Table of Content

25 February 2026, Volume 54 Issue 2

Previous Issue    Next Issue

Computer Science & Technology

Target Navigation Method Based on Multimodal Scene Memory and Instruction Prompting

DONG Min, LAI Youcheng, BI Sheng

2026, 54(2):  1-15.  doi:10.12141/j.issn.1000-565X.250152

Asbtract ( 328 )   HTML ( 15)   PDF (2679KB) ( 85 )  

Figures and Tables | References | Related Articles | Metrics

Target navigation requires robots to autonomously plan paths and accurately reach specified target locations based on natural language instructions or object categories in a working environment. Existing approaches to this task primarily fall into two categories in a working environmrnt: end-to-end learning and planning-based methods. While end-to-end methods can directly learn a mapping from perception to action, they often exhibit limited generalization capability and poor interpretability. Conversely, planning-based methods offer better generalization and interpretability to some extent; however, they are often not optimized for known environments, fail to exploit prompt information embedded in natural language instructions, struggle to achieve precise docking at a specified distance from the target, and generally suffer from low execution efficiency. To overcome these limitations, this paper proposed a novel target navigation method named MEMO-Nav, which leverages multimodal scene memory and instruction prompting to improve navigation performance in known environments. The proposed framework adopts a hierarchical architecture: a high-level planning layer maintains a multimodal scene memory to record environmental information and utilizes a Large Language Model (LLM) to parse target and prompt information from natural language instructions. This information is then combined to enable efficient waypoint selection and navigation planning. A low-level execution layer handles fundamental navigation functions, including robot localization and movement, and integrates an object detection model with a depth camera to achieve accurate target positioning. Together, these two layers form a complete target navigation system, ultimately enabling the robot to locate the target and dock at a specified distance based on natural language instructions. Extensive experiments conducted on the GAZEBO simulation platform and in real-world settings demonstrate that the proposed method significantly outperforms existing approaches in known environments across key metrics, including navigation efficiency, success rate, and docking distance accuracy. In summary, the proposed method offers a feasible, efficient, interpretable, and precise solution for mobile robot target navigation in practical scenarios.

Incomplete Multi-View Clustering Algorithm Based on Sample Complementary Anchor Graph

LIU Xiaolan, XU Yuhong

2026, 54(2):  16-24.  doi:10.12141/j.issn.1000-565X.250145

Asbtract ( 149 )   HTML ( 1)   PDF (1482KB) ( 60 )  

Figures and Tables | References | Related Articles | Metrics

With the widespread application of multi-view data in real-world scenarios, clustering with incomplete views has emerged as a significant challenge in machine learning. Traditional anchor graph-based clustering algorithms rely on complete instances to build the anchor graphs. This dependency leads to insufficient anchors for capturing the underlying data structure under high missing rates, while failing to fully leverage the benefits of anchors when missing rate is low. To address the limitations of traditional methods, including restricted anchor selection, inflexible weight assignment, and high computational complexity, this paper proposed an incomplete multi-view clustering algorithm based on a Sample-Complementary Anchor Graphs (IMVC-SAC). First, the algorithm introduces a cross-view anchor complementation mechanism, which adaptively selects anchors from both shared samples and view-specific samples to enhance data structure representation, particularly under high missing rates. Second, it establishes a missing pattern-aware weighting model that dynamically adjusts the contribution of each view to the similarity matrix based on the missing pattern and degree of the samples. Finally, by leveraging the properties of doubly stochastic non-negative matrix factorization, the time complexity of spectral clustering is reduced from cubic to linear with respect to the sample size. Experimental results on five public datasets demonstrate that the proposed IMVC-SAC algorithm outperforms state-of-the-art methods in clustering performance. Notably, it maintains robust and effective clustering even under high missing rates, validating its superiority.

