Aiming at the basic ant colony algorithm's problems of long path planning time, slow convergence speed, and high number of iterative stabilization in 2D grid maps for mobile robots, an improved adaptive elite ant colony algorithm is proposed. The algorithm improves the heuristic information function by introducing a distance parameter factor, selects the next node by using an adaptive pseudo-random state transfer rule, and also fuses the angle guidance factor into the transfer probability to reduce the search blindness, thus shortening the search time. In addition, an adaptive pheromone weight updating strategy is defined to reward the pheromone only for the optimal paths found in the contemporary search, which further improves the convergence speed. The ablation experiments, comparative experiments under different scales and environments show that the improved algorithm plans better paths and converges faster, verifying the superiority and feasibility of the algorithm.

In addressing the issue of traditional algorithms often exhibiting suboptimalperformance when applied to the intelligent control system of an inverted pendulum, we propose a design for an inverted pendulum intelligent control system based on the salp swarm algorithm (SSA). Focusing on a single inverted pendulum as the controlled object, we establish a mathematical model for the inverted pendulum. The control approach involves initially employing a neural network, followed by the optimization of the neural network's weights using the SSA algorithm. This optimization aims to identify the optimal control parameters. The entire system is simulated using the Python software, with the ultimate goal of achieving stability in the inverted pendulum system.

In view of the problems found in the process of path planning by the traditional Rapid Expansion Random Tree (RRT) algorithm in the process of path planning, such as large randomness, poor goal orientation, too many redundant nodes, slow path planning speed and poor trajectory smoothness, an improved RRT algorithm was proposed to enhance the goal orientation, reduce the redundant redundant nodes and optimize the path at the same time. Firstly, in view of the problems of poor goal orientation and long search time of the traditional RRT algorithm, a probabilistic sampling strategy was added to the sampling to enhance the goal orientation. Secondly, the global adaptive step size method can be used to dynamically adjust the step size according to the spatial size of the obstacles in the map, so as to achieve fast path planning and enhance the exploration ability of the map. In order to solve the problem of too many redundant nodes and slow path planning speed in the planning process, the greedy optimization strategy was combined to reduce the redundant nodes and improve the planning speed. Finally, the cubic B-spline curve was used to smooth the generated path. The experimental results show that the improved RRT algorithm can effectively improve the planning time, path length and smoothness.

Tool wear will cause adverse effects on industrial production. With the development of industrial processing driven by intelligent manufacturing, research on automated tool wear intelligent recognition system has gradually emerged, aiming to improve processing efficiency and prolong the service life of turning tool processing to reduce costs. In this paper, a tool wear classification method for CNC machine turning based on EfficientNetV2 network is used to solve the problems of inaccurate wear information recognition, large amount of calculation and low accuracy of the current model parameters. The EfficientNetV2 network can automatically select features, which is more intuitive and accurate, and achieves a high classification accuracy, so as to distinguish the wear of the turning tool.

Tomato fruit detection is a key issue that needs to be addressed in order to achieve mechanization and automation of tomato harvesting. In response to the complex background, dense fruit distribution, and leaf obstruction in the tomato growth environment, an optimized YOLOv7 mature tomato recognition model is proposed. Based on the YOLOv7 model, the main network's ELAN module is replaced with a P-ELAN module to reduce network parameters and computational load, while enhancing the network's feature extraction capability. Additionally, an LSK attention mechanism is added in front of the detection head to dynamically adjust the receptive field using the feature extraction module, more effectively handling the differences in background information required for different targets. Finally, the EIoU loss function is introduced to more effectively guide the model in learning more accurate bounding box predictions, thereby accelerating the convergence of prediction boxes and improving the regression accuracy of prediction boxes. The improved algorithm not only has high recognition accuracy but is also more lightweight, making it well-suited for application in mature tomato harvesting scenarios.

Aiming at the localization of traditional heat exchange station system and complex client configuration, the remote control system of heat exchange station is developed based on Web technology. Through the OPC server, the data communication between PLC data and Web application is realized, and the operation screen is published with HTML technology. The practical application shows that the system can be accessed remotely through the browser, and the operation is simple, and the purpose of remote online monitoring is achieved.

As a critical piece of industrial production equipment, the operational status of an air compressor directly affects the success of production. However, traditional fault diagnosis methods struggle to accurately obtain fault characteristics. The feature distribution differences between different working conditions are not sufficiently measured by domain adaptation, making it difficult to achieve high recognition accuracy. Additionally, background noise generated during the operation of air compressors introduces interference that impacts fault identification accuracy. To overcome these limitations, a feature fusion-based fault diagnosis method for air compressors is proposed. Firstly, Mel-frequency cepstral coefficients (MFCC) features and wavelet transform features of the air compressor are extracted separately. Then, at the decision layer, confidence scores and predicted bounding boxes are fused late in the process, and the best network model is selected based on evaluation metrics to complete the classification. Comparative experimental results show that this feature fusion method significantly improves fault identification accuracy.

In order to solve the problem that magnetic levitation ball system is susceptible to internal parameter changes and external disturbance, an adaptive inverse sliding mode control algorithm (NDO-ABSMC) based on nonlinear disturbance observer is proposed. Firstly, the state space model of magnetic levitation spherical system is established, and then a nonlinear disturbance observer (NDOB) is designed to observe the external disturbance of the system. It is proved that the observation error of the observer can converge in a finite time. Secondly, in order to overcome the uncertainty of the system, an adaptive inverse sliding mode controller based on nonlinear disturbance observer is designed. Finally, the proposed method (NDO-ABSMC) is verified by simulation. The results show that, compared with the inversion sliding mode control algorithm (BSMC), the adaptive inversion sliding mode control based on nonlinear disturbance observer has obvious suppression effect on the disturbance, the trajectory tracking effect is remarkable,which optimizes the control effect of the magnetic levitation ball system and enhances the robustness of the system.

