Journal of Industrial Information Integration最新文献

Redundant data wastes cloud storage space, especially the multimedia data which comprises a large number of similar files and accounts for the majority of cloud storage. To protect privacy and eliminate redundancy in the cloud, fuzzy deduplication for encrypted multimedia data is practical and feasible. Unfortunately, existing fuzzy deduplications depend on aided server to be against security threats. In this paper, we propose a Practical Fuzzy Deduplication (PFDup) algorithm for encrypted multimedia data and it is secure against brute-force guessing attacks without additional independent severs. With our secure fuzzy deduplication technology, cloud storage can be significantly optimized by using Perceptual Hash (phash) to eliminate large quantities of identical even the similar multimedia data in a secure manner. In addition, PFDup protocol supports label consistency and a non-interactive Proof of Ownership (PO) in order to prevent the server–client collusion attacks. We conduct a series of experiments on numerous real-world datasets and the simulation results show that our deduplication rate for the similar images is over 91.5%.

One of the key aims of Industry 4.0 is to create more responsive systems. Responsiveness enables coping with new market requirements or introducing new products, as demonstrated by the COVID-19 challenges. However, there are currently no effective methods for measuring the responsiveness or reconfigurability of a system, or for quantifying the effort required to adapt it from one state to another. Adapting a production cell from its current state to a new adapted state requires a significant amount of information about dismantling, reintegrating, and handling physical equipment, as well as updating the software controller. Practitioners often only consider adaptation options for simple process parametrization or at the end of a system's life cycle, overlooking many potential adaptation opportunities. This paper proposes an evolvable network graph approach for supporting reconfiguration decisions by estimating the effort required to adapt the physical structure. Two complexity indexes have been developed to quantify the adaptation activities. An estimation algorithm infers the effort from the difference in the adaptation graphs that represent alternative options. The approach is illustrated in a laboratory-scale cell and applied in two industrial-sized cells, quantifying adaptation times of approximately 58, 7, and 122 h, respectively. This is equivalent to £3129.6, £356.04, and £6118.8, utilizing average hourly rates for system integrators and equipment handlers. The results show that the approach can effectively quantify the adaptation effort for different equipment sizes and connections, estimating the adaptation cost and time from the graph change quickly at around a millisecond and with minimal computational resources.

Hit rate is an important quantitative criterion for the process product quality prediction of the integrated industrial processes. The hit rate indicates the percentage of product quantities accepted by the downstream process within the controlled range of the product quality. The optimization model of the hit rate criterion is a non-convex intractable problem. In order to improve the hit rate of the predicted product quality, we define a hit rate optimization problem, and propose a data-driven quasi-convex approach, which converts the original problem into a set of convex feasible problems and achieves the optimal hit rate. The proposed approach combines factorial hidden Markov models, multitask elastic net and quasi-convex optimization. In order to illustrate the advantages of the proposed approach, a Monte Carlo simulation experiment is designed to verify the convex optimization property. Another experiment is carried out on two actual steel production datasets for the temperature prediction in molten iron dispatch. The results confirm that the proposed approach not only exhibits superior performance with the controlled hit rate, but also improves the hit rate by at least 41.11 % and 31.01 %, respectively, compared with the classical models on two real datasets.

With the advancement of unmanned aerial vehicle technology, dynamic monitoring with drones has been widely adopted to enhance multi-source information awareness capabilities. The cooperative strategy among drones still poses a significant challenge. Redundant actions within the drone swarm system can lead to a noticeable decrease in awareness performance. In this work, an advanced cooperative multi-hive drone swarm system is developed, which integrates multiple drones for information awareness and multiple hives for battery replacement. The system response is modeled by a series of discrete system state-action sequence, which follows a parallel system state transition mode. A well-designed simulated annealing-based hybrid algorithm (SA-HA) is developed for system response optimization, of which the simulated annealing process is adopted to coordinate two heuristic operators. To avoid redundant actions, an asynchronous cooperation mechanism (ACM) is proposed to strengthen the collaboration among agents in staggered system time intervals. Computational results indicate that the involvement of ACM can extract more problem-specific knowledge, which makes SA-HA easier to get high-quality system state-action sequences. Through the system redundancy analysis, we found that properly configured drones and hives can achieve high-efficiency global dynamic multi-source information awareness. The proposed system can provide pivotal support for regional situation awareness and analysis.

This paper presents a two-echelon vehicle routing problem (2E-VRP) with vehicle synchronization at meeting points for the reverse logistic network to collect waste in the urban area. Low-capacity vehicles are utilized to perform collection only in the inner part of the city because of restricted access and limited infrastructure to be expanded. While, high-capacity vehicles are used to transform waste from the border of the inner city to the recycling centers located outside of city. Vehicles of the first and second echelons visit each other at meeting points, which do not have storage capacity mainly because of environmental and social consequences of the transshipment of waste. The timing aspect plays a critical role at meeting points to ensure the reasonable sequence of operations. So, temporal synchronization, a critical aspect of 2E-VRP, is considered in this paper, as well as minimization of waiting time of vehicles in meeting points to avoid long dwelling time. Moreover, because the previous study of 2E-VRP has not dealt with space limitation of meeting points, this paper presents a 2E-VRP that incorporates temporal synchronization, waiting time, and temporal capacity simultaneously for the first time to address a more practical model. To solve the mentioned problem a set of capable and latest metaheuristics are proposed as well as the hybrid ones to probe their efficiency and reasonably construct high quality routes. Moreover, the impact of the temporal capacity is investigated compared to temporal synchronization without considering temporal capacity.

