAM attention mechanisms were introduced into the Neck network, and a new PagAM feature concatenation module was constructed to enhance the feature extraction capability under complex conditions. A RepNCSPELAN4-DCNv3 feature extraction and fusion module was built to strengthen the detection of defects at different scales. Experimental results showed that the parameters of the improved YOLOv9s network were reduced by 8.9%, the computational load was decreased by 11.4%, and the mean average precision was increased by 5.0%. The proposed method achieved improved detection accuracy on the premise of algorithm lightweight.

Roughset models over three universes

Weng Shizhou, Lü Yuejin, Huang Zhengwei

Journal of Shandong University(Engineering Science). 2026, 56(5): 13-25. doi:10.6040/j.issn.1672-3961.0.2025.126

Abstract ( 18 )

PDF (1172KB) ( 5 )

Save

References | Related Articles | Metrics

The classical rough set theory faces limitations when dealing with increasingly complex uncertainty decision-making problems. To better handle uncertain information, serve knowledge acquisition, and decision analysis, the rough set theory based on single and dual universes was extended to a triple universe, proposing a triple-universe rough set model. Concepts such as predecessor, successor neighborhood of elements were defined, as well as common neighborhood, and possible neighborhood of sets. Based on this, concepts such as lower approximation, upper approximation, approximation accuracy, and roughness were defined. The relevant properties and theorems satisfied by these concepts were summarized and rigorously derived, supplemented by numerical examples for verification. This provides new insights for the improvement of rough set theory.

An energy-efficient cloud workflow scheduling method combining list schedulind and artificial bee colony algorithm

Wang Yun, Zhu Xiaoze, Jiang Xinhuan, Tian Xiumei, Wu Zhiqiang

Journal of Shandong University(Engineering Science). 2026, 56(5): 26-39. doi:10.6040/j.issn.1672-3961.0.2025.116

Abstract ( 23 )

PDF (7590KB) ( 2 )

Save

References | Related Articles | Metrics

To address the high energy consumption problem caused by the increasing growth of data centers in the field of cloud workflow scheduling, a workflow scheduling method combining list scheduling and artificial bee colonies, called LisABC, was proposed, which aimed to meet the deadline constraints and reduce the execution energy consumption. Considering the impact of task scheduling sequences on the energy consumption of workflow execution, a new honey source(scheduling solution)encoding was designed; by leveraging the characteristics of workflow structure, a hierarchical method for generating effective task scheduling sequences based on workflow layering was proposed; a scheduling rule that took into account both the sub-term deadline constraints of tasks and the minimization of task execution energy consumption was proposed. Combined with list scheduling ideas, the honey source solution was decoded into a scheduling plan. Experimental results showed that compared with existing methods, LisABC had been able to effectively reduce the energy consumption of the workflow while meeting deadline constraints, and had been well adapted to scientific workflows of different scales.

The study of commuter flow identification methods on expressways in megacities based on ETC toll data

Dai Hongna, Li Zhongrui, Feng Haixia, Zhao Huanhuan, Fan Hongxian

Journal of Shandong University(Engineering Science). 2026, 56(5): 40-51. doi:10.6040/j.issn.1672-3961.0.2024.210

Abstract ( 20 )

PDF (10784KB) ( 2 )

Save

References | Related Articles | Metrics

To scientifically analyze the characteristics of highway commuting flows in megacities, a dual-dimensional approach focusing on “passenger flow characteristics-land use structure”was adopted in this study. Highway ETC toll data and land use data within a 5 km radius of toll stations were utilized for analysis. The station types were classified using location entropy and the fuzzy C-means algorithm. Based on vehicles' weekday travel patterns, an improved Canopy-K-means algorithm was applied to identify commuting flows. The highway commuting flows in Jinan were used as a case study for validation and analysis of their spatiotemporal distribution characteristics. The results showed that toll stations could be categorized into six types, including employment-dominated and employment-biased mismatch types. Commuting flows were divided into two categories: conventional commuters who used highways during both morning and evening peak hours, and special commuters who only used highways during the morning peak hour. The commuting flows were primarily composed of traffic from counties to the central urban area. This research provided data and theoretical support for highway planning and differentiated toll policies.

