Two polymers (I3CAld-Ch-POP and Isatin-Ch-POP) were obtained by imination reaction of chitosan with indole-3-carboxaldehyde and isatin, respectively. Their iodine adsorption performance in KI3 aqueous solution and the structure-activity relationship were investigated. The modification of chitosan with indole-3-carboxaldehyde improved the adsorption effect toward iodine species. After adsorption for 48 h in 3 mL of 100 mg/mL KI3 aqueous solution with 50.0 mg of I3CAld-Ch-POP, the I2 adsorption capacity reached 2.358 g/g, and 80.4% of the I2 adsorption efficiency was maintained after three cycles. The adsorption process of I3CAld-Ch-POP toward iodine species involved both physical adsorption and chemical adsorption. The excellent adsorption and regeneration performance make I3CAld-Ch-POP a potential adsorbent for iodine in aqueous phase.

For the problem of defects on steel surfaces, a defect detection algorithm is proposed to improve the YOLOv8n model. First, the backbone network of YOLOv8n is improved using a focused linear attention mechanism to enhance the model's ability to detect smaller targets in steel defects. Second, the model inference delay is reduced and the model performance is improved by introducing the SeaFormer module. Again, the CIoU loss function of the YOLOv8n model is converted to MPDIoU, which simplifies the computation process while taking into account the overlapping or non-overlapping regions, centroid distances, as well as width and height deviations. Finally, the detection accuracy is further improved by the knowledge distillation technique without increasing the model size and the number of parameters. The improved algorithm achieves a mAP50 of 77% on the GC10-DET dataset, which is 12.9% higher than the original YOLOv8n model. And the AP values reached 96.0%, 99.5%, 97.2% and 99.5% in the identification of punched holes, welded seams, creases and waist creases, respectively. The experimental results prove that the improved model has obvious advantages in the detection performance of steel surface defects detection.

This study proposes an improved YOLOv9-based helmet wearing detection algorithm for non-motor vehicle riders, aiming to solve the problems of low accuracy and high missed detection rate under high-density and complex riding scenarios.Firstly, the Slide Loss loss function was incorporated into YOLOv9 to enhance the model's detection performance on small targets and occluded objects. Secondly, Depthwise Separable Convolution (DSConv) modules were introduced into the backbone network, reducing the model's parameter count. Lastly, a DASI small target detection module was added to the model's Head network, improving the detection capability for small targets. Compared to the YOLOv9 algorithm, the improved algorithm achieved increases of 2.3% in Precision, 1.9% in Recall, and 2.8% in mean Average Precision, while the parameter count was reduced by 9.1% compared to the original model, enabling high-precision detection of non-motor vehicle helmet wearing.

Urban rail transit systems often face issues of passenger congestion during periods of insufficient capacity. To address this problem, it is crucial to forecast future network congestion and provide passengers with optimal travel routes based on these predictions, thereby achieving balanced passenger flow in the rail transit network. Therefore, this paper proposes a method for recommending urban rail transit travel routes based on Origin-Destination (OD) prediction. This method utilizes graph convolutional networks and gated recurrent units to predict OD passenger flows in the urban rail transit network. Subsequently, it calculates the future traffic volume of rail sections as an indicator of congestion level. Furthermore, a personalized travel route recommendation method based on route cost is designed to suggest travel routes for passengers. By computing the cost of K shortest-time routes, the method prioritizes future travel routes for passengers. Case studies are conducted using the Shanghai Metro as a research subject to validate the proposed model. The results demonstrate the model's accurate prediction of OD passenger flows and the reasonableness of the recommended personalized travel routes for passengers.