Transmission Tower Bolt Looseness Detection Based on Improved YOLOv5s

WANG Dehong, ZHANG Zixuan

2026, 54(2):  25-37.  doi:10.12141/j.issn.1000-565X.250172

Asbtract ( 178 )   HTML ( 3)   PDF (6249KB) ( 69 )  

Figures and Tables | References | Related Articles | Metrics

As the critical infrastructure in power transmission networks, the structural stability of transmission towers directly impacts the safe and reliable operation of the power grid. During long-term service, bolts in the tower structure are prone to gradual loosening under the coupled effects of multiple factors such as wind loads, temperature variations, and material aging. This paper proposed an intelligent detection model for bolt looseness in transmission towers based on an improved YOLOv5s (named CCSGS-YOLO). The model incorporates several key enhancements: coordinate convolution replaces standard convolutional layers in the backbone network to strengthen the model’s ability to capture positional information of targets; a convolutional block attention module (CBAM) is introduced to strengthen the model’s feature discrimination capability in complex backgrounds through dual channel and spatial attention mechanisms; a slim-neck feature fusion architecture is constructed, leveraging an optimized combination of cross-stage partial connections and depthwise separable convolutions to reduce computational complexity while maintaining detection accuracy; a joint optimization strategy employing the Generalized Intersection over Union (GIoU) loss function and Soft Non-Maximum Suppression (Soft-NMS) improves localization accuracy by considering the geometric overlap characteristics between predicted and ground-truth bounding boxes. Experimental results demonstrate that CCSGS-YOLO achieves a precision of 91.7%, a recall of 89.4%, a mean average precision (mAP) of 95.3%, and an F₁ score of 90.0%. These metrics represent improvements of 1.6, 3.0, 1.4, and 1.0 percentage points, respectively, over the baseline YOLOv5s model. In terms of computational efficiency, the model achieves a detection speed of 74.8 frames per second (FPS), reducing the inference latency to 13.4 ms, which represents an 11.6% improvement compared to the YOLOv5s model. Furthermore, this paper validates the detection robustness of CCSGS-YOLO across various scenarios through field experiments, providing a novel approach for intelligent inspection of loose bolts on transmission towers.

Sound Recognition and Early Warning Mechanism for Liquid Aluminum Leakage Based on Improved EfficientNetV2

LIANG Yanhui, WEN Chengjie, YAN Junwei, ZHOU Xuan, ZHANG Hongtao

2026, 54(2):  38-51.  doi:10.12141/j.issn.1000-565X.250006

Asbtract ( 129 )   HTML ( 0)   PDF (4501KB) ( 66 )  

Figures and Tables | References | Related Articles | Metrics

Liquid aluminum leakage is the direct cause of explosion accidents in aluminum deep-well casting processes. To address the practical engineering challenges of strong lag, low accuracy, and limited monitoring range in existing leakage detection methods, this paper proposed a sound recognition method for liquid aluminum leakage based on an improved EfficientNetV2 model. This method utilizes acoustic characteristics to identify leaks, thereby expanding the monitoring range. The core enhancement involves optimizing the stacking factor and integrating an efficient channel attention mechanism into the EfficientNetV2 architecture to further improve the recognition speed and accuracy. Firstly, a sound database encompassing seven types of acoustic scenes was constructed by collecting audio data under different scenarios using pickups. Then, log-Mel spectrograms were extracted from the sound signal as the feature set and fed into the improved EfficientNetV2 model for training and validation, finally yielding the liquid aluminum leakage sound recognition model. The experimental results show that the recognition accuracy of the improved EfficientNetV2 reaches 95.48%. Compared to the original EfficientNetV2, ResNet, RegNet and DenseNet, the proposed model requires only 12.34%, 8.64%, 11.14%, and 10.80% of the floating point operations, and 11.37%, 9.55%, 15.95%, and 17.24% of the parameters, respectively. Furthermore, it processes 6.53, 6.14, 4.41, and 8.00 times more frames per second in a CPU environment, confirming its fast and accurate recognition performance. In addition, a risk early-warning mechanism for liquid aluminum leakage was established based on the proposed sound recognition method and deployed for real-time risk monitoring in a casting unit. Practical application results verify the effectiveness of both the identification method and the warning mechanism, providing a valuable technical reference for preventing explosion accidents in aluminum deep-well casting.

Recommendation Algorithm Based on Social Diffusion and Adaptive Negative Sampling

CAI Xiaodong, LI Ting, SU Yifeng

2026, 54(2):  52-61.  doi:10.12141/j.issn.1000-565X.250179

Asbtract ( 85 )   HTML ( 1)   PDF (2959KB) ( 58 )  

