As traditional analog industrial automation transitions toward digitalization, the efficient deployment of control systems becomes increasingly critical. Open programmable logic controllers (OpenPLC) and the Modbus communication protocol, leveraging the widespread adoption of containerization technologies, are gradually migrating to native cloud architectures. This paper investigates the deployment of OpenPLC in local environments, Docker containers, and Kubernetes clusters, evaluating several key performance indicators, including resource consumption, Modbus communication response time, throughput, multi-user handling, stability, and fault recovery. By designing experimental platforms tailored to different environments and conducting validation across multiple platforms, the experimental results indicate that, although local deployment performs better in terms of low latency and resource consumption, deployments on the Kubernetes platform exhibit significant advantages in fault recovery, throughput, and system stability. In contrast, Docker-based deployments demonstrate more balanced performance, but lag behind Kubernetes in scalability and fault tolerance. Based on the experimental findings, this study provides valuable insights for selecting the optimal deployment strategy for industrial automation systems, taking into account OpenPLC, containerization technologies, performance requirements, and the integration of cloud-fog automation for enhanced flexibility and scalability.

This article presents a Kubernetes-based database benchmarking framework for Cloud-Fog Automation (CFA) in industrial systems, integrating Locust (dynamic load simulation) and Prometheus (resource monitoring) to evaluate four databases under industrial workloads: write-heavy , query-intensive (real-time analytics), and 6:4 read-write hybrid (control-logic scenarios). Kubernetes, as a representative of cloud-native technologies, is a crucial support for industrial CPS, but the database performance for industrial data management is yet to be fully determined. In 1k-concurrent-user tests (emulating distributed CFA edge nodes), ReductStore delivered 649.63 req/s throughput with <6 ms latency, ideal for fog-level real-time control, while OpenGauss achieved P99 latency <1 ms (critical for PLC synchronization) at 4.69 Gi memory cost—quantifying trade-offs for resource-constrained fog deployments. The framework demonstrates Kubernetes’ role in elastic cloud-fog orchestration, aligning with CFA’s industrial demands: InfluxDB suits massive IIoT data aggregation, while OpenGauss optimizes mission-critical latency. Our results bridge cloud-native scalability with deterministic industrial performance, enabling cost-efficient DBMS selection for smart factories.

Virtual Programmable Logic Controllers (vPLCs), as a newborn technology, are becoming increasingly important in modern industrial automation due to their flexibility and scalability. There is lack of researches on data synchronization and redundancy mechanisms for vPLCs, limiting applications of vPLCs in critical industrial scenarios. This paper designs and implements a data synchronization and redundancy mechanism between vPLCs based on heartbeat detection to enhance the reliability of vPLC systems. The mechanism continuously monitors for failures and synchronizes data between vPLCs to ensure seamless control task takeover in the event of a failure. Experimental results demonstrate the mechanism’s high effectiveness in fault detection and recovery, achieving a redundancy switchover time that meets industrial application requirements.

With the rapid development of the industrial internet of things and automation control systems, supervisory control and data acquisition (SCADA) systems have been widely adopted in industrial manufacturing due to their flexibility and scalability. The cloud-fog automation (CFA) paradigm is emerging to address higher real-time and computing demands in complex industrial environments. To fully leverage the efficiency, flexibility, and scalability of Kubernetes, an open-source container orchestration platform Kubernetes in managing containerized applications, this article investigates methods for deploying SCADA systems on the Kubernetes platform. This approach aims to capitalize on Kubernetes’ benefits, such as automated deployment, elastic scaling, and high availability, to optimize resource management and enhance system performance. To validate the proposed solution, we employs testing tools such as wrk and tc, along with monitoring tools like Prometheus and Grafana, to conduct a comprehensive evaluation of Kubernetes’ advantages in various scenarios. We focus on three key aspects: reliability, scalability, and security. The results demonstrate that Kubernetes can significantly improve the scalability, fault recovery capabilities, and stability of SCADA systems.

In this article, we propose an event-triggeredH∞filtering algorithm based on hysteresis quantized measurements for accurate state estimation in networked mass-switching autonomous marine vehicles (AMVs). The proposed method addresses several critical challenges in networked AMV filtering, including limited communication bandwidth, constrained energy resources, and signal transmission delays. A dynamic model is first established to capture the parameter variations caused by mass changes in AMVs. To alleviate bandwidth limitations and reduce signal chattering, a hysteresis-based quantization scheme is introduced. Furthermore, an energy-efficient event-triggered mechanism is designed to be both energy-efficient and capable of preventing excessively long periods without triggering, thereby reducing the frequency of data transmission. To facilitate rigorous stability analysis, a novel augmented Lyapunov-Krasovskii functional is constructed, and Wirtinger-based inequalities are employed to handle the time-delay-dependent integral terms. Based on this, a co-design strategy is also developed to jointly solve both the filter and the event-triggering mechanism. Simulation results demonstrate that the proposed method achieves accurate state estimation for networked mass-switching AMVs while reducing the transmission rate by 75%, significantly outperforming conventional logarithmic quantizers in suppressing chattering.