This study investigates the scheduling problem of operational states within a cluster of dryers. The dryer cluster consists of multiple dryers, where each dryer must perform a regeneration operation after completing its drying tasks. There is a cooperative relationship between the dryers, where the regeneration process has a certain time flexibility after the drying operation of one tank is completed. Given the impact of China’s time-of-use electricity pricing policy, significant differences in electricity costs exist between different time periods. By adjusting the timing of the regeneration operation to suitable periods, it is possible to meet production requirements while reducing electricity expenses. To address the challenge of training models for dryer clusters, this paper proposes an electricity scheduling model for dryer cluster regeneration based on a Markov decision process. To optimize the scheduling of the dryer cluster, a method using the Proximal Policy Optimization (PPO) algorithm is introduced for training the dryer cluster. Experimental validation was conducted using a cluster of seven dryers under different working conditions. The results show that the fewer tanks involved in the drying operation, the more significant the optimization effect. After 240 hours of continuous operation with different numbers of tanks performing drying tasks, electricity costs were reduced by more than 20% compared to the original costs.

A multi strategy improved Dung Beetle Optimizer (IDBO) is proposed to address the slow convergence speed and low optimization accuracy of the Dung Beetle Optimizer (DBO) algorithm. Firstly, the population is initialized using Piecewise chaotic mapping to improve diversity, avoid early convergence to local optima, and make the initial solution distribution more uniform to improve the coverage of the search space and quickly find better solutions in complex problems; Secondly, adaptive inertia weights are used to update the position information of the beetle, thereby improving the global exploration and local development capabilities of the algorithm. Finally, adaptive t-distribution mutation is used to dynamically adjust the population position to improve the convergence rate of the algorithm. In the experimental simulation, by comparing the convergence curves, optimal values, means, and standard deviations of six benchmark test functions, the results showed that the improved beetle algorithm exhibited significant advantages in convergence speed, accuracy, stability, and optimality, verifying its superiority. Future research will be based on this improved algorithm to explore its application in engineering optimization, in order to solve practical engineering problems.

Aiming at the problem of large time delay in temperature control of industrial process, the control principle and mathematical model of predictive PI control algorithm are studied. Laplace inverse transform and differential separation method are adopted to obtain the mathematical model of predictive PI controller incremental discrete control algorithm in the time domain. The PLC program of predictive PI control algorithm is designed and written, and the controller algorithm program packaged with FB function block is tested. Simulation experiments show that the control quality of temperature control system with large time delay can be effectively improved, and it has good reliability and robustness.

The temperature control of the drying room had been an important link in the coating process and had had a significant impact on the quality of automobile coatings. In this paper, aiming at the temperature characteristics and control difficulties of the automobile coating drying room, the control algorithm of fuzzy PID had been written in the PLC, and the temperature model of the drying room had been built in Simulink. Data exchange between the PLC and Simulink had been achieved through OPC communication, and the temperature response curve had been generated to conduct simulation verification on the temperature control system of the drying room. The Simulation results show that the fuzzy PID controller is superior to the traditional PID controller in terms of stability and adjustment time, and has certain application value in solving complex temperature control problems of the drying room.

To address the issue of reduced accuracy in trajectory tracking due to link oscillations in the two-degree-of-freedom serial flexible manipulator, an active disturbance rejection control method based on an improved tracking differentiator is proposed. By establishing the dynamic model of the system, a novel tracking differentiator is designed using the Sigmoid function. Combined with a nonlinear state error feedback control law and an extended state observer, a complete active disturbance rejection control scheme is constructed. Simulation results demonstrate that this method effectively suppresses oscillation phenomena and improves trajectory tracking control performance.

Safety is the basis of human survival and development, especially in banks, railway stations, museums and other densely populated places, danger may occur at any time, aiming at the problems of missed detection and false detection of dangerous goods caused by the lack of attention of front-line security personnel, this paper proposes to improve the YOLOv8s on the detection of dangerous goods in X-ray images. Firstly, a dynamic convolution module (ODCov) is introduced on the basis of the original model to improve the accuracy of dangerous goods detection and feature extraction ability, especially in complex environment detection. Secondly, an efficient Multi-scale Feature Fusion Network (BiFPN) is introduced, which can better integrate features of different scales and improve the detection ability of the model for multi-scale targets. Finally, the Exponential Moving Average attention mechanism is introduced to improve the target detection performance and obtain more detailed information of dangerous goods. The improved YOLOv8s model was tested on the SIXray dataset, and the average accuracy of the test results reached 93%, which was 1.8% higher than that of the original model,Achieve an 81.7% increase in frames per second.