Figures and Tables | References | Related Articles | Metrics

Social recommendation algorithms based on Graph Neural Network (GNN) leverage social networks to improve recommendation performance. However, most existing methods directly integrate the raw social graph into the recommendation system, which often introduces noise as they overlook the presence of non-homophilous social connections. Furthermore, prevailing negative sampling strategies typically select negative samples with a fixed level of hardness, which is prone to generating false negatives and consequently limits the model’s ability to effectively discriminate between user preferences. To address these issues, this paper proposed a novel recommendation algorithm based on social diffusion and adaptive negative sampling. First, forward diffusion and interest-guided denoising were performed on the social network to derive user representations that reflect homophilic social relations. Subsequently, a multi-view representation alignment approach was employed to maximize the mutual information among user representations from the denoised social graph, the original social graph, and the user-item interaction graph, thereby enhancing the quality of user embeddings. Finally, negative samples of adaptive hardness were selected based on the predicted scores of positive samples, enabling dynamic calibration of the similarity boundary between positive and negative pairs to improve overall model performance. Extensive experimental results demonstrate that the proposed algorithm significantly outperforms state-of-the-art recommendation baselines. On the Douban dataset, it improves recall and NDCG by 11.99% and 10.54%, respectively; on Epinions, by 15.62% and 11.14%; and on Yelp, by 13.80% and 14.90%. These results validate its effectiveness in alleviating social noise and enhancing the differentiation between positive and negative samples.

A Long-Range Lane Detection Method with Enhanced Spatial Perception

WANG Yaoqi, LU Yaqi, WANG Xiaopeng

2026, 54(2):  62-76.  doi:10.12141/j.issn.1000-565X.250181

Asbtract ( 77 )   HTML ( 0)   PDF (6632KB) ( 59 )  

Figures and Tables | References | Related Articles | Metrics

Lane detection serves as a core technology for the visual navigation systems of intelligent vehicles. Its performance directly impacts a vehicle’s path guidance and steering control, which is of great significance for improving traffic safety and navigation efficiency. In lane images, background information often dominates, especially when distant lane markings exhibit challenges such as small feature size, absence, or occlusion, coupled with perspective-induced width variations. These issues make distant lane detection considerably more challenging than detecting nearby lanes. To address this problem, this paper proposed a lane detection method enhanced with spatial perception. Firstly, conside-ring the elongated morphology of lane markings in images, a strip-pooling module was incorporated into the backbone network to refine the representation of lane information. Furthermore, an Enhanced Spatial-Aware Optimizer (ESAO) was integrated with a Lane Multi-Scale Aggregator (LMSA) to suppress irrelevant background interference and enhance the features of distant lane markings, thereby improving the accuracy and robustness of lane detection. Finally, a global and local slope consistency loss function is designed to adaptively adjust the shape and position of lane lines, maintaining geometric consistency between the predicted lanes and the ground truth. Experimental evaluations conducted on the TuSimple and CULane benchmarks demonstrate that the proposed approach outperforms the state-of-the-art methods in comparative experiments. Specifically, it achieves improvements of 0.58 percentage points in F₁ score and 0.19 percen-tage points in accuracy on the TuSimple dataset, and 1.14 percentage points in F₁@50 on the CULane dataset. Notably, the method exhibits more stable performance, particularly in long-range road scenarios.

A Breast Ultrasound Images Lesion Segmentation Network Based on Channel-Wise Spatially Adaptive Selective Kernel Convolution and Bidirectional Boundary-Aware Mechanism

WANG Jie, LI Luyao

2026, 54(2):  77-90.  doi:10.12141/j.issn.1000-565X.250230

Asbtract ( 212 )   HTML ( 0)   PDF (2290KB) ( 59 )  

Figures and Tables | References | Related Articles | Metrics

Breast cancer is the most common malignancy among women worldwide, and accurate lesion segmentation is of great importance for its early diagnosis and treatment. However, due to the high morphological variability of lesions and the inherent complexity of ultrasound imaging, existing deep-learning-based methods still face significant challenges in achieving satisfactory segmentation accuracy for breast ultrasound images. To address this limitation, this study proposed a novel breast-lesion segmentation network, termed CWSASKM-BBAM-Net, which is built upon the classical U-Net architecture. First, a Channel-Wise Spatially Adaptive Selective Kernel Convolution Module (CWSASKM) was introduced to adaptively adjust the receptive field size for each spatial location based on channel-specific semantic features, thereby enhancing the network’s multi-scale representation learning. Second, a Bidirectional Boundary-Aware Mechanism (BBAM) was designed to jointly model salient regions and their boundaries by integrating forward and reverse attention. This mechanism progressively improves the discrimination between non-salient areas and lesion areas, thereby refining boundary delineation. Extensive experiments were conducted on three public breast ultrasound datasets (BUSI, UDIAT, and STU). On BUSI dataset, the proposed method achieved Jaccard index, precision, recall, and Dice similarity coefficient of 71.97%, 82.85%, 81.40%, and 80.44%, respectively, outperforming the second-best approach by margins of 1.69, 1.05, 1.28, and 1.84 percentage points. On UDIAT, it attained 78.14%, 88.31%, 86.73%, and 86.10% on these metrics, with improvements of 2.75, 2.04, 0.56, and 2.01 percentage points, respectively. The proposed method also demonstrated superior overall performance on the external STU dataset. These results collectively demonstrate that CWSASKM-BBAM-Net achieves state-of-the-art performance in breast ultrasound image segmentation tasks.

