Advanced Control for Applications最新文献

This research introduces an advanced algorithm based on convolutional neural networks for the detection and categorization of surface defects in manufacturing processes. At its core, the algorithm employs a deep learning model that integrates residual networks and attention mechanisms to effectively extract features. Additionally, we have developed a novel feature selection method, named NR, which synergistically combines neighborhood component analysis and ReliefF techniques. This approach enables the selection of more representative deep features for subsequent analysis. For the classification task, we utilize the support vector machine technique, which demonstrates versatility in handling both binary and multi-class classification scenarios. The reliability and superiority of our algorithm are further validated through a comparative analysis using a dataset specifically tailored for this context. The results indicate that our approach outperforms existing algorithms in accurately identifying manufacturing defects.

The IoT-based monitoring system is used to collect, transmit, analyze, and monitor industrial data in real time, as well as to monitor and operate industrial field equipment. Reduced manufacturing costs, better managerial decision-making, and increased production efficiency are all very important. The main research results are as follows: (1) The current development of the IoT technology at home and abroad is analyzed, the overall design of the laboratory equipment monitoring system with a three-layer three-dimensional architecture is proposed, and its working mode and functional composition are discussed. (2) A set of PLC control system with S7-300 PLC as the core is designed by using the laboratory CS4000 process experiment device, which realizes the functional requirements of the laboratory equipment monitoring system on-site monitoring. The intelligent IoT gateway suitable for the system is selected as the data collection and aggregation device of the system, which fully demonstrates the great role of the IoT technology in the monitoring field. (3) A laboratory remote monitoring platform that can remotely monitor experimental data is designed and its software and hardware functions are tested. The results show that all indicators have excellent performance.

As an important component of modern logistics industry, aviation logistics parks address the problems of unreasonable layout and low efficiency in traditional planning and construction methods. The study first constructed a grid model for optimizing the layout of functional areas in the aviation logistics park, and selected the layout scheme with the highest comprehensive correlation as the optimal planning. At the same time, three important model parameters were determined: prediction of cargo throughput, size of functional areas in the park, and comprehensive correlation values between functional areas. Subsequently, in response to the optimization problem of functional area layout in aviation logistics parks, a cosine adaptive genetic algorithm based on adaptive reversal operation was introduced to solve the model. According to the findings, the research mention algorithm has an average utilization rate of up to 69.64% in the international freight zone, which is a 12.38% improvement over the conventional genetic algorithm. Additionally, it has an average domestic cargo area utilization rate of 67.93%, 11.24% greater than the genetic algorithm. This demonstrated that the functional area layout scheme of the aviation logistics park produced by examining the suggested algorithm is extremely viable and offers new ideas and techniques for the planning and designing of aviation logistics parks.

The emerging trend in sustainable industrial processes focuses on energy-efficient drives to enhance performance in both intermediate and continuous operating conditions. For decades, researchers have focused on permanent magnet synchronous machines (PMSM) for industrial applications in order to retain consistency in performance and efficiency under wide speed ranges. Common bus regenerative control is an energy-efficient method used in industrial systems to manage power flow between multiple devices on a shared DC bus. It captures and reuses excess energy generated during braking/deceleration, reducing waste and improving overall system efficiency. The experimental behavior of the drive scheme has been observed as per the standards of the laboratory. In this article, regenerative control of a PMSM drive is using the voltage vector control (VVC+) technique and the SCADA based condition monitoring for the PMSM drive system is integrated to supervise the parameters including current, voltage, power, speed, torque, and temperature under dynamic operating conditions. As a primary step of implementing the drive condition monitoring is carried out through Danfoss's motion control tool software. Further, the SCADA control system is interfaced to FC302 PMSM drive through RS485 Modbus communication to extract the necessary electrical attributes and monitor the same.

The electricity load prediction is closely related to production and daily life. The electricity load prediction is also a very important task. With the widespread application of smart grids, load data shows an exponential growth trend. The huge amount of data in the load makes power prediction even more difficult. On the basis of traditional prediction algorithms, a power load prediction model based on machine learning and neural networks is designed. Because the single model prediction has the unstable results, a combined model is obtained based on the ensemble learning idea and two single model prediction method. The prediction results are detected by the load data. From the experimental results, the mean absolute percentage error (MAPE) of the AdaBoost-GRU data fusion model is 0.066%. Compared to the AdaBoost-GRU data fusion model, the MAPE decreases by 1.59% and 1.12%, respectively. The relative mass scores of the two groups decrease by 132.57% and 89.14%, respectively. The prediction accuracy is improved, which has advantages compared to traditional combination models. It can effectively enhance the accuracy of short-term power grid load forecasting. It is an important scientific and practical reference for power grid decision-making.