To reduce backflow and improve the output flow rate of the valveless piezoelectric pump, this paper develops a structure resembling a fish tail shape, which increases the resistance to backflow through an internal built-in flow-blocking structure. This special design promotes liquid flow, reduces backflow, and imparts unidirectional flow characteristics. The influence of key parameters of the fish-tail-shaped flow-blocking structure in the pump on the output flow rate is analyzed, and a theoretical formula for the flow rate is established. The velocity and pressure distributions within the pump chamber are simulated, and the simulation results verify that this structure effectively suppresses backflow in the valveless piezoelectric pump. A series of prototypes are fabricated to test the flow rate and output pressure of this valveless piezoelectric pump. By comparing the experimental results, the optimal parameters for the fish-tail-shaped pump are determined. The research findings indicate that the fish-tail-shaped flow-blocking valveless piezoelectric pump can achieve unidirectional fluid transport. The pumping performance is optimal when the included angle (α angle) of the fish-tail-shaped flow-blocking structure is 120°. At a driving voltage of 300?V and a driving frequency of 18?Hz, the maximum output flow rate is 38.85?mL/min. At a driving voltage of 300?V and a driving frequency of 9?Hz, the maximum output pressure is 460Pa.

Taking a linkage-type robotic am as the research object and based on its actual parameters, an improved D-H parameter method was adopted to establish the mathematical model of the robotic arm. The forward and inverse kinematic equations of the mechanical arm were analyzed and derived, Using the Robotics Toolbox in MATLAB, the robotic arm was modeled and analyzed to verify the forward and inverse kinematic solutionsn. Trajectory planning in the joint space was performed to analyze the motion characteristics of the robotic arm. The workspace of the robotic am was obtained using the Monte Carlo method. Finally, the manipulability ellipsoid was constructed through the Jacobian matrix to provide a theoretical basis for the control of the robotic arm.

To address the issues of inadequate data pattern mining and low efficiency in utilizing existing information, we propose the Basisformer Time Series Forecasting Model integrating Pattern Decomposition. First, a novel decomposition architecture is designed. This architecture breaks the convention of using sequence decomposition as a preprocessing step, integrating it as a basic internal block of the deep model, thereby enhancing the model's decomposition capability for complex time series. Then, the Basisformer model, which possesses powerful forecasting capabilities, is utilized to predict the time series patterns of different modes, yielding efficient and accurate predicted data. Finally, a linear layer module is designed to fuse the predicted data for output, consequently generating the final prediction result. This process improves the model's fitting ability and generalization ability, thus further enhancing the overall forecasting performance. The proposed model is validated using four public datasets covering energy, economics, and weather aspects. The experimental results demonstrate the superior forecasting performance of the proposed algorithm.

As a key preprocessing step of recognition technology, the accuracy of face detection is easily affected by complex environment. In this paper, an interactive improved interface based on MATLAB is proposed for the traditional Viola-Jones algorithm. The Haar feature template is used to extract the multi-level structural features of the face, and the cascade classifier is used to achieve fast region screening. An adaptive threshold adjustment mechanism is designed to allow users to adjust the detection sensitivity of facial features through the Graphical User Interface interface. Finally, the integrated detection system supports multi-site independent detection and result visualization. The experiment is verified by ORL and WIDER FACE database. The results show that in the single face scene, the face detection rate is 100 %, and the eyes, nose and mouth are 92.5 %, 97.5 % and 95 % respectively. The face detection rate in the multi-person scene is 96.36 %, which is 12.7 % higher than that in the fixed threshold mode. The comprehensive performance is close to the MTCNN model, and the memory usage is only 25 % of the latter, which provides a practical solution for accurate detection in complex scenes.