To prevent the excitation system from misregulating due to the primary fuse of PT and the primary system ground fault, the sampling voltage of PT break logic basically adopts the line voltage method. The high-voltage primary fuse of the generator outlet PT is a special operating condition. Under this condition, the operating unit has a high risk of false over-excitation tripping. In order to quickly identify the PT slow-fuse fault of the excitation system, some excitation devices add the excitation slow-fuse logic. The PT slow-fuse logic mostly adopts the dual-PT comparison method. Since the PT slow-fuse set value of the excitation system is based on the line voltage, and the logic criterion is based on the phase voltage, there is a conversion relationship between the two base voltages, resulting in the irrationality of the default set value of the excitation system PT slow-fuse on site, which is not conducive to the rapid judgment of PT slow-fuse. Through on-site tests and the derivation of the relationship between the sampling voltage of the excitation system and the excitation display voltage, optimization suggestions are put forward for the PT disconnection, PT slow-fuse logic, and PT disconnection and slow-fuse setting values of the excitation system, providing reference experience for the safe and stable operation of the excitation system.

In recent years, computer-assisted medical imaging diagnosis has gradually become a research hotspot in this field. In order to better classify and identify medical image features, this study proposes an image recognition method that integrates convolutional neural networks and improved iterative deep learning based on adaptive learning. In the process, a randomized fusion improved convolutional neural network is introduced to cope with the multimodal feature extraction of medical images, and combined with improved iterative deep learning to avoid the loss of image data information, and finally complete the recognition of image information. The results show that the research method is experimented on the training set and the validation set. When the iteration is carried out to the 28th and 17th times, the system begins to stabilize, and the corresponding loss function values are 0.0124 and 0.0112 respectively. When the precision of the four algorithms is 0.900, the recall rates of the improved deep learning model, LeNet-5CNN model, IYolo-v5 model and the research method are 0.6232, 0.5791, 0.6774 and 0.8369 respectively. The recognition accuracy of the research method for the five diseases is significantly higher than 95%. The above results indicate that the research method has a fast convergence speed and accuracy, and can be widely used in image diagnosis and recognition of various types of diseases.

In recent years, Transformer has gradually become the mainstream architecture in computer vision. Its broad expressiveness and high parallelism give it the ability to match the performance of convolutional neural networks (CNNs). However, there are two main problems in applying the attention mechanism to computer vision at the current stage: high computational complexity and the need for a large amount of training data. To address these issues, a category-query based visual Transformer model (OB_ViT) is proposed. The innovation lies in two aspects: the introduction of learnable category queries and the use of a loss function based on the Hungarian algorithm. Specifically, a learnable category query is used as input to the decoder, which allows reasoning about the relationship between target categories and the global image context. In addition, the Hungarian algorithm is used to enforce unique predictions, ensuring that each category query learns only one target category. Experimental results on the Cifar10 and 5-class Flower image classification datasets show that the OB_ViT model achieves significantly improved learning accuracy while reducing the number of parameters compared to ViT and ResNet50. For example, on the Cifar10 dataset, there is a 15% reduction in parameters and a 22% improvement in accuracy. 

Aiming at the problems of slow convergence speed, insufficient exploration ability and easy to fall into local optimum of sparrow search algorithm (SSA), an improved sparrow search algorithm (OSSSA) based on multi-strategy fusion is proposed. Firstly, the diversity of population is initialized with the help of Tent chaotic map to improve the quality of initial solution; Secondly, the first stage exploration strategy of osprey algorithm is introduced in the location update of discoverer to improve the exploration ability of population to local search; Finally, cauchy mutation and variable spiral search strategy are introduced to update the follower position to improve the search efficiency and global search performance of the algorithm, reduce the probability of the algorithm falling into the local optimal solution and enhance the global optimization ability of the algorithm. On this basis, eight benchmark functions are simulated to evaluate the optimization performance of the algorithm. Through the analysis of simulated images and data, the improved sparrow search algorithm has greatly improved the convergence speed and optimization accuracy, which verifies the effectiveness of the improved strategy.

A three-dimensional two-phase non-isothermal serpentine flow field PEMFC model is established by COMSOL to study the effect of rough contact interface on the performance of serpentine flow field PEMFC. The influence of contact thermal resistance on polarization curve, temperature distribution and saturation distribution of liquid water is analyzed with the contact thermal resistance among all components being coupled into the three-dimensional model. The results show that with contact thermal resistance has a significant effect on the temperature distribution of PEMFC,which improves the gas transmission and drainage capacity and provides better performance when compared with that of PEMFC without contact thermal resistance.

Aiming at the citrus picking manipulator, the overall scheme is proposed, the mechanism is designed, and the statics model is established. Use CATIA software to carry out detailed design and 3D modeling, then carry out kinematics analysis on the manipulator, verify the manipulator workspace, and finally check the strength of the key stressed parts of the manipulator to ensure that the manipulator can meet the design requirements.

In order to solve the problem of old clothing recycling device, which causes inconvenient storage and bacterial growth in the recycling process. This paper designs a recycling device that can compress and pack old clothes. According to the shortcomings of the traditional old clothing recycling device, the overall device design is given, and the theoretical calculation and simulation analysis of the components to achieve the compression packing, the device can realize the compression packing of clothing and drying sterilization function.