Accurate deflection prediction is vital for structural health monitoring of large-span bridges yet remains challenging due to complex nonlinear environmental couplings. This paper proposes a hybrid deep learning framework, BGCO-PIC-DA-LSTM, for precise bridge deflection prediction. First, a Prior-Informed Correlation (PIC) strategy incorporating temperature lag terms is introduced to enhance the statistical consistency of input features. Second, a dual-stream residual Bi-LSTM network integrating adaptive temporal attention is developed to simultaneously capture long-term evolutionary trends and instantaneous dynamic fluctuations. Furthermore, a Bayesian-Gradient Cooperative Optimization (BGCO) strategy is employed to automatically configure optimal hyperparameters. Validation using in situ data from a large-span cable-stayed bridge demonstrates that the proposed method significantly outperforms baseline algorithms in prediction accuracy and robustness. Additionally, the prediction residuals exhibit characteristics approximating zero-mean Gaussian white noise, establishing a reference baseline for structural state evolution and providing a certain basis for identifying potential performance shifts.

Contact sensor-based stress monitoring of transmission towers under strong winds is often limited by complex installation and maintenance procedures and the risk of local structural damage. With the development of non-contact displacement monitoring technologies, such as laser measurement and machine vision, this study proposes a wind-induced stress surrogate model for transmission towers using MISSA-BPNN, aiming to rapidly calculate stresses in vulnerable regions from macroscopic displacement responses. First, finite element analysis is conducted to investigate the wind-induced responses of a tower-line system under different operating conditions, identify vulnerable regions, and construct a dataset using displacement and stress responses. Then, a multi-strategy improved sparrow search algorithm (MISSA) is developed to optimize the initial weights and biases of the BP neural network, thereby establishing the MISSA-BPNN model. The constructed dataset is used to train the model and build the wind-induced stress surrogate model. Results show that the vulnerable regions are mainly located at the leeward tower foot and the middle-lower tower body. Compared with the conventional BPNN model, the proposed MISSA-BPNN surrogate model reduces the MAE, RMSE, and MAPE by 22.43-24.33%. This method provides a new approach for health monitoring of transmission lines under strong winds.

During the construction of mass concrete, thermal cracking can occur due to temperature gradients, significantly affecting structural stability. To mitigate thermal cracking, accurately predicting concrete temperature and monitoring its internal thermal evolution during construction are essential. This study develops a PO-BP concrete temperature prediction model based on temperature monitoring data collected during the construction of an arch dam. The method employs a novel meta-heuristic algorithm called the Parrot Optimizer (PO) to perform global optimization of the initial weights and biases in the BP network. By analyzing factors influencing concrete temperature, six input parameters are selected: ambient temperature, initial concrete temperature, cooling water flow, inlet water temperature, concrete age, and adiabatic temperature rise. The influence of each input parameter on the concrete temperature is quantified using Sobol' global sensitivity analysis. Prediction accuracy of five models—PO-BP, BP, CNN, LSTM, and SVR—is compared using evaluation metrics. The results show that the PO-BP model delivers the best prediction accuracy and stability on the test set (R2 = 0.983 ± 0.007, RMSE = 0.265 ± 0.037 °C, MAE = 0.184 ± 0.022 °C). This confirms its feasibility for predicting concrete temperature. Further detailed analysis of the PO-BP and BP prediction outcomes reveals that PO reduces extreme errors in the BP model and improves model stability. This study provides a high-accuracy predictive tool for concrete temperature during arch-dam construction. Using these predictions, it guides the implementation of temperature-control measures. This approach contributes to better prevention of thermal cracks in dam concrete.

Physics-Informed Neural Networks (PINNs) are sensitive to how collocation and labeled points are distributed, especially in unsteady flows with localized high-gradient dynamics. Existing sampling strategies are mainly uniform or residual-driven and may underutilize physically critical regions while oversampling low-information areas. We propose a process-oriented sampling framework that jointly optimizes collocation and labeled-data placement by prioritizing feature-rich regions. The framework supports two region definitions: (i) empirical regions identified by domain knowledge, such as the Karman vortex street, and (ii) automatically detected feature-dominant regions obtained through KDTree-based neighborhood search and vorticity-divergence analysis. The framework is evaluated using a two-dimensional cylinder wake at Re = 3900 and a high-Reynolds-number wind-flow case at Re = 4.8 × 10⁷. In the cylinder-wake case, increasing collocation density in feature-rich regions reduces reconstruction errors by up to one order of magnitude. For labeled data, a “fewer but better” effect is observed: strategically placing fewer labels in feature-rich regions outperforms placing more labels in non-feature regions. In the high-Re case, feature-dominant sampling improves prediction accuracy for u, v, and p, reducing errors by 14.7

