Computers in Industry最新文献

Design-by-Analogy (DbA) is a powerful approach to innovative conceptual design. Cross-domain analogies are the stimulated sources of inspiration for generating new concepts. Therefore, how to efficiently retrieve them becomes an essential issue to be solved. This paper proposes a tool based on a structure-mapping function model (SMFM) for analogy retrieval. Inspired by the structure-mapping theory, SMFM considers two in-depth features: functional relation and causal relation between functions, which are expressed by function and meta-function concepts, respectively. SMFM serves to capture design knowledge from instances (i.e., engineering or biological systems) to establish a case database. The SMFM ontology is constructed for linking knowledge representation and analogy retrieval. Its meta-function, function, and flow terms are used to index the modeled instances so as to establish an index database. Analogy retrieval is realized by the ontology-based query expansion, vector space model and weighted cosine similarity. By inputting function or meta-function queries, we retrieve the required cross-domain analogies, whose design knowledge is displayed by function models to inspire target functions or meta-functions realization. Finally, the tool is evaluated for its effectiveness in cross-domain analogies retrieval. An application example illustrates its practicality in aiding innovative design. Additionally, an experiment was conducted to test the tool. The results indicate that our tool can assist users (e.g., engineering designers) to more quickly generate new schemes to realize target meta-functions and the quality of schemes is higher.

Anomaly detection of gas turbines faces the significant challenges of data imbalance and inter-class overlap. In this paper, we develop a novel data augmentation method, namely deep attention synthetic minority over-sampling technique with the Encoder-Decoder (DA-SMOTE-ED), which serves as a key step in our hybrid re-sampling scheme. To reduce the risk of generating noise data, on one hand, the DA-SMOTE-ED leverages an Encoder-Decoder to learn a class-separable feature space to weaken the effect of inter-class overlap. On the other hand, an attention module is applied to assign proper interpolation factors to generate synthetic samples that stay off the aggregation area of normal samples. Moreover, synthetic samples are generated in the learnable feature space, mapped back to the original space, and merged with under-sampled samples to form the balanced dataset. Finally, the superiority of the developed method is validated through two case studies including the real monitoring data of gas turbines and the modified version of the commercial modular aero-propulsion system simulation (C-MAPPS) dataset. More specifically, its average balanced accuracy is 91.77 % on the gas turbine dataset, yielding 3.67 %, 6.4 %, and 5.56 % improvements compared to the SMOTE-ENN, TimeGAN, and AugmentTS, respectively.

Industrial anomaly detection (IAD) is crucial for automating industrial quality inspection. The diversity of the datasets is the foundation for developing comprehensive IAD algorithms. Existing IAD datasets focus on diversity of data categories, overlooking the diversity of domains within the same data category. In this paper, to bridge this gap, we propose the Aero-engine Blade Anomaly Detection (AeBAD) dataset, consisting of two sub-datasets: the single-blade dataset and the video anomaly detection dataset of blades. Compared to existing datasets, AeBAD has the following two characteristics: (1.) The target samples are not aligned and at different scales. (2.) There is a domain shift between the distribution of normal samples in the test set and the training set, where the domain shifts are mainly caused by the changes in illumination and view. Based on this dataset, we observe that current state-of-the-art (SOTA) IAD methods exhibit limitations when the domain of normal samples in the test set undergoes a shift. To address this issue, we propose a novel method called masked multi-scale reconstruction (MMR), which enhances the model’s capacity to deduce causality among patches in normal samples by a masked reconstruction task. MMR achieves superior performance compared to SOTA methods on the AeBAD dataset. Furthermore, MMR achieves competitive performance with SOTA methods to detect the anomalies of different types on the MVTec AD dataset. Code and dataset are available at https://github.com/zhangzilongc/MMR.

Ontologies can represent technological enablers for knowledge elicitation and management in different kinds of organizations, especially with the exponential growth of sources and types of data fostered by digital transformation. However, their adoption in business applications is still limited, with existing Ontology Engineering Methodologies (OEMs) lacking adequate support during knowledge elicitation, authoring and reuse phases. This paper introduces a novel agile ontology engineering methodology (AgiSCOnt) to support ontologists (especially novice ones) in ontology development workflow, fostering collaboration with domain experts in an iterative, flexible and customizable approach. AgiSCOnt combines macro-level instructions with micro-level guidance, leveraging existing techniques and a management framework to help novice ontologists throughout the whole ontology engineering process. The methodology is compared to existing OEMs and assessed with three other agile methodologies (UPONLite, SAMOD, and RapidOWL). The evaluation is conducted with a sample of novice ontologists in a learning environment on Industry 4.0 technologies. Both the development process with a methodology from a user perspective and the quality of the developed ontologies were considered in the evaluation. Preliminary results show that AgiSCOnt effectively supports authoring and reuse, with developed ontologies of good quality. It is perceived as clear and simple, while being flexible and adaptable enough, thus supporting knowledge management and sharing in industrial organizations through the documentation of the ontologies.

