"Software is eating the world, and all the functions realized by specific hardware in the past will now be realized by software." The functions of traditional closed industrial control systems realized by specific hardware will be replaced by software in the era of Industrial Internet or Industry 4.0.
content
1. The core PLC of the industrial control system. 3
2. Development Trend of Industrial Control System... 6
3. PLC virtualization and software definition... 11
zero, preface
The Internet is a subversive and aggressive ideology and system. In just over a decade since its birth, it has hit multiple industries across borders and achieved great success. To sum up, the core idea of the Internet is economies of scale, and economies of scale are formed, that is, in the initial stage, large-scale capital investment is made, products that meet basic needs are provided, and customers are developed on a large scale in order to achieve economies of scale. And when it reaches a certain scale, it starts to replicate this kind of scale economy in a diversified batch, forming a diversified scale development. Because the cost of storage and marketing of each new product added to the Internet can approach zero, this characteristic of the Internet determines that it must be naturally aggressive and subversive for various industries.
After the Internet gradually penetrated into the field of industrial control systems, marketing concepts and systems such as the Industrial Internet, Industry 4.0, and the Industrial Internet of Things were born. These concepts are buzzing, and the city is full of storms for a while. After the hustle and bustle, there may be silence, but the intrusion from the Internet is unstoppable.
The general trend of the world is mighty and mighty, those who follow it will prosper, and those who go against it will perish. As industrial control system practitioners, we should put aside the fog of marketing concepts and pursue the technological transformation of industrial control systems under the Internet concept. In such an era of change, the realization of the concepts and technologies advocated by Industry 4.0, smart factories, and industrial Internet will have to be reconfigured in a short period of time. Therefore, it is necessary to reconfigure the core equipment PLC of the industrial control system flexibly. How to flexibly configure PLC becomes the support behind the industrial Internet to truly realize industrial interconnection. Therefore, from this perspective, this article describes the core technologies behind the Industrial Internet in the era of the Industrial Internet.
1. The core PLC of the industrial control system
Programmable Logic Controllers, also known as Programmable Logic Controllers in English, or PLC for short (hereinafter referred to as PLC in this article) are small industrial computers with modular components designed to automate custom control processes. The control process I understand is the process of controlling physical equipment through a program, and this program is realized in the form of logical expression (ladder diagram or other PLC programming language). Inside the PLC, the real physical equipment is represented logically by a symbol or a string, so the program written is the process of programming and combining and controlling these logics sequentially. This control process is programmable and customizable. Hence the name Programmable Logic Controller (PLC).
PLC has been in the development, so far it has not been under the final definition. The International Electrotechnical Society (IEC) successively released the first, second and third drafts of the PLC standard draft in November 1982, January 1985 and February 1987. In the third draft, PLC is defined as follows: Programmable Logic Controller (PLC) is an electronic system for digital operations, designed for use in industrial environments. It uses programmable memory to store instructions for performing logic operations, sequence control, timing, counting and arithmetic operations, and controls various types of machinery through digital and analog inputs and outputs. or production process. Programmable logic controller (PLC) and its related peripheral equipment should be designed according to the principle of being easy to form a whole with the industrial control system and easy to expand its functions.
In the field of industrial control systems, the physical machines and production lines of an industrial production environment are usually controlled by hardware PLCs, which are also considered to be the most optimized solution today and have driven industrial automation processes for many years. To better understand the purpose of PLCs, let us look at a brief history of PLCs.
Industrial automation started before PLCs. In the early to mid-20th century, industrial automation was often accomplished using complex electromechanical relay circuits. Electromechanical relay is an electronic control device, which has a control system (also known as an input loop) and a controlled system (also known as an output loop). When the input quantity (such as voltage, current, temperature, etc.) reaches a specified value, the controlled An appliance whose output circuit is turned on or off. However, there are many problems with the amount of relays, wires and space required to create simple automation with this architecture. The implementation of a simple factory control process requires thousands of relays! It's even more catastrophic if something in the logic circuit needs to be changed.