Biological Engineering

Prediction of Target Inhibitor Activity by Integrating Machine Learning and Metaheuristic Algorithms

LING Fei, GU Xuerong

2026, 54(2):  91-101.  doi:10.12141/j.issn.1000-565X.250020

Asbtract ( 66 )   HTML ( 0)   PDF (3674KB) ( 56 )  

Figures and Tables | References | Related Articles | Metrics

Traditional machine learning (ML) and deep learning (DL) play a key role in predicting the activity of target inhibitors. Many models based on existing datasets can predict compound bioactivity. However, debate persists regarding whether ML or DL performs better for such prediction tasks. In this study, datasets were constructed based on different molecular representations. Ten metaheuristic algorithms were applied to optimize the hyperparameters of eleven ML and DL models, aiming to systematically compare their predictive performance and identify the optimal ones. The results show that ML and DL models whose hyperparameters were optimized by metaheuristic algorithms significantly outperformed those optimized using the traditional grid search method. Furthermore, in low-dimensional feature spaces, graph-based DL models, such as SSA-GAT and SSA-Attentive FP, can automatically extract informative features from data via an end-to-end learning mechanism, yielding better performance than ML models. In contrast, in high-dimensional feature spaces (e.g., the feature space formed by combining RDKit descriptors with ECFP, AtomPairs, and MACCS fingerprints), ML methods, leveraging the complementary information in molecular features and the powerful optimization capability of metaheuristic algorithms, can effectively capture complex feature interactions. Consequently, ML methods often demonstrate higher accuracy and robustness in high-dimensional modeling. These findings provide valuable guidance for selecting between ML and DL approaches for target inhibitor activity prediction.

Effect of Chicken Peptide-Ferrous Chelate on Iron Deficiency Anemia in Mice

LIU Ruili, SONG Xueying, MIAO Jinxin, ZHAO Mouming, SU Guowan

2026, 54(2):  102-111.  doi:10.12141/j.issn.1000-565X.250302

Asbtract ( 219 )   HTML ( 0)   PDF (5623KB) ( 58 )  

Figures and Tables | References | Related Articles | Metrics

This study systematically investigated the ameliorative effect of Chicken Peptide-Ferrous Chelate (CMP-Fe) on anemia symptoms in mice with iron deficiency anemia (IDA). The intervention effects were comprehensively evaluated through multiple dimensions, including changes in body weight, blood routine parameters, iron metabolism indicators, inflammatory responses, and tissue protection. The experimental results showed that, compared to the model group, all CMP-Fe dose groups (with low-dose, medium-dose, and high-dose administered at Fe 1.0, 2.0 and 3.0 mg/kg, respectively, based on mouse body weight) exhibited significant improvements in body weight, blood routine parameters (including red blood cell count RBC, hemoglobin HGB, hematocrit HCT, mean corpuscular volume MCV, mean corpuscular hemoglobin MCH, mean corpuscular hemoglobin concentration MCHC, and red blood cell distribution width-coefficient of variation RDW-CV), and serum iron metabolism markers (including serum iron SI, total iron-binding TIBC, transferrin receptor TRF, ferritin FER, transferrin saturation TSAT, and unsaturated iron-binding capacity UIBC), with the high-dose group demonstrating the most pronounced effects. The ameliorative effects of CMP-Fe on erythrocyte-related parameters such as RBC, HGB, and HCT were comparable to those of the positive control group and exhibited a clear dose-dependent trend (low, medium, and high dose gradients). In terms of inflammation regulation, CMP-Fe could suppress the production of serum and colonic pro-inflammatory factors (IL-6, TNF-α, and CRP), while elevating the levels of the anti-inflammatory factor IL-10 and the intestinal mucosal immune marker sIgA. Specifically, the levels of pro-inflammatory factors such as IL-6, TNF-α, and CRP in the colonic tissues of mice in the CMP-Fe group were significantly lower than those in the model control group (P < 0.05), suggesting that CMP-Fe effectively modulates the inflammatory response asso-ciated with IDA. Notably, under high-dose CMP-Fe intervention, the recovery of sIgA levels in mice outperformed that of the model control group (P < 0.05) and even exceeded that of the positive control group. Histopathological examination shows that CMP-Fe causes no significant pathological damage to organs such as the heart, lungs, spleen, and kidneys in mice, indicating its favorable biosafety. Meanwhile, this chelate also significantly alleviates pathological injuries in intestinal and liver tissues caused by iron deficiency. In conclusion, CMP-Fe effectively alleviates iron metabolism disorders, inhibits inflammatory responses, and mitigates intestinal and liver tissue damage in mice with IDA, while demonstrating good safety. As a promising novel organic iron supplement, it holds potential for application in the development of specialized pet foods and nutritional supplements for various animals such as cats and dogs.