Label distribution learning based on label-specific features and common features

Kong Shu, Chen Linlin, He Qiang, Wang Hengyou

Journal of Shandong University(Engineering Science). 2026, 56(5): 52-61. doi:10.6040/j.issn.1672-3961.0.2025.132

Abstract ( 18 )

PDF (2905KB) ( 4 )

Save

References | Related Articles | Metrics

Label distribution learning(LDL)is an innovative machine learning paradigm that focuses on the label descriptiveness of specific instances. Existing LDL models were built on a simple feature space where all features were shared by all class labels. However, this traditional data representation strategy tended to select features that were distinguishable across all labels while overlooking label-specific features that were relevant and distinctive for each class label. In this paper, we proposed a label distribution learning method based on label-specific features and common features. First, we calculated the label correlation matrix for label pairs, and for each label pair, all instances were partitioned based on descriptiveness, clustering was performed for each class, and label pair-specific features were constructed. Next, clustering analysis was performed on all instances to construct common features for the labels. Finally, training and testing were conducted by querying the clustering results. Extensive experiments on multiple real label distribution datasets validated the superiority of the method and the effectiveness of label-specific features.

Semantic description of ancient architecture images guided by visual cues

Chen Xin, Yang Fan, Cui Di

Journal of Shandong University(Engineering Science). 2026, 56(5): 62-70. doi:10.6040/j.issn.1672-3961.0.2025.224

Abstract ( 22 )

PDF (4634KB) ( 2 )

Save

References | Related Articles | Metrics

In the field of digital services, users faced difficulties in accurately understanding the deep semantic information contained in cultural images such as ancient architectures. To effectively represent the semantic information of ancient architectural images, a method was proposed to generate descriptions of ancient architecture images using visual cues guided by a multi-modal large language model. First, the S²-wrapper mechanism was utilized to extract multi-scale visual features from images. A multi-label recognition network based on Transformer was constructed by introducing a label-mask training strategy with predefined label states, by which deep relationships between images and labels, as well as among labels themselves were established. Subsequently, the generated multiple labels were transformed into structured textual prompts and defined as visual cues. Finally, the visual cues and images were jointly input into the multi-modal large language model to generate a series of interpretations related to ancient architectural knowledge, without requiring any additional cross-modal training. Quantitative analysis was conducted to compare the recognition performance of different methods, and qualitative analysis was performed to evaluate the quality of the generated descriptions. The results showed that the proposed method achieved an average precision of 82.7%, a class-average F1 score of 72.7%, and an overall F₁ score of 85.7% on our self-built ancient architecture image dataset. Furthermore, ablation experiments further verified the effectiveness of each proposed module.

Multimodal sentiment analysis of images and texts based on auxiliary knowledge of large language models and contrastive learning strategy

Tang Kaiyu, Zhang Hengyuan, Aizierguli·Yusufu

Journal of Shandong University(Engineering Science). 2026, 56(5): 71-78. doi:10.6040/j.issn.1672-3961.0.2025.098

Abstract ( 20 )

PDF (5179KB) ( 8 )

Save

References | Related Articles | Metrics

With the rapid development of large models in the field of artificial intelligence, multimodal large language models(MLLM)have shown strong understanding capabilities for text and visual inputs, and they also have quite rich common sense and reasoning characteristics.Inspired by this, this paper designed a method for multimodal sentiment analysis by using a multimodal large language model LLaVA(large language and vision assistant)to generate auxiliary knowledge. At the same time, the paper designed a multi-task framework based on auxiliary knowledge and contrastive learning strategy(BAK-CLS)was designed to implement multimodal sentiment analysis tasks on pre-trained language models through the text + text paradigm. Masked language modeling tasks and contrastive learning tasks were included in the multi-task framework. The model performance was improved by learning the sentiment knowledge of data samples through masked language modeling. Meanwhile, a contrastive learning task was incorporated during model training to improve the models sentiment recognition capability in scenarios where textual and visual semantics were inconsistent, thereby enhancing its robustness. Experimental results showed that compared with the traditional image-text feature fusion network, thanks to the common sense learned by the large model and its powerful reasoning ability, the sentiment classification results obtained by the method proposed in this chapter on the two public datasets MVSA-Single and MVSA-Multiple were significantly improved in multiple indicators.