With the development of Industry 4.0 and smart manufacturing, improving production automation, intelligence, and digitalization has become a research trend in the Wire Arc Additive Manufacturing (WAAM) field. This study introduces a digital shadow that aims to improve the adaptiveness and dimensionality of monitoring systems in WAAM. Three sensors are used in the digital shadow: a welding electric signal sensor, a camera, and a laser profilometer to collect welding current and voltage data, image data, and point cloud data. The collected multi-scaled data are time and spatially synchronized by sampling multiple points along the welding path. Three ML algorithms are used for decision-making: Multi-layer Perceptron (MLP) classifier and YOLOv5 are used for time and spatial-scale detection, respectively, and a Variational Autoencoder (VAE) is used for the decision-level fusion. The system performance is then tested to detect defects and geometric errors in practical experiments and the results show that the overall F1 score is 0.791, including detecting, classifying, and analyzing the cause of defects. Additionally, the total predicting time is within 0.5 s, which is suitable for an in-process monitoring system.

The recognition and clustering of deformation modes are key to constructing impact deformation constraints for thin-walled structures. This paper transforms the clustering and recognition problem of structural impact deformation modes into a problem of clustering and recognition of point cloud sequences based on pseudo-labels. The effectiveness of the method is assessed, and the experimental results show that the accuracy of the proposed method can reach up to 92.17 % when using a pre-training deep neural network feature extractor, which is not only close to the 98.50 % accuracy of supervised learning classification models but also has a 16.84 % improvement in accuracy compared to the deep clustering method based on K-Means. Under different clustering conditions, the proposed method can effectively classify and recognise samples with similar deformation modes and has the ability to summarise and induce new deformation modes when the number of clusters exceeds the number of manual labels. Furthermore, this paper presents a multi-objective optimisation method for structural crashworthiness under impact deformation constraints based on the NSGA-II algorithm. This method constructs impact deformation constraints using surrogate models and deformation clustering and recognition models. The experimental results show that the proposed method can effectively constrain the generation of the population. It is found that there are a large number of Pareto solutions that do not belong to the expected impact deformation mode under the condition of no deformation mode constraint. In contrast, almost all the obtained Pareto solutions conform to the expected impact deformation mode under the condition of deformation mode constraint. In summary, under the condition of impact deformation constraint, the obtained Pareto solutions can satisfy the crashworthiness requirements while conforming to the expected impact deformation mode.

The swift growth of the wind power industry necessitates comprehensive evaluation and efficient fault prediction of wind turbines. Given the challenges of integration and optimization of reliability evaluation and fault prediction models, a systematic method of reliability fuzzy evaluation and fault prediction based on the Supervisory Control and Data Acquisition (SCADA) data is proposed. A mid-to-long-term reliability fuzzy evaluation model is constructed using Fuzzy Comprehensive Evaluation (FCE). The mid-term evaluation results in ten failure modes reveal that the model's hazard ranking results match the situation better than the RPN method. And the long-term evaluation results of 5 years in the operating mode show that the model effectively gathers the evaluation information each year and provides a clear and accurate reflection of reliability. Meanwhile, fault prediction is studied using alarm logs because they are better at expressing the status of wind turbines than monitoring data. And the tree-based algorithms and unsupervised statistical learning methods are used to mine the mapping relationship between input variables and predefined tags. The fault prediction achieves both accuracy and recall of 0.784 and saves over 163k Euros based on local wind turbine maintenance expenditures. Overall, the reliability evaluation and fault prediction complement each other, which may either affect future wind farm management or prevent unnecessary maintenance costs.

As manufacturers are adopting data-driven decisions and processes, manufacturing is becoming more vulnerable to digital threats, making data security a major challenge for Industry 4.0. More specifically, data manipulation is considered a serious threat to organizations with significant and damaging consequences. Adopting a cybersecurity framework is essential to protect organizations against these cyber threats. These frameworks are often generic, unopinionated, and only provide high-level guidance for mitigating cyber risk. Our objective is to find a framework that offers a solid foundation we can customize to support our needs for addressing data manipulation risk. This paper aims to review cybersecurity frameworks and identify the most customizable, yet opinionated, option.

OPC Unified Architecture (OPC UA) is widely considered a key enabler of “Industry 4.0” and one of the most promising standardized platforms for industrial communications from sensor to cloud. One of its key features is a powerful framework for information modeling that allows to compose semantic models and enables self-describing information provisioning. However, building OPC UA information models can be a tedious task, requiring deep understanding of both the OPC UA meta-model and the application domain to be modeled. Therefore, a wide range of methods for automatically generating OPC UA information models has been described in the literature, either from relational databases, from application-domain specific models, tools, or languages, or by aggregating multiple component-level models into a single, system-level information model. This paper reviews the state-of-the-art in tools and methods for automated generation of OPC UA information models. It is argued that enriching the tool landscape and interoperability, in particular with industrial engineering tools, will be a prerequisite for unleashing the full potential of OPC UA.