Architecture & Civil Engineering

Research and Application of an Evaluation System for the Flexural Cost Performance Ratio of UHPC Beam Components

FAN Xueming, YE Xiaohang, ZHOU Xiaopeng, YIN Shiling

2026, 54(2):  112-122.  doi:10.12141/j.issn.1000-565X.250163

Asbtract ( 298 )   HTML ( 1)   PDF (1007KB) ( 57 )  

Figures and Tables | References | Related Articles | Metrics

Ultra-high performance concrete (UHPC) has demonstrated significant application potential in beam components due to its excellent mechanical properties, durability, and environmental friendliness. However, the high material cost often leads to the misconception that UHPC structures offer low cost-effectiveness, which hinders their widespread engineering adoption. Existing studies mostly focus on optimizing the mechanical properties of UHPC components, but there is a notable lack of systematic evaluation regarding their flexural cost performance ratio. To address this gap, this paper established a comprehensive evaluation system for the flexural cost performance ratio of beam components, aiming to provide theoretical support for the rational design and engineering application of UHPC structures. Firstly, with “performance” and “cost” as the core considerations, an evaluation indicator for flexural cost performance ratio was constructed. Subsequently, the ultimate bending moment was selected to characterize the flexural performance of components, while the material cost per unit length of the pure bending segment was used to represent economic cost. Based on these, a quantifiable cost performance ratio indicator was established. Subsequently, the classic reference method was employed, taking conventional single-reinforcement rectangular beams with appropriate reinforcement as the reference benchmark. The cost performance ratio indicator was then rendered dimensionless throgh the range standardization method. On this basis, integreating probability theory and the K-means clustering algorithm, a grading system for flexural cost performance ratio was established, ensuring both scientific rigor and objectivity. Finally, the evaluation system was applied to comparatively analyze the differences in flexural cost performance ratio between conventional beam members and UHPC beam members. The research results show that the flexural cost performance ratio of reinforced concrete beams is independent of beam width but exhibits a positive correlation with both beam depth and material strength grade. In component design, priority should be given to combinations of higher-strength concrete and highe-strength steel reinforcement, and the substitution of tensile reinforcement with prestressed steel strands can also be considered. Furthermore, if UHPC is simplely used as a direct replacement for ordinary concrete in developing new components, it does not demonstrate an advantage in flexural cost performance ratio. Only through tailored designs that leverage the mechanical characteristics of UHPC and other construction materials can component forms with a high flexural cost performance ratio be achieved.

Enhancing Effect of Rectangular Spiral Stirrups on the Fire Resistance of Concrete-Filled Rectangular Steel Tubular Columns

QIU Peiyun, ZANG Jianbo, WANG Xinxiang

2026, 54(2):  123-132.  doi:10.12141/j.issn.1000-565X.250138

Asbtract ( 201 )   HTML ( 1)   PDF (4103KB) ( 59 )  