Motif-based graph pattern-aware sampling

Hu Bin, Wang Xin, Xie Wenbo, Zhang Chong, Yan Sisi, Xiao Junfeng

Journal of Shandong University(Engineering Science). 2026, 56(5): 79-91. doi:10.6040/j.issn.1672-3961.0.2025.252

Abstract ( 22 )

PDF (10685KB) ( 4 )

Save

References | Related Articles | Metrics

To address the problem that existing graph sampling methods were unable to obtain high-quality sampled graphs supporting downstream tasks under the rapid growth of graph data, a Motif-based graph pattern-aware sampling(MGPA)method was proposed. A Motif-informed weighted sampling(MIWS)strategy was designed, in which the walking process was guided by node weights so that high-value regions were preferentially visited, achieving targeted sampling of regions rich in frequent patterns. A neighborhood-matching-based node weight prediction model(EchoModel)was constructed to jointly learn node attributes, local structures, and an external Motif pattern set, and output the weight of the guiding walk for each node to accurately reflect node importance. A neighborhood-matching-guided node feature generation algorithm(DomainMatching)was designed. By quantifying the matching degree between each node's neighborhood and a predefined Motif pattern set, a target vector enriched with neighborhood information was produced for every node by the algorithm, and reliable supervisory information was thereby provided for the training of EchoModel. Biased random walks were performed by MIWS according to the node weights, and regions rich in frequent patterns were preferentially explored. Experimental results showed that superior performance was achieved by MGPA across multiple evaluation metrics, and a mining accuracy of up to 96% was attained for the top-k(k=100)frequent patterns under a sampling rate of only 20%.

Design and reaction mechanisms of blast furnace slag-carbide slag-desulfurization ash ternary solid waste cementitious material based on extreme vertices method

Chen Cong, Wan Tiantao, Shan Hongri, Li Hui, Zhang Yong, Song Shuai, Ge Zhi

Journal of Shandong University(Engineering Science). 2026, 56(5): 92-101. doi:10.6040/j.issn.1672-3961.0.2025.084

Abstract ( 27 )

PDF (5987KB) ( 5 )

Save

References | Related Articles | Metrics

To achieve efficient resource utilization of ground granulated blast-furnace slag(GGBS), carbide slag(CS), and desulfurization ash(DA), the mix proportions of a ternary solid waste-based cementitious material were designed and optimized using the extreme vertices method. The influence of raw material composition on the compressive strength was investigated. The hydration products and pore structure characteristics of the materials were characterized using X-ray diffraction, thermogravimetric analysis, and mercury intrusion porosimetry. It was indicated by the results that the optimal mix proportion of GGBS, CS, and DA was 70.0∶19.4∶10.6. A negative correlation with the compressive strength was exhibited by the GGBS content across all curing ages, while enhanced later-age strength was contributed by DA. The interaction between GGBS and DA showed the lowest contribution, whereas the CS-DA interaction demonstrated the highest contribution. Within the GGBS-CS-DA ternary system, CaSO₃ from DA continuously reacted with GGBS in the alkaline environment provided by CS, forming ettringite and C—S—H gel. These hydration products densified the microstructure of the cementitious material through cementation and pore-filling effects, consequently improving its compressive strength.

Rock strength prediction model based on multi-layer perceptron and drilling parameters of rock drilling jumbo

Peng Liuyi, Liu Hao, Zhang Qi, Liu Zhihui, Sun Zizheng

Journal of Shandong University(Engineering Science). 2026, 56(5): 102-110. doi:10.6040/j.issn.1672-3961.0.2025.027

Abstract ( 31 )

PDF (6705KB) ( 6 )

Save

References | Related Articles | Metrics

To address the problem that existing methods for rapidly obtaining the uniaxial saturated compressive strength(USCS)of rocks based on drilling parameters of rock drilling jumbos rarely considered the influence of impact frequency, leading to insufficient prediction accuracy during the construction of drill-and-blast tunnels, this research conducted experiments on a full-scale rock drilling jumbo, took impact frequency, impact pressure and other parameters as feature variables, and constructed a USCS prediction model for rocks based on a multi-layer perceptron and drilling parameters. The correlation coefficients between drilling parameters and rock strength were used to assign weights to each parameter in the model, which improved the model's accuracy and enabled real-time prediction of the rock USCS. The developed perceptual model relied on easily collected on-site drilling parameters of rock drilling jumbos. Test and verification results showed that the overall prediction accuracy of the model reached 86.67% within an allowable error range of ±3 MPa, indicating high prediction precision.