In 1968, the first programmable logic controller (PLC) came out, replacing the industrial automation control realized by complex relay circuits in industrial production. The first to come up with a clear idea was General Motors. In 1968 they wanted a device that could replace the relay control. The following year, American Digital Equipment Corporation developed the first programmable controller PDP-14 for General Company, and the trial was successful, which was the first PLC in the world. By the late 1970s, PLC began to enter a stage of rapid development, with a rapid increase in operating speed and substantial progress in miniaturization. In the early 1980s, it was widely used in western countries and grew rapidly. That period was the golden age of PLC. After that, mainframes and ultraminiatures were developed. By the 21st century, the scale of PLC has been expanding, the number of I/O points has increased, multiple CPUs work in parallel, large-capacity storage, high-speed scanning, etc., modularization and standardization have become the mainstream, the cost has been greatly reduced, and the application has become more extensive.
PLCs are designed to be easily programmed by control engineers and technicians familiar with relay logic and control schematics. One of the earliest implementations was ladder logic, which was designed to simulate a control circuit schematic. The ladder diagram looks like a control circuit where electricity is energized by closing the contacts from left to right to energize the relay coil. As shown in Figure 1 below:
Figure 1 - Example of Ladder Logic
The ladder diagram above looks like a simple control circuit schematic, with input sources such as switches, buttons, sensors, etc. shown on the left, and output sources on the right. Realizing complex automation processes through intuitive interface programming such as ladder logic is more convenient and efficient than using previous relays, and the learning cost of transitioning to PLC is minimized. PLC is the product of the combination of microcomputer technology and traditional relay contact control technology. It overcomes the shortcomings of complex wiring, low reliability, high power consumption, poor versatility and flexibility of mechanical contacts in the relay contact control system. Make full use of the advantages of microprocessors, and take into account the skills and habits of on-site electrical operation and maintenance personnel, especially PLC programming, which does not require special computer programming language knowledge, but adopts a set of relay ladder diagrams as the basis. The simple instruction form makes the user programming image, intuitive, convenient and easy to learn; debugging and error checking are also very convenient. After purchasing the required PLC, the user can flexibly and conveniently apply the PLC to production practice only by doing a small amount of wiring and simple user programming according to the instructions in the manual.
For programming based on such a ladder diagram logic, it is only necessary to program the ladder diagram logic according to the production process of the on-site factory environment, thereby realizing the programmability of the control process. Although this programmable logic controller (PLC) is programmable, there are still some differences from the current software definition, mainly because the programmable logic controller (PLC) has certain limitations. That is to say, to program each programmable logic controller (PLC), specific programming software is required to implement the program, and then the final control process program upload is implemented by communicating with the programmable logic controller (PLC). The programmable logic controller (PLC) is between traditional hardware devices and software definition, realizing the programmability of the data plane, but the control plane is not separated to achieve unified centralized control.
PLCs were developed to deal with the complex machine control of electromechanical relays. The aim was to develop a more flexible control system that would reduce machine downtime and perform logical functions with this new device. From the development of PLC to the present, it has indeed achieved the purpose of initial design and development. PLCs have been quietly dedicated to industrial automation for decades, and they have achieved reliability in controlling machines even in safety-critical applications. So that almost all modern industrial automation controllers are realized by PLC. In the industrial environment, PLC is almost omnipotent.
2. Development Trend of Industrial Control System
The first industrial revolution occurred in the 18th and 19th centuries, and promoted social progress by creating new manufacturing processes that improved production processes. The manufacturing industry at that time mainly relied on the production of goods by hand, but the first industrial revolution born in the United Kingdom changed this situation, allowing the machine manufacturing industry to make better use of water and steam power to promote productivity. And these improved innovative ideas and systems have naturally played a great role in the second and third industrial revolutions. The ongoing industrial revolution is the Fourth Industrial Revolution, also known as Industry 4.0 (proposed in Germany) or the Industrial Internet (proposed in the United States). The basic concepts of Industry 4.0 are the same as other industrial revolutions: by improving business processes and manufacturing processes, reducing production time, reducing the cost of production materials, reducing the number of defective products manufactured, and making industrial manufacturing easier by creating machines that can replace human jobs .