Deep learning-based methods for industrial fault detection and diagnostics (FDD) depend strictly on good quality and sufficient quantity of condition monitoring data. However, in real-world industrial settings, data collection is usually limited, leading to sparse and insufficient data to train a data-driven model. Therefore, this work proposes a new methodology to address this issue by leveraging multimodal data anddomain knowledge to develop a data-driven solution. Particularly for large, complex machinery, unimodal sensors may not fully capture the health state information. In such cases, multimodal data may provide complementary insights into the machine degradation. However, challenges mentioned above need to be addressed before these data can be useful. The multimodal learning method presented within the methodology can benefit from useful information from different data modalities and from domain expert knowledge, even when these data are of low volume. The performance of the proposed methodology is investigated through a real industrial case study involving energy production systems. The obtained results demonstrate the potential of the proposed methodology in augmenting the FDD accuracy and tackling the sparse data challenge.

In the last decade, damage detection approaches swiftly changed from advanced signal processing methods to machine learning and especially deep learning models, to accurately and non-intrusively estimate the state of the beam structures. But as the deep learning models reached their peak performances, also their limitations in applicability and vulnerabilities were observed. One of the most important reason for the lack of trustworthiness in operational conditions is the absence of intrinsic explainability of the deep learning system, due to the encoding of the knowledge in tensor values and without the inclusion of logical constraints. In this paper, we propose a neuro-symbolic model for the detection of damages in cantilever beams based on a novel cognitive architecture in which we join the processing power of convolutional networks with the interactive control offered by queries realized through the inclusion of real logic directly into the model. The hybrid discriminative model is introduced under the name Logic Convolutional Neural Regressor and it is tested on a dataset of values of the relative natural frequency shifts of cantilever beams derived from an original mathematical relation. While the obtained results preserve all the predictive capabilities of deep learning models, the usage of three distances as predicates for satisfiability, makes the system more trustworthy and scalable for practical applications. Extensive numerical and laboratory experiments were performed, and they all demonstrated the superiority of the hybrid approach, which can open a new path for solving the damage detection problem.

Industry 4.0 determined the emergence of technologies that enable data-driven production planning and control approaches. A digital model can be used to make decisions based on the current state of a manufacturing system, and its efficacy strictly depends on the capability to correctly represent the physical counterpart at any time. Automated model generation techniques such as process mining can significantly accelerate the development of digital twins for manufacturing systems. However, complex production environments are characterized by the convergence of different material and information flows. The corresponding data logs present multiple part identifiers, resulting in the wrong finding of the system structure with traditional process mining techniques. This paper describes the problem of discovering manufacturing systems with complex material flows, such as assembly lines. An algorithm is proposed for the proper digital model generation, aided by the new concept of object-centric process mining. The proposed approach has been applied successfully to two test cases and a real manufacturing system. The results show the applicability of the proposed technique to realistic settings.

To address the challenge of predicting mechanical properties due to the unavoidable and multi-characteristic nature of defects in additive manufacturing lattice structures, an improved ensemble learning prediction model is proposed. The objective is to predict the true value of the yield stress of the lattice structure by using the data obtained from finite element simulation. The prediction model is constructed using the diversity and randomness of defects in the lattice structure as the input features of the model and the yield stress as the output. In order to improve the prediction capability of the model for multi-defect features, the Boosting module is added to the stacking model. To further improve the data-defect fit capability, feature transformation and feature combination methods are used to increase the number of data features, which in turn enhances the generalization performance of the model. In addition, the model has the ability to analyze the effect of defect characteristics and distribution on stress. The experimental structure shows that the model proposed in this paper can predict the yield stress of defects in defective lattice structures with an $R^2$ of 0.844. The proposed model reduces the time required for preparation and the cost of testing while ensuring prediction accuracy and enabling small samples of simulation data to predict true values. The research idea of this paper provides a research basis for industrial inspection and evaluation of lattice structures used in additive manufacturing.

Cyber–physical systems (CPSs) constitute the backbone of critical infrastructures such as power grids or water distribution networks. Operating failures in these systems can cause serious risks for society. To avoid or minimize downtime, operators require real-time awareness about critical incidents. However, online event identification in CPSs is challenged by the complex interdependency of numerous physical and digital components, requiring to take cyber attacks and physical failures equally into account. The online event identification problem is further complicated through the lack of historical observations of critical but rare events, and the continuous evolution of cyber attack strategies. This work introduces and demonstrates CyPhERS, a Cyber-Physical Event Reasoning System. CyPhERS provides real-time information pertaining the occurrence, location, physical impact, and root cause of potentially critical events in CPSs, without the need for historical event observations. Key novelty of CyPhERS is the capability to generate informative and interpretable event signatures of known and unknown types of both cyber attacks and physical failures. The concept is evaluated and benchmarked on a demonstration case that comprises a multitude of attack and fault events targeting various components of a CPS. The results demonstrate that the event signatures provide relevant and inferable information on both known and unknown event types.

Predicting the lumber products that can be obtained from a log allows for better allocation of resources and improves operations planning. Although sawing simulators make it possible to anticipate the production associated with a log, they do not allow processing many logs quickly. It was shown that machine learning can be used in place of a simulator. However, prediction quality is still lacking and information rich log representations are seldomly used in the literature for machine learning purposes We compare several log representations that can be used (industry know-how-based features, 2D projections, and 3D point clouds) and several neural network architectures able to process these log representations (multilayer perceptron, residual network and PointNet). We also propose a new way to implement a loss function that improves prediction of sparse object count in regression. This new approach achieves a 15% improvement of F1 score compared to previous approaches.