Figures and Tables | References | Related Articles | Metrics

Internal stirrup confinement can significantly enhance the fire resistance of concrete-filled steel tubular (CFST) columns. Even without any fire protection, this approach still enables the columns to meet the design requirements for the corresponding fire resistance rating, providing a novel method for structural fire protection design and enhancing the safe service life of structures. This paper designed and fabricated three full-scale rectangular CFST columns and conducted fire test on them under the combined effects of standard fire exposure and axial loading. The study investigated the fire performance enhancement of such members via internal rectangular spiral stirrups, and examined the influence of stirrup spacing on their fire resistance. Based on the existing mechanical constitutive models of materials at elevated temperature, this study developed a numerical model for rectangular CFST columns and revealed the mechanism by which the spiral stirrups constrain the core concrete to enhance the fire resistance. The research results indicate that, with equivalent or slightly lower total steel usage, the internal rectangular spiral stirrups can significantly improve the fire resistance of such columns, with an increase in fire resistance of up to 104.1%. By reasonably designing the stirrups parameters, this type of axially compressed column can meet the Class Ⅱ fire resistance rating requirements for 150 minutes without the need of a fire protection layer on the steel tube surface. When the stirrup spacing increases from 40 mm to 80 mm, its effect on the fire resistance of such members is relatively small. The established numerical model can effectively predict the fire resistance of rectangular CFST columns with internal rectangular spiral stirrups, but its accuracy in predicting deformation at different heating time is less satisfactory.

Damage Creep Constitutive Model of Carbonaceous Slate with Different Moisture Content

HU Taotao, LIU Kemeng, ZHAO Yulong, LI Hao, GAO Xianchao, WANG Lei

2026, 54(2):  133-144.  doi:10.12141/j.issn.1000-565X.250055

Asbtract ( 110 )   HTML ( 0)   PDF (1350KB) ( 59 )  

Figures and Tables | References | Related Articles | Metrics

To explore the creep characteristics of carbonaceous schist under different moisture content conditions, based on the indoor graded loading creep test data, this paper constructed a viscoplastic body capable of describing the nonlinear accelerated creep stage throughout the entire creep process. This is achieved by parallelly connecting a nonlinear viscous element with a plastic element that characterizes yield behavior, based on data from indoor graded loading creep tests. Subsequently, this nonlinear viscoplastic body was integrated in series with the classical Nishihara model. By incorporating the softening patterns of elastic modulus and viscosity coefficient, four damage factors were introduced to establish a damage-based creep constitutive model (i.e., an improved Nishihara model) that describes the entire creep process of carbonaceous slate under different moisture content conditions. Through secondary development of a user-defined material subroutine (UMAT) in ABAQUS finite element software, numerical simulations of triaxial creep tests on carbonaceous slate under varying moisture content conditions were implemented. The applicability of the model was validated by comparing experimental creep data from rock samples with numerical simulation results. The research results show that the constructed improved Nishihara model can significantly improve the simulation accuracy in the accelerated creep stage. The graded loading creep test curves under different moisture content conditions are in good agreement with the numerical simulation curves, and the correlation coefficients are all greater than 0.9. The numerical simulation cloud map accurately reproduced the creep deformation evolution process of rock samples with different water contents, verifying the correctness and effectiveness of the proposed creep constitutive model of carbonaceous shale considering water damage and the development of the UMAT subroutine. This research achievement can provide theoretical support for the long-term stability assessment and disaster early warning of deep-water buried carbonaceous slate tunnel projects.

Investigation on Acoustic Environment of Experimental Teaching Rooms in Universities

PENG Jianxin, WU Weizhen

2026, 54(2):  145-151.  doi:10.12141/j.issn.1000-565X.250091

Asbtract ( 237 )   HTML ( 0)   PDF (1060KB) ( 57 )  

Figures and Tables | References | Related Articles | Metrics

Speech intelligibility is an important acoustic index for the acoustic environment quality in speech-oriented halls. Excessive reverberation time (RT) and high background noise levels (BNL) can significantly degrade speech intelligibility in rooms. While numerous studies have investigated the acoustic environments of university classrooms, research focusing on experimental teaching spaces remains relatively scarce. This study conducted objective acoustic measurements in 14 physics experimental teaching rooms at a university in Guangzhou, collecting room acoustic parameters including BNL, early decay time (EDT), RT, early-to-late sound energy ratio (C₅₀), and speech transmission index (STI) to analyze the acoustic characteristics of such rooms. The results show that the average BNL in unoccupied experimental teaching rooms was 37.1 dBA, which increased to 57.9 dBA during student experiments. The average values of EDT, RT, C₅₀, and STI across the measured rooms were 1.23 s, 1.26 s, -1.89 dB, and 0.54, respectively. Comparative test shows that under the same spatial dimension conditions, experimental teaching rooms exhibit shorter RT than conventional classrooms, with relatively flatter frequency characteristics. This study also found that large lab tables with cavities can help reduce RT in experimental teaching rooms and improve their RT frequency response. Further analyzed results show that 13 out of the 14 measured experimental teaching rooms failed to meet the recommended values specified in current relevant standards. It is recommended to implement appropriate acoustic treatments in these rooms to reduce RT and BNL, enhance room acoustic conditions and speech intelligibility, ensure compliance with the acoustic requirements for university experiment teaching, and facilitate effective verbal communication between instructors and students.