Industry 4.0 or Industrial Internet is the term for the ongoing industrial revolution. It originally referred to the digitization of manufacturing, but it actually also refers to the digitization of other industries such as healthcare, logistics, and oil and gas. Also refers to the concepts we often hear about smart factories, smart cities or smart devices. Industry 4.0 is about the convergence of Internet of Things (IoT), Cyber-Physical Systems (CPS), Information Technology (IT) and Operational Technology (OT), where changes are first initiated from the field of information technology, cloud computing, machine learning and big data, etc. IT technology guides modern information enterprises to adopt new business models, improve their own business processes and operational efficiency, and enhance their core competitiveness. The development of these new IT technologies has solved a common demand of Internet companies and traditional companies, that is, to solve the challenges of continuous scale expansion and rapid business changes, while effectively controlling costs. Among traditional businesses, other types of businesses are willing to take risks by deploying new technologies at an early stage, while industrial businesses may be more cautious. Due to the particularity of the industrial environment, it is still unknown whether this appeal of industrial enterprises can learn from the success of Internet enterprises. In order to overcome this threshold, the industry needs innovation, so the rise of concepts and systems like Industry 4.0 is aimed at conducting a large amount of research, testing and implementing these technological changes to guide industrial enterprises.
Regarding recent developments in practice, we illustrate by analyzing the traditional pyramid model of automation. The traditional automation pyramid (Figure 2) represents a typical model in the field of industrial control systems today. All physical devices from sensors to actuators are at the field level, and the data and actions used to control these field-level physical devices are at the second level, which controls the physical hardware at the field level by using physical hardware such as PLCs. The third level is a data acquisition and monitoring level that allows users to monitor and control their industrial control processes through SCADA systems. SCADA is an acronym for Supervisory Control System for Data Acquisition and Control, and a typical SCADA architecture includes the first three levels of the traditional automation pyramid. MES and ERP systems are on top of SCADA architecture. MES stands for Manufacturing Execution System and it refers to a system that monitors manufacturing data in real time. The MES system can track the goods situation throughout the production process. Enterprise resource planning (ERP) systems provide the highest level of the automation pyramid. ERP systems manage real-time monitoring of core business processes such as production or product planning, material management, and financials.
Figure 2 - Traditional Automation Pyramid
With the arrival of Industry 4.0 and cyber-physical systems, the traditional industrial control system architecture based on this pyramid model is changing. First, the top-level ERP and MES are gradually interconnected and integrated, realizing the upper-level linkage of production data, and finally using the advantages of cloud computing, big data and even artificial intelligence in data storage and computing to deeply mine and process production data. And finally output optimized production data for improving production efficiency. The equipment and systems at the lower production execution layer are also in the stage of intelligence and reconstruction, such as the terminal of final production data and changes, including running shoes worn by people and smart production lines in smart factories, from sensors to actuators. All the physical equipment of the company has been developing in the direction of digitization and intelligence and has made achievements. Therefore, it can be found that in the era of Industrial Internet or Industry 40, the final system change of the industrial control system is to change the traditional pyramid model from both ends. To be precise, the Industrial Internet focuses on the technological transformation of upper-level production data, that is, the integration of industrial control systems with cloud computing, big data, artificial intelligence, etc. Data mining and analysis to optimize the production process. Therefore, in the implementation of the current industrial Internet architecture, the implementation of the system architecture is either the industrial Internet platform directly collects the production data on the PLC with the PLC equipment, or the industrial Internet platform collects the data in the real-time database of the industrial control system, or the development A data acquisition gateway is created. After all data is collected through the data acquisition gateway, the gateway uploads the data to the industrial Internet platform. Therefore, the Industrial Internet is essentially a cloud computing platform that aggregates all industrial control system data. Industry 4.0 focuses on the digitization and intelligence of all physical devices at the bottom, from sensors to actuators, and enables these terminal devices to directly upload this data. The uploaded platform may be MES, and the historical database may also be an industrial Internet platform. For most companies, the first step in the realization of Industry 4.0 is the vertical integration and digitization of all relevant subsystems of the production system through MES systems to achieve real-time transparency of factory operations. At the same time, horizontal integration also includes the connection of functional areas. Here MES functions as the core element of the information carousel, collecting, analyzing, processing big data and exchanging data to support other systems.
Before the birth of automated control, the production of systems and machines had to be done manually. The advantage of automation is that the use of automatic control to realize the links that require repeated operations frees up human hands and achieves many advantages - from shortening time to market to reducing faulty products, it can well demonstrate the advantages of automatic control . Nevertheless, with the continuous growth of market demand, people still dislike the existing automation control can not meet the needs of production efficiency, and the essence is the same. People need more flexibility to ensure rapid product production and marketing, and flexibility is now the key and focus of industrial automation. These flexibility are manifested in: more and more factory data should be reusable, logic code should be easy to move and reusable, systems should be modular and extensible, and production companies should choose them according to their needs The preferred supplier instead of the current bundled sales and so on.
