Cloud task scheduling and resource allocation (TSRA) constitute a core issue in cloud computing. Batch submission is a common user task deployment mode in cloud computing systems. In this mode, it has been a challenge for cloud systems to balance the quality of user service and the revenue of cloud service provider (CSP). To this end, with multi-objective optimization (MOO) of minimizing task latency and energy consumption, we propose a cloud TSRA framework based on deep learning (DL). The system solves the TSRA problems of multiple task queues and virtual machine (VM) clusters by uniting multiple deep neural networks (DNNs) as task scheduler of cloud system. The DNNs are divided into exploration part and exploitation part. At each scheduling time step, the model saves the best outputs of all scheduling policies from each DNN to the experienced sample memory pool (SMP), and periodically selects random training samples from SMP to train each DNN of exploitation part. We designed a united deep learning (UDL) algorithm based on this framework. Experimental results show that the UDL algorithm can effectively solve the MOO problem of TSRA for cloud tasks, and performs better than benchmark algorithms such as heterogeneous distributed deep learning (HDDL) in terms of task scheduling performance.

This study aimed to develop a 5G + "mixed computing" + deep learning-based next-generation intelligent health-monitoring platform for an ethylene cracking furnace tube based on 5G communication technology, with the goal of improving the health management level of the key component of ethylene production, that is, the cracking furnace tube, and focusing on the key common technical difficulties of ethylene production of tube outer-surface temperature sensing and tube slagging diagnosis. It also integrated the edge-fog-cloud "mixed computing" technology and deep learning technology in artificial intelligence, which had a higher degree in the research and development of automation and intelligence, and was more versatile in an industrial environment. The platform included a 5G-based tube intelligent temperature-measuring device, a 5G-based intelligent peep door gearing, a 5G-based edge-fog-cloud collaboration mechanism, and a mixed deep learning-related application. The platform enhanced the automation and intelligence of the enterprise, which could not only promote the quality and efficiency of the enterprise but also protect the safe operation of the cracking furnace device and lead the technological progress and transformation and upgrading of the industry through the application.

In order to achieve a good generalization ability across different prediction tasks and ensure the reliability of prediction results, this paper proposes a GAC-BiLSTM-AM prediction method based on ensemble learning. This method employs a dual-channel convolutional neural network to comprehensively extract and integrate sample features, which are then input into a composite analysis module consisting of a bidirectional long short-term memory, an attention mechanism, and a fully connected layer to achieve in-depth analysis and processing of information, jointly constructing an efficient base learner. Meanwhile, during the parameter iteration tuning process of the base learner, a mistake correction mechanism is introduced to dynamically adjust sample weights in order to correct potential attention biases, and further combined with the gold rush optimizer to optimize the key structural parameters of the model. The application evaluation in multiple benchmark prediction tasks confirms that the GAC-BiLSTM-AM method possesses excellent generalization performance.

Offshore wind turbines face harsh conditions and high maintenance costs, necessitating precise fault classiffcation based on deep learning. However, gearbox variability, scarce fault signals, and poor interpretability challenge this task. Therefore, this paper proposes a multimodal time-frequency enhancement synergistic network (TF-MESN). First, a Time-Frequency Graphic Block Compression Enhancement Method (TF-BCEM) is introduced, to highlight local feature structures and macroscopic time-frequency domain features of signals. Second, a Cascaded Heterogeneous Convolution Module (CHCM) is proposed, which uses heterogeneous convolution kernels through cascaded residual structures of standard convolution and separable convolution to build a convolutional neural network, capable of capturing signal features comprehensively. Third, an Improved Multiscale Statistics and Statistical Dynamic SCRM Module (MSDSM) is proposed, which can reasonably segment effective signals from redundant ones, making the model more suitable for complex scenarios where steady-state and transient faults coexist. Finally, multi-channel feature fusion information is input into a Multi-scale convolutional weighted classiffer (MSWC), which utilizes stride-based multi-scale down sampling convolution and multiple statistics for fault classiffcation adapted to complex information representation. Experiments on two gearbox datasets, validated via visualization, show that TFMESN outperforms four representative methods, with superior generalization and small-sample fault identiffcation capability, meeting offshore wind gearbox fault diagnosis needs.

With the development and popularity of cloud computing, various artificial intelligence algorithms have been applied more and more widely in the field of cloud computing. While the complexity of artificial intelligence technology itself is also increasing, modern people are not satisfied with directly using artificial intelligence algorithm to solve complex practical problems, more and more researchers turn their eyes to explore the theoretical basis behind artificial intelligence algorithm, and try to explain artificial intelligence algorithm from various angles. This paper selects the latest application mode in the current field of cloud computing, and helps relevant researchers to clarify the development context of artificial intelligence technology in this typical mode by sorting out the latest artificial intelligence algorithm, so as to further explain the purpose of artificial intelligence technology.