The influence of freeze-thaw cycles and salt content on microscopic and shear characteristics of saturated silty soil

Lü Zhonghua, Zhang Qi, Chen Guolong, Chen Rong, Chen Youhui, Wang Ziru

Journal of Shandong University(Engineering Science). 2026, 56(5): 111-123. doi:10.6040/j.issn.1672-3961.0.2025.032

Abstract ( 20 )

PDF (11050KB) ( 4 )

Save

References | Related Articles | Metrics

To investigate the effects of freeze-thaw cycles on the micro pore distribution, the strength and modulus of saturated soil, freeze-thaw cycles were applied to saturated silt samples with different salt contents, followed by nuclear magnetic resonance testing and consolidated undrained triaxial tests. The results showed that freeze-thaw cycles damaged the structure of saturated soil, leading to the formation of medium and large pores. However, the compaction effect was improved and the damage was partially repaired by consolidation. After six cycles of "damage-repair", the strength of the saturated soil decreased slightly, while the modulus decreased significantly after one freeze-thaw cycle. Salt content had a significant impact on the pore size distribution of the soil; cations influenced the electrical double layer, resulting in the formation of large pores, but also improved consolidation. Under the combined effects of salt-induced structural damage and consolidation repair, the strength of saline soil decreased with increasing salt content, with a more pronounced decrease under higher stress levels. Freeze-thaw cycles caused salt cementation, which played a role in repairing damage; one freeze-thaw cycle could increase the strength of saline soil. This study could provide a basis for determining stability analysis parameters for ditch slope protection, shore engineering, and low-permeability foundations in seasonally frozen regions.

Research on the mechanism of Ordovician limestone water inrush from the floor induced by local goaf mining in deep multi-seam coal mines

Wang Wei, He Enlei, Liu Zewei, Li Haiyan, Zhang Bo, Yan Xianzhen, Xia Maozhe,Li Bingbing

Journal of Shandong University(Engineering Science). 2026, 56(5): 124-136. doi:10.6040/j.issn.1672-3961.0.2025.058

Abstract ( 22 )

PDF (12873KB) ( 4 )

Save

References | Related Articles | Metrics

This research was focused on Ordovician limestone water inrush accidents induced by local goaf mining formed in adjacent small coal mines within the Fengfeng Mining Area. It was aimed at revealing the characteristics of Ordovician limestone water inrush from the floor under local goaf mining conditions in deep multi-seam coal mines, and a corresponding targeted exploration and treatment technology was proposed. By combining theoretical calculation, numerical simulation, and engineering practice, a mechanical model for floor stress along the working face strike was constructed, and the floor failure characteristics and water inrush behavior were analyzed under different combinations of local goaf mining, including local goaf mining of the 4# coal seam only, combined local goaf mining of the 4# and 9# coal seams, and combined local goaf mining of the 4#, 6#, and 9# coal seams. The results showed that the direct floor of the goaf was mainly characterized by tensile failure, and the deep rock stratum and the advance abutment stress zone were mainly dominated by shear failure. The theoretical calculation results showed that the failure depths of 4 # coal, 6 # coal and 9 # coal floor were 31.8 m, 32.2 m and 32.7 m respectively, which were highly consistent with the standard formula and numerical simulation results. Further results indicated that when local goaf mining was limited to the 4# coal seam, or to the combined mining of the 4# and 9# coal seams, the water inrush coefficients obtained from theoretical calculation, empirical formulae specified in regulations, and numerical simulation ranged from 0.012 5 to 0.047 0 MPa/m, which were all lower than the critical threshold of 0.06 MPa/m, indicating no water inrush risk was identified. In contrast, when combined local goaf mining of the 4#, 6#, and 9# coal seams was conducted, the effective aquiclude thickness was sharply reduced to less than 4 m, and the water inrush coefficient was increased to 5.88-7.12 times the critical threshold(0.352 5-0.427 3 MPa/m), thereby forming a continuous failure pathway and posing an extremely high risk of Ordovician limestone water inrush. Based on the research results, a three-level exploration and treatment technology system of "regional positioning-targeted locking-precise plugging" was proposed. Through geophysical prospecting, abnormal areas were delineated, and through three-dimensional exploration using a vertical horizontal hole network and perfusion plugging technology, the water inrush and water inflow were successfully reduced to a safe range. The research outcome can be used as theoretical value and technical reference for the prevention and control of Ordovician limestone water disaster in deep mines.