Unmanned aerial vehicles (UAVs) have been widely used to transmit data to Internet of Things (IoT) devices in various industrial, civil and military applications because of their flexibility and mobility. Some emergency situations pose strict requirements for UAV mission completion time. Therefore, this paper investigates a UAV-supported data distribution network, where a UAV is dispatched to distribute data to a group of IoT devices. We propose a device attribution (DA) based cluster-by-cluster (CBC) communication strategy. The objective is to minimize UAV mission time while satisfying the required data amount of all devices. To this end, we propose a mission time optimization algorithm (MTOA), whose key lies in invoking DA mechanism to determine the device belonging in the coverage of overlapping clusters. Numerical results demonstrate that the proposed strategy can effectively reduce the mission time compared with the baseline ones, offering an innovative method for solving complex device attribution issues. Furthermore, the proposed strategy is expected to exhibit a significant potential in scenarios involving the high-density IoT device deployment.

The article presents the conceptualization, development, and implementation of a sophisticated mobile robotic system with the purpose of providing aid in the medical care of those afflicted with severe cases of COVID-19. The robotic system will engage in verbal communication with the patient and provide updates regarding the external environment. Additionally, it will facilitate the delivery of food, beverages, and other consumable items to the isolation room. The robot has the capability to navigate to its intended location using two distinct modes: autonomous mode and online control mode. The hardware is constructed within a mobile robot system that is interconnected to the Internet over a 4G mobile network. The system employs a client–server software architecture wherein the transmission of data between the client and the server occurs through various transport protocols. Experimental simulations were conducted in an active treatment room to evaluate the performance of autonomous operating mechanisms, including obstacle avoidance positioning, safe destination navigation, and feedback display for remote robot operation.

The next generation of low observable (LO) unmanned combat aerial vehicle (UCAV) with highly autonomy to implement a penetration mission requires advanced methods for flyable and safe route planning (i.e., respecting physical capability of vehicle and threat coverage by hostile air defense radars) at a real-time manner. Currently, the main challenge of real-time route planning for LO UCAV is to achieve computationally efficiency under dynamic (pop-up/moving) threats by air defense radars. In this paper, a real-time planning paradigm in compliance with complex penetration requirements is proposed, and a complete modeling of route planning for LO UCAV's penetration as an optimal control problem is designed. The paper at first devises a direct method to transform the optimal control problem into a nonlinear programming (NLP) problem and then solves the formulated NLP problem under a moving planning horizon. The proposed method can give computationally efficient route planning results for LO UCAV's penetration under multiple kinds of radar threats. Numerical test results based on F-16 uninhabited platform demonstrate the effectiveness of the proposed method.

In intelligent vehicles, road environment perception technology is a key component of autonomous driving assistance systems. This component is the foundation for vehicle decision-making and control, and is a guarantee of safety during the driving of the vehicle. The existing environment perception technology mainly targets well-lit environments and requires visible light imaging equipment. Therefore, in low visibility environments, this technology cannot make good judgments about the external environment. Many existing perception systems mainly rely on sensors. Under low visibility conditions, these sensors weaken their effectiveness due to signal transmission, reflection, or absorption, resulting in incomplete or distorted data collection. Reduced visibility often affects the sensing range of various sensors, hindering the system's ability to detect and recognize distant objects, thereby limiting the necessary advance warning and response time for safe navigation. In response to this issue, this study proposed a combined method of infrared imaging and polarized imaging to collect feature data on road conditions in low visibility environments. Then, the obtained images were denoised and enhanced. The processed images were input into the system for recognition, and the images were analyzed and recognized using a low visibility road situation semantic segmentation algorithm based on deep learning. The research outcomes denoted that the pixel accuracy, average pixel accuracy, and average intersection ratio of the variable weight combination model in polarized degree images were 91.2%, 89.1%, and 71.6%, respectively. Those in infrared images were 83.6%, 90.6%, and 62.1%, respectively. The various indicators of the variable weight combination model were higher than those of the U-shaped neural network model, indicating its performance is relatively excellent. The research results indicated that infrared imaging helps to acquire information at night or in low light conditions, while polarized imaging can provide better adaptation to cluttered light and reflections, enabling the system to provide more robust environmental sensing in complex weather conditions. It fills a critical gap in perception for autonomous driving systems in adverse weather conditions.