These implementations of the Industrial Internet and Industry 4.0 support the need for flexibility and scalability in future industrial control systems. The Industrial Internet enables large-scale centralized storage of our production data, realizes previously impossible big data, and utilizes the unprecedented computing power of cloud computing platforms to analyze these big data, mine and optimize production efficiency. Industry 4.0 has made field devices, machines and factories "smarter", so we can talk about smart devices, smart machines and smart factories. However, we will find that neither the Industrial Internet nor Industry 4.0 has made any further technological changes to the "brain" PLC of the industrial control system. This kind of phenomenon that the two ends are heavy and the middle is light is like the toll station on the expressway. The expansion of the expressway is far from being able to achieve greater vehicle throughput. The toll stations standing on the expressway are the bottleneck on this road. . Therefore, it is now necessary to set up more charging windows for toll stations, realize electronic toll collection and other measures to adapt to the rapid and rapid growth of traffic flow. The same applies to the field of industrial control systems. The core PLC equipment for control cannot be flexibly expanded. Undoubtedly, this limitation will greatly reduce the flexibility and scalability of industrial control systems.
Therefore, the industry is currently exploring technical systems such as Industrial Internet and Industry 4.0, and the focus will be on PLC.
Specifically, there are two main solutions to achieve PLC flexibility and scalability:
1. Realize PLC virtualization. Using PLC virtualization is virtual PLC (vPLC) to replace traditional hardware PLC;
2. PLC hardware reconstruction, realizing the next generation of new and intelligent PLC equipment, replacing traditional hardware PLC.
These two technologies are the current technological hotspots and research directions for the transformation of industry to the Internet. At least from the perspective of theoretical realization, the realization of PLC virtualization will maximize the protection of the existing interests of most existing manufacturers and their users, and Based on the successful cases of the information system, everyone is full of confidence in its realization. In addition, the PLC hardware reconstruction mainly takes the idea of software definition as the core, and separates the logic operation plane and logic control plane of the PLC hardware. The PLC hardware will realize the general logic operation, and the logic control and logic of the control plane will be unified by the controller. manage.
3. PLC virtualization and software definition
Virtualization and cloud computing have achieved great success in the ICT field, and their level of innovation has even subverted the entire old ICT architecture system, turning it over from the inside out. The effect is also quite remarkable. At least the operating cost of the current ICT environment has been significantly reduced by using virtualization and cloud computing technology. As for how much and how much, I believe this is a difficult figure. But at least businesses using virtualization and cloud computing have enjoyed the benefits, and current virtualization and cloud computing technologies have become the most advanced solutions in the office and corporate world. But it is not easy to deploy these technologies and solutions in industrial applications, because in industrial environments, the requirements are usually high, and system failures, real-time performance, etc. are critical to industrial production and applications. The Industrial Internet and Industry 4.0 used to focus on solving the transformation and technological realization of the application of advanced IT innovation technologies such as virtualization and cloud computing to the industrial field and improving industrial production efficiency. In the practice of Industrial Internet and Industry 4.0, whether Industrial Internet and Industry 4.0 technologies are so developed and reliable at present that they can be used in industrial control environments that need to meet high requirements such as stability and real-time computing , which is questionable. But judging from some published cases and statistics, manufacturers have begun to benefit from the Industrial Internet and Industry 4.0 technologies, and some of these benefits are mainly by using Industrial Internet or Industry 4.0 technologies, they can use real-time production data to help more Efficiently plan production processes to increase productivity and reduce operating costs. The success of these cases has encouraged people to actively invest in the research and development of the next technological solutions and improve the implementation of the current technology. At present, industrial control equipment manufacturers such as Siemens and virtualized cloud computing manufacturers in the IT field are focusing on trying to virtualize the control plane, using software instead of physical hardware to reduce operating costs and have a more flexible control environment. PLC virtualization or software-defined PLC, in other words, using virtual PLC (vPLC) or software-defined PLC is their next research and realization goal. At least from the perspective of the current technology implementation and research direction and the successful experience of IT, the general trend is like this.
The technical idea of PLC virtualization or software-defined PLC is to decouple traditional dedicated hardware functions. Because in the era of Industrial Internet or Industry 4.0, data integration of machines with higher-level applications on the factory floor is mainly done using traditional protocols that lack support for flexible integration of new equipment. There is always a contradiction between the two, how to solve this contradiction? Only an intermediate layer can be introduced between the two, where more flexible reconfiguration can be achieved by decoupling the control logic from the machine by virtualizing the PLC controller or software-defined PLC.