Method for locating fire sources and predicting maximum temperature rise in cable trenches

Yang Hongmin, Wang Zheng, Zou Liang, Zhang Zhaochen, Su Tongfei, Wang Xuefei, Suo Wenxuan

Journal of Shandong University(Engineering Science). 2026, 56(5): 137-146. doi:10.6040/j.issn.1672-3961.0.2025.199

Abstract ( 22 )

PDF (6837KB) ( 7 )

Save

References | Related Articles | Metrics

To address the challenge of predicting and locating the position of the fire source with difficulty when the fire occurred in a cable trench, a method of fire source location using a single-exponential function with correction parameters and maximum temperature rise prediction was proposed. Based on a full-scale test platform and finite element simulation analysis, the relationship between the ceiling temperature rise and the location of the fire source was obtained, effectively enabling the location of the fire source and the prediction of the maximum temperature rise. The experimental results showed that the temperature of the ceiling exhibited an exponential change pattern longitudinally. The maximum deviation of fire source location was 0.098 m, with an average deviation of 0.044 m, and the average accuracy reached 98.67%. The maximum error in predicting the maximum temperature rise of the ceiling was 30 ℃, with an average deviation of 23.33 ℃, and the average accuracy was 96.8%. Compared with traditional fire prediction methods, the proposed method offered higher precision, providing a theoretical basis for the prevention of early cable trench fires. For different sizes of cable trenches, the proposed method's horizontal applicability was generally good, although its vertical applicability was limited to some extent. This limitation could be addressed by setting up temperature measurement points every 10 m to ensure accurate predictions.

Feature-directed surrogate-accelerated parameter identification method for the J-A hysteresis model

Xu Jianwen, Zang Ying, Gao Teng, Zhang Zhiqiang, Meng Lingzhuochao, Liu Zhizhen, Sun Weize

Journal of Shandong University(Engineering Science). 2026, 56(5): 147-153. doi:10.6040/j.issn.1672-3961.0.2025.167

Abstract ( 24 )

PDF (1646KB) ( 4 )

Save

References | Related Articles | Metrics

To address the prohibitive per-iteration evaluation cost of evolutionary algorithms in Jiles-Atherton(J-A)hysteresis model parameter identification, a feature-directed surrogate-accelerated parameter identification method was proposed. A six-dimensional physical feature vector replaced hundreds of B-H loop sampling points as the identification target, eliminating inter-sample redundancy. A radial basis function(RBF)surrogate model with dimensionless-ratio input engineering was introduced to compress single-evaluation time to microsecond-scale inference. A feature-directed differential evolution(FD-DE)algorithm was proposed to exploit the RBF analytical gradient via weighted multi-feature direction with reverse annealing, that was, exploration dominated early search while gradient guidance was progressively introduced. Experiments on a J-A simulated dataset constructed from typical grain-oriented electrical steel parameter ranges showed R²>0.99 for all six features of RBF surrogate model, input engineering improved R² from 0.983 0 to 0.997 3, FD-DE accelerated convergence by 42% over standard differential evolution(DE), with all feature-level errors meeting hysteresis loop reconstruction requirements. The proposed method accommodated incomplete data and the surrogate model served multiple identification tasks after one training.