The magnetic components made of thin laminated silicon steels are viable candidates to work within the medium-frequency range, and are often exposed to the power electronic rectangular voltage excitations. The related energy loss of thin laminated silicon steels produced in that context should be predicted accurately and efficiently. However, the traditional loss models, for instance Steinmetz equations, are empirical in nature, thus requiring a lot of experimental data to extract their coefficients. Contrarily, the numerical loss models are prohibitively time-consuming. In this article, a fully analytical loss model for sinusoidal excitations is firstly derived by applying the statistical theory of losses and the fractional derivatives for the classical eddy current loss component. Then, the relation between sinusoidal excitation loss and rectangular excitation loss of duty cycle 0.5 is constructed. Thereby, a pure analytical loss model of laminated silicon steels for rectangular voltage excitations with arbitrary duty cycle is developed with the concept of separately treating the positive and negative parts of this kind of excitation. Finally, the predicted losses of two laminated silicon steel samples under the rectangular voltage excitations with different duty cycles by our proposed model, complex permeability model and improved generalized Steinmetz equation are compared with the measured ones, verifying the superiority of the proposed model, and only a few data measured under normal sinusoidal excitations are needed to identify it.

To address the dynamic response prediction of transmission tower-line systems under wind-rain coupling, this study proposes a hybrid framework integrating multi-scale decomposition, improved swarm intelligence optimization and deep learning. Variational Mode Decomposition and Complete Ensemble Empirical Mode Decomposition are adopted to extract multi-scale features and alleviate mode mixing. A multi-channel gated recurrent unit network is constructed for time-series modeling, and the improved Improved Boomerang Aeronautical Ellipse Optimization algorithm is introduced for global hyperparameter optimization. Based on simulation data, the model reduces prediction error by over 69% compared with a single gated recurrent unit and by an additional 10% compared with hybrid models without optimization, performing well under extreme conditions and demonstrating the effectiveness of the proposed framework for dynamic response modeling under wind-rain coupling conditions.

For the pre-assessment of cracking risk in concrete dams, deterministic analysis is traditionally performed based on temperature control design data. However, numerous uncertainties inherent in the construction process inevitably affect the reliability of temperature and stress field simulations. Accordingly, this study considers the uncertainties of thermal parameters and boundary conditions of concrete during construction. Based on structural reliability theory, a reliability analysis method is developed by combining the Killing-strategy Adaptive Beluga Whale Optimization (KABWO) algorithm with the Support Vector Machine (SVM). The proposed method is applied to an engineering case. It establishes the response relationship between the temperature-stress field and influencing factors of the concrete dam during construction. Based on this relationship, the cracking risk of the target block is assessed in advance, and the reliability index is used to quantitatively evaluate the cracking risk. The results indicate that the SVM optimized by the KABWO algorithm can construct accurate response surface equations linking random variables to thermal stress. As a result, it can serve as an efficient surrogate for time-consuming temperature-stress field simulations. This approach enables a more scientific assessment of cracking risk during dam construction and provides both theoretical significance and practical guidance for temperature regulation and crack mitigation.

The model for simulating the dynamic radar echo from wind turbines plays an important role in enhancing the radar antijamming capabilities of radar. However, the extensively used scattering point spacing model can hardly simulate the influence of the wind turbine's irregular blades, which makes it insufficiently accurate to simulate the radar echo from wind turbines. Therefore, the transverse arrangement is introduced to construct the irregular structure of the wind turbine's blades. Based on the waveform matching, the parameters of the transverse offset are determined, with which the improved model of the wind turbine can be constructed. The Monte Carlo algorithm is used to accelerate the process of searching for the best parameters of the transverse offset. Finally, the scaled model experiment with a GW 77/1500 wind turbine was carried out to verify the accuracy of the improved model, and the calculation times for simulating the radar echoes from 1 to 10 wind turbines were compared, which shows that the improved model can increase the accuracy of radar echo by 40% compared to the traditional model without increasing the calculation scale.