PLC virtualization or software-defined PLC enables easy scalability and system modularity by allowing users to replace or add components without affecting the rest of the system. PLC Virtualization or Software Defined PLC is designed as an open platform that allows users to choose preferred components and solutions, which means users have the flexibility to choose different suppliers (no vendor lock-in). In an Industry 3.0 system, using components from multiple vendors in the same architecture is not so easy, or simply impossible. There are usually no hardware dependencies in PLC virtualization or software-defined PLCs, so it is easy to migrate and reuse software. PLC virtualization or software-defined PLC uses virtualization or software-defined technology, using more software instead of hardware, and because less hardware is required, it reduces cost and footprint. The goal is to provide off-the-shelf and saleable COTS (COTS = commercial off-the-shelf) software/hardware products for flexible choice and scalability of the overall solution. Virtualization technology and software-defined architecture are software-centric models that have advantages in centralized system management, network processing, and security. Remote monitoring reduces operating costs, and maintenance engineers or operators do not always need to be on-site to check the status of machines. Centralized management simplifies remote monitoring as only one software platform is required to manage your assets. Through cloud computing and the use of smart sensors (including sensors with communication capabilities and on-board diagnostics), machine data is pushed to the cloud, where the data can be accessed through a user interface (HMI). Machine data can be used for predictive maintenance, which means that machine data can be used to estimate when a machine is about to fail.
PLC virtualization here does not refer to software PLC (SoftPLC) or the software PLC installed in the virtual machine, but refers to the decoupling of the PLC execution environment from the I/O modules, and the standardization, modularization and virtualization of the PLC execution environment. accomplish. The software definition abstracts the logic of the PLC, uses the realization idea of the software-defined network, and realizes the program development and management of the PLC execution action through an application store, so that the application program defines the function of the hardware PLC. That is, the logic control, program storage and IO module of PLC are separated, and the application program is used to realize the logic control part.
Industrial Internet or Industry 4.0 uses the following technologies when implementing PLC virtualization or software-defined PLC:
l Hypervisor or container-based virtualization
l Software Defined Networking (SDN)
l Network Function Virtualization (NFV), etc.
The typical architecture and implementation of PLC virtualization in academia and industry The typical architecture and description are:
Figure 3 PLC virtualization architecture
In PLC virtualization, where the PLC I/O bus is replaced by high-speed network functions, SDN allows flexible virtual channels to be created on the I/O fabric to accommodate the flow of connections between vPLC instances and I/O modules, such as sensor interfaces Or run the controller, and SDN-based control can provide flexible business isolation. Furthermore, due to recent advances in field programmable gate array (FPGA) and application specific integrated circuit (ASIC) technologies, such I/O modules can be constructed with lower complexity using FPGA or ASIC modules. In this architecture, SDN reconfiguration is managed by the SDN controller via a high availability (HA) server (not shown in the figure) connected to its northbound. The HA server continuously monitors SDN switch statistics and path reachability, triggering the reconfiguration process in case of performance degradation or failure.
A major networking consideration in the implementation of this PLC virtualization architecture is that this decentralized model has similarities to remote or distributed I/O PLC topologies, where network I/O modules act as extensions to the PLC rack. Advances in cut-through switching and Remote Direct Memory Access (RDMA), especially in the case of converged Ethernet, have allowed port-to-port delays in 10G Ethernet switching fabrics to shrink to a few hundredths of nanoseconds and applications Latency shrinks down to microseconds. In addition, resources such as Intel's Data Plane Development Kit (DPDK), Cisco's VPP, etc. allow the implementation of low-latency, high-throughput packet processing mechanisms that bypass the kernel, bring the networking stack into user space, and enable adapters to perform direct Memory access operations to application memory. This makes it possible to meet the requirements of jitter and flicker in transfers in microseconds, allowing bare-metal performance on commodity server hardware.