Review on nonlinear dynamics of metal cutting system

Ren Yongsheng, Yao Donghui, Zhang Jinfeng

Journal of Shandong University(Engineering Science). 2026, 56(5): 154-172. doi:10.6040/j.issn.1672-3961.0.2025.325

Abstract ( 18 )

PDF (1929KB) ( 1 )

Save

References | Related Articles | Metrics

Cutting chatter is an undesirable dynamic state frequently occurred in high-speed machining and high-precision machining system. As a complex physical phenomenon, cutting chatter has the characteristics of high dimension, non-smooth, strong nonlinear and variable parameters as well as time-delay effect, which makes it becomes the key difficulties and challenges to study for dynamics of mechanical system. The nonlinearity that has always been recognized as an essential element in machining is the direct reason for nonlinear chatter occurrence. Understanding nonlinear dynamics helps to reveal the essence of unstable phenomena in cutting processes and provide ideas for eliminating chatter problems. A critical review of the nonlinearities in machining system including cutting force and machine tool structure, friction between tool and chip/workpiece, loss of contact between tool and workpiece, etc., was presented, and the available approaches for solving nonlinear models were summarized. Nonlinear bifurcation, periodic and quasiperiodic response and chaos phenomena in cutting process wre briefly introduced. Some open problems and a few possible research directions in future were suggested.

Analysis of two-phase flow and structural response characteristics in gas turbine lubricating oil pipes

Zhang Bin, Zhou Guangjie, Ren Guanyu, Xin Gongming, Liu Bo

Journal of Shandong University(Engineering Science). 2026, 56(5): 173-181. doi:10.6040/j.issn.1672-3961.0.2025.153

Abstract ( 23 )

PDF (5464KB) ( 3 )

Save

References | Related Articles | Metrics

To prevent gas locking and the consequent blockage of the lubricating-oil return pipeline in the first 300 MW-class F-class heavy-duty gas turbine, as well as pipeline deformation induced by unstable internal flow, the two-phase flow and structural response characteristics of the oil return pipeline were systematically investigated. A numerical model of gas-liquid two-phase flow covering the entire oil return pipeline was developed using the shear stress transport(SST)k-ω turbulence model and the volume of fluid(VOF)method. The flow patterns, two-phase flow regimes, and gas-liquid distributions in the pipeline were analyzed under different lubricating-oil temperatures and liquid volume fractions at the oil return inlet. Based on the simulated flow-field data, a fluid-structure interaction dynamic model of the oil return pipeline was subsequently established to investigate the structural response characteristics of the pipeline and its supports under different flow conditions. The numerical results indicated that the lubricating oil in the return pipeline primarily exhibited two characteristic flow patterns, including stratified flow and circumferential swirling flow. In the pipeline section characterized by circumferential swirling flow, the maximum liquid volume fraction occurred at the downstream end cross-section. The liquid volume fraction in the central cross-section of this section was positively correlated with the liquid volume fraction at the oil return inlet. As the lubricating oil temperature increased, the overall mean liquid volume fraction in the pipeline generally decreased. When the inlet liquid volume fraction was below 0.50, the system-averaged liquid volume fraction increased with increasing inlet liquid volume fraction. Structural mechanical analysis further demonstrated that the stress levels in both the pipeline system and its supports satisfied the requirements of the relevant design codes. The findings provided a theoretical basis for the optimal design and safety assessment of lubricating oil return systems in heavy-duty gas turbines and contributed to improving the operational reliability and stability of gas turbine units.

Single-phase numerical simulation research of plate-fin heat exchangers for new energy vehicles

Qu Huige, Liu Yan, An Chunguo, Li Kefeng, Zhang Jingzhi

Journal of Shandong University(Engineering Science). 2026, 56(5): 182-191. doi:10.6040/j.issn.1672-3961.0.2025.061

Abstract ( 19 )

PDF (8448KB) ( 2 )

Save

References | Related Articles | Metrics

Numerical simulation analysis of single-phase flow and heat transfer in fin channels of plate-fin heat exchangers was conducted to investigate the velocity distribution, average surface heat transfer coefficient, and pressure drop characteristics of plain and serrated fins, as well as serrated fins with different structural parameters. CFD software with the shear stress transport(SST)k-ω turbulence model was employed to numerically simulate the single-phase heat transfer flow of a 50% ethylene glycol aqueous solution in the fin channels, and experimental validation was performed. The results showed that the fin type and structural parameters significantly influenced the flow and heat transfer, with serrated fins exhibiting better flow and heat transfer characteristics than plain fins. By altering the structural parameters of the serrated fins, it was found that an increase in flow velocity led to a rise in both heat transfer and pressure drop. Reducing the fin height, spacing, or discontinuity length, or increasing the fin thickness, was beneficial for heat transfer, but also resulted in an increase in pressure drop.

