Diqing Nonferrous Metals Co., Ltd. (hereinafter referred to as "Diqing Nonferrous") is located in the Shangri-La Tibetan area of Yunnan Province, in a national forest park at an altitude of more than 3,000 meters. It produces more than 50,000 tons of copper concentrate annually.
The production process of Diqing Nonferrous Metals employs an extremely complex and large-scale process monitoring and automatic control system. Each production piece of equipment is controlled by a programmable logic controller (PLC) provided by the equipment manufacturer, and then centrally monitored and controlled in a central control room via a distributed control system (DCS). During production, automation engineers face numerous challenges related to individual equipment maintenance and system optimization. The production equipment comes from different manufacturers, using PLCs such as Siemens S7-200SMART, S7-200, S7-1200, and S7-1500; GE's 90-30 and RX3i; AB's Micrologix 1500 and 1756 series; as well as Siemens PCS systems and Honeywell DCS system platforms.
Due to the above problems, hyper-converged infrastructure and virtualization technologies are applied to industrial control platforms. By integrating servers, networks, virtualization and other components into an easily managed integrated system through software, and by reducing manual operation through automated operation and maintenance, security is improved, human error is reduced, thereby reducing implementation and operation and maintenance risks and operating costs.
1. Background Analysis of the Current Status of Automation Control
Diqing Nonferrous Automation covers the entire production process of mining, beneficiation, and tailings, and adopts a redundant central processing unit (CPU) system for DCS system programming, configuration, testing, and system integration.
The control system mainly includes a rail-guided unmanned driving system, SABC grinding automatic control, copper-molybdenum mixed flotation automatic control, copper-molybdenum separation flotation automatic control, concentrate dewatering automatic control, and tailings conveying control system.
2. Industrial Control System Optimization Methods
2.1 Introduction to Hyperconverged Infrastructure Technology
Hyperconverged infrastructure (HCI) is inspired by the construction of large-scale data centers through software-defined technologies. It combines virtualization technology with enterprise IT scenarios to provide enterprises with scalable IT infrastructure. Currently, Lenovo's HCI appliance in China integrates server clusters using distributed storage and compute virtualization technologies to provide IT infrastructure with computing, storage, and networking resources, helping customers significantly reduce the complexity of data centers of various sizes.
A cloud data center based on hyperconverged infrastructure (HCI) refers to the extensive use of software-defined technologies to decouple computing, storage, networking, and dedicated hardware, achieving true convergence of IT infrastructure and removing the last obstacle to the implementation and deployment of cloud computing data centers. Providing cloud computing services using HCI is an inevitable trend. The principle behind HCI technology is essentially replacing the traditional SAN in a converged system with software-defined storage (SDS). Its main components are software-defined storage built on standard server hardware plus (server) virtualization.
2.2 Introduction to Virtualization Technology
Server virtualization features: Server virtualization divides a physical server into multiple smaller virtual servers. With server virtualization, multiple servers can rely on a single physical machine. The most common server virtualization method is to use virtual machines, which make a virtual server behave like an independent computer. IT departments typically use server virtualization to support various tasks.
By abstracting physical server resources into logical resources, a single server can become several or even hundreds of isolated virtual servers. We are no longer bound by physical boundaries, but instead turn hardware such as CPU, memory, disk, and I/O into dynamically manageable "resource pools," thereby improving resource utilization, simplifying system management, achieving server consolidation, and making IT more adaptable to business changes—this is server virtualization.
2.3 Optimize Control Strategy
Virtualization allows different operating systems to run simultaneously on the same control host, such as Windows XP, Windows 7 (and UNIX and LINUX systems), enabling a multi-homed architecture that was previously impossible. In industrial control applications, this can meet the needs of various business systems.
The snapshot feature supports creating multiple snapshot copies per volume in a virtualized environment and distributing these point-in-time copies to other virtual machines. This allows users to create powerful test environments simply by creating snapshot copies of production data and allocating them to virtual machines dedicated to testing. Once a virtual server environment is set up, virtual test servers can be created quickly. Testers can not only obtain realistic "real-time" data within seconds but also take data snapshots throughout the testing process.
3 Industrial Applications
3.1 Control System Platform
The following section uses VMware vSphere platform setup as an example.
3.1.1 Building a server hardware resource pool
Configure a suitable server hardware resource pool according to actual usage needs. Depending on the usage requirements, you can choose multiple low-configuration HP MicroServer Gen8 or similar machines as servers, with each server running only 1-2 VMS (virtual machine instances). Alternatively, you can configure 2-4 large servers such as DELL R920 or HP ML350PGen9 (depending on the actual needs of the project), with each running 4-6 VMS. You can also build blade server platforms such as DELL C6320 or HP SL230S, as well as fault-tolerant server hardware platforms such as StratoS (USA), NEC (Japan), and Haide (China).
3.1.2 Setting up the VMware vSphere platform
Deploy the vSphere platform strictly according to the deployment manual and the vSphere version recommended by the hardware manufacturer. Compatibility, scalability, and fault tolerance should be fully considered.
3.1.3 Setting up an automated software environment within a virtualized environment
Virtualization environments must be deployed strictly according to the methods recommended by each automation system manufacturer. Software requiring network interface card (NIC) passthrough must be configured; for software not requiring NIC passthrough, NIC passthrough is recommended. To minimize compatibility risks, the newest version of the automation software should be used whenever possible.
3.1.4 Establishing an automated control network
Some PLC system networks can be consolidated onto a single switch, while others require a separate set of one or two switches. When there are many system vendors, the network can become quite disorganized, requiring extensive experience to manage. Systems that can be consolidated should be consolidated as much as possible; for systems that cannot be consolidated, they should be isolated as much as possible. It's advisable to deploy more complex DCS systems on one server and simpler PLC systems on another. For easier packet monitoring, high-performance, compatible, and advanced managed switches with port mirroring capabilities should be selected.
3.2 Application Effect
(1) Meets the needs of multiple business systems: Virtual operating platforms can be created between computer platforms and end users through virtualization software. Automation engineers can configure multiple virtual systems according to the different software compatibility of the PLC on site, and can flexibly switch between multiple operating systems.
(2) Convenient cloning mechanism: Once a virtual machine is cloned, the same virtual system, including all software and configuration, can be reinstalled on any host without considering hardware and software compatibility issues. For industrial control applications that require stability, the cloning mechanism can ensure sufficient backup copies.
4 Conclusion
In an era of rapidly evolving computer hardware, industrial control technology, due to its unique characteristics, often lags behind mainstream products in both software and hardware platforms, leading to incompatibility issues between the host computers of various industrial control processor manufacturers. A general-purpose platform built using hyper-converged infrastructure and virtualization technologies can support the vast majority of software and hardware environments, providing rapid restore and system snapshot functions, significantly improving the compatibility of industrial control platforms and the efficiency of engineers in troubleshooting.