This study proposes an iterative method based on thermal equilibrium equations to calculate the radial temperature distribution of long-span overhead transmission lines under forced convection. This paper takes the ACSR 500/280 conductor as the research object, establishes the three-dimensional finite element model considering the helix angle of the conductor, and carries out the experimental validation for the LGJ 300/40 conductor under the same conditions. The model captures internal temperature distribution through contour analysis and examines the effects of current, wind speed, and ambient temperature. Unlike traditional models assuming uniform conductor temperature, this method reveals internal thermal gradients and introduces a novel three-stage radial attenuation characterization. The iterative method converges and accurately reflects temperature variations. The results show a non-uniform radial distribution, with a maximum temperature difference of 8 °C and steeper gradients in aluminum than in steel. Increasing current raises temperature nonlinearly, enlarging the radial difference. Higher wind speeds reduce both temperature and radial difference, while rising ambient temperatures increase conductor temperature with a stable radial profile. This work provides valuable insights for the safe operation and optimal design of long-span transmission lines and supports future research on dynamic and environmental coupling effects.

In the simulation of concrete thermal stress fields, thermal parameters are crucial for calculating the concrete temperature field. In actual construction, due to the adjustment of the concrete mixing ratio and the changing external environment (temperature fluctuations, cooling conditions, solar radiation, thermal insulation measures, etc.), there are significant differences between the thermal parameters obtained in tests and the actual working conditions, which affect the simulation accuracy. Therefore, the inverse analysis of concrete thermal parameters under real working conditions can be carried out based on the measured temperature data. A method for inverse analysis of thermal parameters of arch dams using the walrus optimization algorithm (WaOA) is proposed. To verify the accuracy of the inversion parameters, twelve classical test functions are used to compare the three algorithms to evaluate their fitness. The efficiency difference is analyzed by nonparametric methods such as Fredman and Wilcoxon rank sum test. The results consistently indicate that the walrus optimization algorithm performs better. Furthermore, the WaOA is utilized for the parameter inversion of an arch dam in the downstream area of the Jinsha River. We bring the inversion results into different dam sections to calculate the temperature field during construction, which effectively verifies the efficient solution ability of the WaOA for the inverse analysis of concrete thermal parameters under complex engineering backgrounds.

During the construction period, the phenomenon of stress concentration at the top arch of the corridor is obvious, which is easy to produce surface cracks. As a consequence, it is of great importance to research the temperature and stress changes of the corridor during the construction period, analyze the cracking risk of the top arch of the corridor, and put forward the corresponding temperature control measures. This study, the temperature gradient test of the top arch of the corridor is conducted, utilizing the distributed optical fiber buried in the construction place of Baihetan Arch Dam. Considering the influence of creep, the finite element method is used to simulate the thermal stress of the top arch concrete of the corridor, and the cracking risk of the top arch concrete is evaluated. According to varying temperature drops, it is suggested that different temperature control measures to maintain the heat of the top arch of the corridor. The recommended measures are as follows: when the temperature drops by 6 degrees C, 12 degrees C and 18 degrees C, the roof vault of the corridor shall be covered with 20 mm, 40 mm and 50 mm polyethylene insulation covers respectively, and the interior of the corridor shall not be sprayed with 20 mm and 40 mm polyurethane respectively. Temporary storm doors shall be set at the entrances and exits of the corridor. The control measures of temperature can meet the requirements of temperature control and crack prevention of the concrete of the dam body of Baihetan Arch Dam.

The simulation accuracy of concrete temperature fields is limited by thermal parameters, which often differ significantly from real values due to environmental temperature fluctuations, cooling water variation, solar radiation, and surface insulation in actual construction. Traditional inversion methods may fall into a local optimum and have low convergence efficiency; it is difficult to meet the requirements of high-dimensional nonlinear optimization. Therefore, this paper proposes a Multi-Strategy Improved Sand Cat Swarm Optimization algorithm (MISCSO). The population is initialized using a cubic chaotic map to enhance randomness, and the triangle walk mechanism, Levy flight, and lens imaging reverse learning strategy are integrated to balance the global exploration and local development capabilities. The advantages of MISCSO performance are verified by twelve benchmark function tests and nonparametric tests. Key thermal parameters were screened by Sobol global sensitivity analysis. A finite element model considering the ambient temperature, boundary conditions, and cooling water was built through an engineering example, and the thermal parameters were obtained by using the optimization algorithm. After substituting the inversion parameter value into the numerical model, the absolute error between the calculated temperature and the monitored value is within 1.0 degrees C, which verifies the effectiveness of the inversion method.