Thermo-economic analysis and optimization of a solar-driven polygeneration system

Ma Bingyu, Guo Lei, Rong Fanhua, Sheng Xuerou, Hou Yuehua, Yuan Xueliang

Journal of Shandong University(Engineering Science). 2026, 56(5): 192-200. doi:10.6040/j.issn.1672-3961.0.2026.020

Abstract ( 17 )

PDF (7666KB) ( 5 )

Save

References | Related Articles | Metrics

A solar-driven multi-generation system was proposed to achieve efficient and sustainable integrated energy supply for large-scale public infrastructures, such as airports. The system employed a parabolic trough solar collector(PTSC)to harvest solar energy for driving a transcritical carbon dioxide cycle(TRCC), and was further integrated with a liquefied natural gas cold energy utilization module to enhance overall system performance. Through multi-energy coupling and energy cascade utilization, the system enabled the coordinated production of electricity, cooling, and heating. A comprehensive evaluation of the system was conducted from the perspectives of energy, exergy, and economic performance, and a multi-objective optimization framework combining backpropagation neural networks, the Non-dominated Sorting Genetic Algorithm II, and the Technique for Order Preference by Similarity to an Ideal Solution was applied. Results indicated that under design conditions, the power generation efficiency, energy efficiency, and exergy efficiency reached 9.015%, 35.56%, and 11.22%, respectively, with a payback period of 2.923 a. The PTSC subsystem was identified as the primary contributor to exergy destruction and cost. Parametric analysis showed that TRCC cycle parameters, heat transfer oil mass flow rate, and solar irradiance significantly affected system performance and economic feasibility. After multi-objective optimization, the energy efficiency and exergy efficiency increased to 38.10%, and 13.30%, respectively, while the payback period was reduced to 2.645 a. The proposed system demonstrated strong potential for providing clean and efficient integrated energy supply to large public buildings such as airports.

Mechanistic study on the effect of carbonate on the degradation of trimethoprim in UV/H₂O₂

Wei Fenghua, Wang Chenglin, Gu Qingyuan, He Maoxia

Journal of Shandong University(Engineering Science). 2026, 56(5): 201-214. doi:10.6040/j.issn.1672-3961.0.2025.162

Abstract ( 26 )

PDF (11343KB) ( 1 )

Save

References | Related Articles | Metrics

To elucidate the regulatory mechanisms and pH dependence of carbonates in the UV/H₂O₂ degradation of trimethoprim(TMP), theoretical calculation methods were adopted to study the differences in the initial reaction mechanisms of TMP with CO^·-₃ and HO^· under UV/H₂O₂ conditions, as well as the influence mechanism of carbonates on their reaction mechanisms. The research results showed that CO^·-₃ had strong selectivity for molecules in different dissociation forms(neutral, diprotonated and monoprotonated forms of TMP), and only underwent single electron transfer reaction(SET)with neutral TMP. Compared with the HO^· system, its SET activation energy barrier was reduced by 30.98%. Therefore, compared with the multi-site hydroxylation characteristics of HO^· on the benzene ring, imino and pyrimidine ring, although CO^·-₃ could not directly achieve the complete mineralization of TMP, it could significantly reduce the activation energy of the subsequent reaction of HO^· by generating a radical cation intermediate. There was a threshold effect in carbonate concentration. At low concentrations, carbonate acted as an electron transfer medium between HO^· and TMP, which could accelerate the degradation of TMP. High concentrations, on the other hand, exhibited a quenching effect on HO^· and inhibited degradation. Toxicity assessment of the intermediate products indicates that, with the exception of P12, the toxicity of most hydroxylated and carbonylated intermediates is lower than that of the parent compound. Carbonate showed a concentration-dependent effect during UV/H₂O₂ degradation of TMP, low concentrations promoted TMP degradation via CO^·-₃-mediated electron transfer, whereas high concentrations inhibited it by quenching HO^·

Table of Content