This paper solves the problem of difficulty in achieving satisfactory results with traditional methods of bearing fault diagnosis, which can effectively extract the fault information and improve the fault diagnosis accuracy. This paper proposes a novel artificial intelligence fault diagnosis method by integrating complementary ensemble empirical mode decomposition (CEEMD), energy entropy (EE), and probabilistic neural network (PNN) optimized by a sparrow search algorithm (SSA). The vibration signal of rolling bear was firstly decomposed by CEEMD into a set of intrinsic mode functions (IMFs) at different time scales. Then, the correlation coefficient was used as a selection criterion to determine the effective IMFs, and the signal features were extracted by EE as the input of the diagnosis model to suppress the influence of the redundant information and maximize the retention of the original signal features. Afterwards, SSA was used to optimize the smoothing factor parameter of PNN to reduce the influence of human factors on the neural network and improve the performance of the fault diagnosis model. Finally, the proposed CEEMD-EE-SSA-PNN method was verified and evaluated by experiments. The experimental results indicate that the presented method can accurately identify different fault states of rolling bearings and achieve better classification performance of fault states compared with other methods.

Starting from the current problems facing Industry 4.0, this article analyzes the changes in the macro and industrial environment that Industry 4.0 faces and explains the problems, opportunities, and strategies for the manufacturing industry in the external environment. First, the reference system of the intelligent manufacturing system, the current status, and the existing problems of industrial production management are analyzed through the investigation of the status quo of industrial production and management. This puts forward the detailed requirements of the industrial intelligent manufacturing system in the data acquisition layer, data storage layer, and analysis and decision support layer and then designs the hierarchical structure of the industrial intelligent manufacturing system. Subsequently, it adopts design methods and lists product manufacturing costs, pointing out that Industry 4.0 requires industrial transformation, and finally proposes the strategic direction of smart manufacturing in combination with the Industry 4.0 network strategy. At the same time, in view of the problems of long parameter measurement time and untimely system feedback in the existing koji-making process, an online parameter measurement method based on network optimization is proposed. On the basis of the neural network, an industrial neural network with double hidden layers and self-feedback of the output layer is proposed. Through algorithm comparison experiments, the proposed parameter prediction model based on industrial neural network has better prediction results and higher accuracy. Finally, a comparison of cost, quality, delivery time, etc., before and after the implementation of Industry 4.0 intelligent manufacturing is carried out. An intelligent solution is proposed, the implementation goal is formulated, and the implementation is gradually implemented in stages, and finally an intelligent upgrade and transformation are realized. It is shown in many aspects that intelligent manufacturing provides a powerful means for enterprises to achieve agility, virtualization, lean, integration, and collaboration, and it can bring efficiency, reliability, and safety to the manufacturing process of enterprises.