Arch closure grouting is an essential procedure for attaining structural completion during the construction of a concrete arch dam. After the closure, due to the continuous heat emission from cement hydration, the internal temperature of concrete rises rapidly, which affects stress distribution and structural stability. In order to accurately predict the temperature evolution of concrete pouring blocks after arch closure, this paper conducted a comparative study using neural networks and finite element methods. First, a hybrid model, CNN-BiLSTM, was constructed. This model integrates a Convolutional Neural Network (CNN) and a Bidirectional Long Short-Term Memory (BiLSTM). The Weighted Mean of Vectors algorithm (INFO) was then introduced to optimize the model parameters. The temperature variation trend of concrete pouring blocks after arch closure was predicted using this approach. Simultaneously, considering the factors such as external temperature, cooling water, adiabatic temperature rise and concrete age, a three-dimensional finite element model of concrete pouring blocks was established to simulate the temperature field distribution of concrete. The comparison results indicate that both methods can achieve the prediction accuracy required by the project (with an error of less than 2 degrees C). Among them, the finite element simulation performs better in terms of stability (with a difference of less than 1 degrees C from the measured value). At the same time, the INFO-CNN-BiLSTM model exhibits significant temperature fluctuations during certain periods and demonstrates insufficient generalization ability. However, it offers the advantage of high computational efficiency.

PurposeDuring the construction process, temperature cracks are easily generated at the dam orifice, which greatly threaten the stability of the dam structure. Therefore, to mitigate the adverse effects of a sudden temperature drop, it is crucial to implement temperature monitoring and develop specific temperature control measures.Design/methodology/approachIn this paper, based on the monitoring data obtained from the concrete surface temperature monitoring test conducted on the floor of the flow channel of Baihetan Arch Dam, the temperature field and thermal stress of concrete at the floor of the flow channel are calculated. Subsequently, the stress field distribution is analyzed under the influence of a sudden temperature drop. Finally, corresponding temperature control measures are proposed for different temperature drop patterns, and the effectiveness of the corresponding temperature control effects is analyzed.FindingsIn total, during the temperature dropped by 8 degrees C, 12 degrees C and 16 degrees C within 3 days, the use of polyethylene coil insulation material with a thickness of at least 2 cm, 3 cm and 4 cm can weaken the adverse effects of a sudden temperature drop on the floor of the flow channel structure.Practical implicationsThis study provides some references and suggestions for the preparation of temperature control measures for the construction of the ledger wall of the flow channel structure at the construction site.Originality/valueCompared with previous studies, this study uses numerical simulation methods to calculate the temperature field and analyze the stress field distribution under the influence of a sudden temperature drop. In addition, the temperature control effects of different insulation measures were compared and optimized for selection.

Introduction During the construction of a tension wire across a live line, the pulled wire may fall and collide with the collision sealing net structure, posing a risk to the safety of the sealing device. It may also come into contact with, or in close proximity to, the live line, potentially causing a flashover. Therefore, it is very important to accurately obtain the changes in the strength and vertical displacement of the sealing net structure during the impact of falling wires for the design and construction of sealing net devices.Methods According to the construction process of the overhead line, the accidental process of the broken line is analysed, and for the assumed defects of the existing simulation and the simulation difficulties, the sequence solution is introduced, and through the implicit-explicit sequence solution method, combined with the theory of nonlinear dynamic analysis, the simulation flow of the conductor-seal network collision is established.Results Based on the relevant parameters presented in this paper, the equivalent load and the overhead line state equation were used to calculate the bending stress and cable tension under accidental conditions. The results showed that the bending stress in brace section 1 was -119.4 MPa, and the bearing cable tension was 32,329 N. These values differed from the simulation results by less than 5%, which is within the permissible error range, thereby demonstrating the feasibility and validity of the simulation calculations.Discussion This study used numerical simulation to analyze the impact of a falling wire on the sealing network structure, highlighting displacement and stress changes. The implicit-explicit sequential solution method successfully simulated the entire process and addressed multi-step loading issues that traditional methods couldn't handle. The results provide key insights for sealing network design and advance transmission line construction safety research. However, the use of specific Dyneema ropes and epoxy resin sealing networks may limit the generalizability of the simulation results. Future studies should compare multiple materials and consider environmental factors affecting material performance.Conclusion This study effectively addresses the multi-step simulation challenges related to structural prestressing and wire-network collision. The complete collision process between the wire and the sealing network has been accurately modeled and simulated.