Inverse Method for Assembly Tolerance Itervals of Cable-Stayed Bridge Based on Proactive Fault Tolerance

WANG Xiaoming, SUN Chenjing, ZHU Chuanchao, FENG Boshun, GAO Lixiang, QIU Hongjie

2026, 54(2):  152-166.  doi:10.12141/j.issn.1000-565X.250099

Asbtract ( 84 )   HTML ( 0)   PDF (4279KB) ( 56 )  

Figures and Tables | References | Related Articles | Metrics

To enhance construction efficiency, prefabricated and assembled industrialized construction methods are widely adopted in bridge engineering. However, on-site procedural adjustments can easily lead to assembly failures, resulting in schedule and cost overruns, which contradict the original intention of rapid construction. For cable-stayed bridge prefabricated components produced according to the original single-segment cyclic process, how to achieve assembly coordination under the adjusted dual-segment cyclic process is a key challenge when complying with overall schedule decisions. Integrating the unstressed state method and tolerance allocation approach, this paper first develops a machine learning-based inversion method for the assembly tolerance intervals of cable-stayed bridges. Then a non-negative dimensionless indicator is introduced as an independent optimization objective to characterize construction operability. Next, a GA-BPNN surrogate model combined with the NSGA-Ⅱ multi-objective optimization algorithm is adopted to conduct a trade-off optimization between structural safety and design optimality. By incorporating prior error reserves, the passive assembly tolerance ranges for various components of a cable-stayed bridge are inversely derived. Finally, integrating field-measured data from preceding construction segments, an assembly tolerance design framework for cable-stayed bridges oriented toward active fault tolerance is established. Results demonstrate that the proposed interval inversion method effectively enhances on-site operability while ensuring structural safety and design optimality. Compared with passive tolerance design, the active tolerance design approach increases the tolerance interval for the girder splicing angle of segment G2 by 1.4 times and that for the unstressed cable length of stay S16 by 2.1 times. Moreover, the active tolerance analysis framework enables adaptive adjustment in assembly failure scenarios by modifying tolerance ranges of subsequent components, thereby reducing the occurrence of work stoppages and reworks.

Simplification and Reconstruction Method for Corrosion Pits on Steel Wires Based on Spherical Harmonics

HE Tiantao, NIU Tianyi, WANG Dalei

2026, 54(2):  167-174.  doi:10.12141/j.issn.1000-565X.250061

Asbtract ( 50 )   HTML ( 0)   PDF (4145KB) ( 61 )  

Figures and Tables | References | Related Articles | Metrics

As a critical load-bearing component of bridges, suspension cables are prone to corrosion under environmental exposure during service. The geometric morphology of corrosion pits, particularly those caused by pitting corrosion, induces stress concentration that directly impacts the safe operation of in-service bridges. Therefore, developing an accurate method for fitting the morphology of corrosion pits on steel wires is a prerequisite for conduc-ting refined fatigue life analysis of suspension cables. The geometry of corrosion pits is complex and irregular. Traditional methods that approximate pits with regular geometric shapes fail to account for stress concentrations in locally irregular areas, while directly using the original point cloud data, though preserving complete geometric information, entails excessively large data volumes. To efficiently and accurately assess the geometric morphology of corrosion pits, this study employed a super-depth microscope to collect point cloud data of corrosion pits. The spherical harmonic function was introduced to characterize their three-dimensional morphology of corrosion pits, simplify the dataset, and achieve geometric reconstruction. An ABAQUS finite element model was established and compared with models based on real pit data and traditional ellipsoid fitting to validate the effectiveness of the simplification and reconstruction method. The results show that using a 20 th-order spherical harmonics function can effectively simplify the corrosion pits, reducing the storage space of the corrosion pit dataset by an average of 99.47% while retaining the geometric information with high fidelity. With fatigue life as the evaluation metric, the average error is 0.80%. This study provides a theoretical foundation for the construction of mesoscale models of corrosion pits and the simplification of corrosion pit datasets.