Diagnosing complex air compressor systems with traditional data-driven deep learning models often results in isolated fault diagnosis, ignoring correlations between concurrent faults. This paper introduces a knowledge graph construction approach for the air compressor fault diagnosis field, using after-sales business data as the source. We propose a model based on Robustly Optimized Bidirectional Encoder Representations from Transformers (RoBERTa), specifically tailored for constructing a knowledge graph for air compressor fault diagnosis. By integrating Whole Word Masking (WWM) technology, Bidirectional Long Short-Term Memory (BiLSTM), and Conditional Random Fields (CRFs), our approach effectively extracts specific entities from unstructured data. On our dataset, the model achieved an average accuracy of 0.7962 and an F1 score of 0.7956, demonstrating notable improvements in both accuracy and recall for entity recognition tasks. The extracted entities were subsequently stored in a Neo4j graph database, facilitating the construction of a domain-specific knowledge graph for air compressor fault diagnosis.

The digital twin workshop is the basic unit for realizing smart manufacturing. Its model is the prerequisite for applying digital twins in workshops. Geometric model assembly is the primary concern in workshop-level digital twin modeling. However, current research lacks sufficient focus on the efficiency and application modes of equipment mesh model assembly features in geometric model assembly, hindering the construction of the workshop-level digital twin model. The equipment-level digital twin model typically only contains assembly information between individual parts and is generally represented by polygonal meshes; the lack of assembly information between equipment models impedes the assembly of the workshop-level digital twin model. To improve the efficiency of workshop-level digital twin modeling and the reusability of the original 3D assembly information, this paper proposes an assembly feature construction method of equipment mesh model for digital twin workshops. A 3D assembly information model is constructed to describe the information related to the equipment geometric model and the assembly features. Then, the assembly features of the equipment mesh model are gradually positioned preliminarily and precisely, achieving precise mapping of the assembly features in the 3D information model to the equipment mesh model and efficient construction of equipment mesh model assembly features. Finally, with the construction and application of equipment mesh model assembly features in an arc welding workshop and storage workshop as an example, the feasibility of the method in improving the modeling efficiency is verified through relevant comparative experiments.

Aiming at the problems that the measurement point planning in the contact-type on-machine measurement (OMM) process mainly relies on the knowledge and experience of the site engineers, and the measurement points’ information of the historical model is not sufficiently reused, a contact OMM points planning method based on similarity of 3-D process element is proposed. By analyzing the requirements of the integrated machining–measurement scenario of modular product parts and using processing features as the measurement unit based on the expression of model-based definition (MBD) information, several types of process factors that affect quality measurement are analyzed and quantitatively evaluated. Then, under the similarity evaluation standard of the constructed 3-D process element including multilevel information, relying on the developed computer-aided inspection planning (CAIP) system, the final OMM points planning scheme is generated through several reuse ways of historical model measurement point information. Finally, the feasibility of this scheme is verified by the OMM experiment of the A-plate of an injection mold for making the stud terminal and the related comparative experiments.

Air compressor valves are prone to mechanical wear and elastic fatigue during long-term operation, often leading to poor sealing or leakage. Such leakage is a typical failure mode of reciprocating compressors, introducing strong nonlinearities into cylinder pressure dynamics and significantly increasing the difficulty of state monitoring. To address this issue, this paper presents an adaptive prediction method for compressor cylinder pressure dynamics under valve leakage failure, based on a physics-guided Variational Autoencoder Convolutional Neural Network State Space Model (VAE-CNN-SSM). In this framework, a VAE with embedded physical information is employed to construct the state equation and generate latent variables reflecting valve motion degradation, while a CNN-based observation equation is established to map latent states to cylinder pressure. This hybrid modeling strategy enables accurate prediction of cylinder pressure dynamics and effective characterization of valve degradation behaviors under valve leakage failure conditions. Comparative experiments against conventional models demonstrate that the proposed method achieves superior predictive accuracy, robustness, and generalization. These findings provide a new approach for analyzing valve leakage failures and offer technical support for condition monitoring, health management, and predictive maintenance of reciprocating compressors.