Currently, most of the monitoring of the construction process of tension wire construction relies on the experience of the construction personnel, and there are great construction hazards. For this reason, combined with the current physical networking technology, the development of a set of traction machine and tension machine release and tension size, traction walking plate force, and attitude of the danger of allround monitoring. In addition, there is an intelligent monitoring system for tension wire construction with real-time information transmission and warning. After analyzing the construction process and construction procedures of tension wire construction, it is determined that allround safety monitoring of tension wire construction is carried out from the aspects of traction speed of traction machine, traction force of traction machine, release speed of tension machine, tension of tension machine, force of traction walking plate, and tilting angle of traction walking plate, etc., and then it builds up the overall structure of the system’s monitoring points. Aiming at the characteristics of field construction of tension wire, a real-time data collection and processing model is designed, and LoRa communication technology and 4G communication technology are selected to interconnect the self constructed local area network and 4G wide area network, so as to complete the on-site interaction and long distance transmission of data. In view of the characteristics of the system with many types of monitoring volume and wide distribution, the on-site data transmission and early warning program adopts a polling mode to realize the orderly utilization of each channel through time division. The test results indicate that the system has an error of less than 2

The blasting excavation of a new tunnel significantly impacts the lining of existing tunnels during the construction of zero-spacing twin tunnels. In this study, a field test was first carried out to study the vibration velocity of an existing lining in zero-spacing twin tunnels. LS-DYNA software was then employed to investigate the dynamic response characteristics under various blasting factors. Finally, the use of damping layer was considered to analyze its effects on the dynamic response of the existing tunnel lining. The maximum vibration velocity was found at the hance adjacent to the new tunnel, with a value of 17.51 cm/s in the tests. The maximum principal tensile stress(PTS) from the simulation was also shown at the same hance of the existing lining to have an intensity of 3.27 MPa. Additionally, the blasting of the rock mass close to the existing tunnel in the first step would result in a higher PTS intensity of 3.54 MPa within the lining, compared to the blasting of the rock mass away from the existing tunnel in the first step. However, the cumulative stress disturbance caused by the former to the lining was smaller than that of the latter. The maximum PTS at the hance was also found to linearly decrease with the reduction of the excavation footage and the maximum PTS was 2.3 MPa at an excavation footage of 0.9 m. The damping layer of the existing tunnel would contribute to a significant reduction of PTS. Therefore, a damping layer thickness of 2.5 cm was suggested in the study considering the effects of the damping and mechanics of the lining.

For the first time, low-heat cement was used in the entire dam section of Baihetan Dam, but the thermal properties of low-heat cement under construction conditions have yet to be fully studied. The thermal parameter values of low-heat cement may differ significantly from the indoor test values or specification values due to factors such as ambient temperature, cooling through water, and surface insulation under actual site conditions. Therefore, in order to obtain more accurate values of the thermal parameters, the hybrid swarm intelligence algorithm and field temperature monitoring data are used to identify the concrete thermal parameters of Baihetan arch dam. To overcome the shortcomings of Particle Swarm Optimization (PSO) that is easy to fall into local optimum and Artificial Bee Colony (ABC) that has insufficient development ability, an Integrated Algorithm Based on ABC and PSO (IABAP) is established. Through eight different test functions and comparing with other different algorithms, it is verified that the IABAP algorithm has certain advantages in terms of convergence speed and accuracy. Considering the influence of ambient air temperature and multi-shift water cooling during construction, IABAP is applied to the inversion of concrete thermal parameters with the same strength, different strength and different gradation of Baihetan arch dam. The computational results show the good performance of the IABAP algorithm in engineering applications on the one hand, and the applicability and reliability of the parameter inversion on the other hand, which can meet the accuracy requirements of practical engineering. At the same time, the conjecture that the thermal parameters are consistent in adjacent dam sections was verified by bringing the thermal parameters into the adjacent dam sections for simulation calculations. Finally, the experimental values of thermal parameters of low-heat cement concrete of Baihetan Dam were compared with the inverse values to analyze the change law of thermal parameters, and it was found that the final adiabatic temperature rise of low-heat concrete was smaller than the indoor experimental values during the actual construction.

The temperature of concrete arch dam during construction is affected by factors such as ambient temperature, cooling through water, and surface insulation. There is often a significant discrepancy between the actual thermal parameters and the values measured in indoor experimental tests. In this study, intelligent identification of thermal parameters is performed based on temperature monitoring data and an intelligence optimization algorithm so that the thermal characteristics of concrete in a real pouring environment can be obtained in real time. A hybrid algorithm of particle swarm optimization and grey wolf optimizer (HPSOGWO) is proposed. The performance of the hybrid particle swarm algorithm is verified to have some advantages over the other seven algorithms by twelve different test functions. Considering the influences of environmental temperature changes and multistage cooling water, HPSOGWO is utilized for inverse analysis of the thermal parameters of concretes with the same strength grade and concretes with different strength grades. The analysis results show that the concrete temperature values calculated based on the inversion of the HPSOGWO algorithm are in good agreement with the measured values, and the HPSOGWO algorithm has good adaptability in the inversion of thermal parameters of arch dams. The inversion results are of essential to clarify the relationship between temperature changes and thermal parameters.