The main computational factors to consider in the realization of this PLC virtualization architecture are, first of all, due to the adoption of technologies such as hardware-assisted virtualization, the server can make use of nearly raw performance, low-latency I/O mechanisms or ISA extensions suitable for digital signal processing tasks. Increased availability, modern x86 or ARM processors have been able to replace microcontrollers in stand-alone PLC applications. Second, the availability of real-time static partition hypervisors such as Jailhouse (Siemens), Xtratum, PikeOS, etc. supports hosting RTOS guest VMs for real-time workloads. That is, in terms of computing, PLC virtualization mainly considers real-time, security and performance requirements. For PLC virtualization, its running real-time static partition management program, considering these characteristics of the industrial control system, its virtualization management program design needs to consider the mixture of several modes at the same time:
1. Full virtualization mode, in which the operating system runs completely unmodified in a secure partition. In this mode, the hypervisor ensures that the operating system running within it does not in any way harm or affect other operating systems running in parallel, and that the operating system running on it can run on the hypervisor without any modification . However, this comes at the expense of a slight performance penalty. This mode mainly runs industrial applications or enterprise applications that are not very real-time.
2. Para-virtualization mode, in order to ensure the hard real-time performance and stability of real-time code or real-time operating system, the hypervisor also has a deployment mode called "privileged mode or para-virtualization". In privileged or paravirtualized mode, the operating system retains full hardware access and uses the paravirtualized interface provided by the hypervisor. This allows the operating system to run at native speed without any latency added by the hypervisor.
3. That is, in the industrial control system environment, the virtualization hypervisor designed must be a hypervisor with para-virtualization characteristics. Therefore, in the era of the Internet of Things or the Industrial Internet, Xen's inherent advantages have begun to lead and Other hypervisors.
After reviewing the design of the current industrial control system virtualization hypervisors of foreign manufacturers, most of the typical cases use a mixture of full virtualization mode and paravirtualization mode, some of which run monitoring-level functions, and others run hardware Real-time control function. Paravirtualized or privileged mode does not introduce any latency, thus making it suitable for real-time applications. Communication between operating systems is achieved through virtual networks or SDN networks, shared memory.
The PLC of the software-defined architecture emphasizes a system and an implementation idea. That is, like the software-defined network and the software-defined world, the intelligence and standardization of PLC equipment is a typical software-defined PLC, including the easy connection of the PLC to the Internet; APP and analysis results are embedded in the machine and cloud to realize intelligence and self. Awareness; change and upgrade PLC equipment functions without replacing PLC hardware, provide users with intelligence and achieve continuous improvement; expand industrial Internet platform applications through APIs and ecosystems.
A typical implementation architecture is to first have an industrial machine that can be used to test the entire production process. This machine can be seen as a set of inputs and outputs that can be controlled via an OT control protocol. This is an ideal real-time protocol because it ensures that messages arrive within a certain time window. A fog computing or edge computing layer is then developed to communicate with machines at runtime via industrial control protocols. Its runtime will send the read data from the machine to the virtual PLC, and then return the output of the PLC to the machine. As shown in Figure 4 below:
Figure 4 PLC implementation under software-defined architecture
According to the technical solution tested by this technical framework abroad,
Figure 5 Technical solution
The technical solution chooses to combine Raspberry Pi with UniPi expansion board to simulate industrial machines. UniPi expansion board provides digital input and relay control for Raspberry Pi. Using CODESYS control software, these inputs and outputs are mapped to Modbus registers and finally passed by the upper layer. Industrial control protocol Modbus to realize industrial control operations. Then at the edge layer, or fog computing layer, act as a logic controller using OpenPLC, a standardized software PLC that can run structured text (ST) programs. OpenPLC contains a web server through which we can upload our PLC program to the PLC to run. The communication between the simulated industrial machine and the PLC is done via Modbus. Finally, the PLC needs to be connected to the industrial cloud. This step is realized through the Node-RED tool to combine with OpenPLC. Node-RED is a traffic-based IoT tool. It can connect different devices, APIs and other services. In the edge layer or fog computing layer, the PLC and the OPC UA protocol server are connected to realize communication through Node-RED, and the OPC UA client is installed in the cloud to realize the communication between the edge layer or the fog computing layer and the cloud.
The edge layer or fog computing layer uses Node-RED as the runtime and OpenPLC as the virtual PLC. Modbus messages can be sent and received using external packets. The machine's input is read in through a Modbus node and sent to the virtual PLC through another Modbus node. The virtual PLC processes this data and writes the results to its internal Modbus registers. Node-RED then polls the output of the PLC and sends the results back to the machine. With the help of an OPC UA node, data can be sent to an OPC UA server, or its own server can be hosted during runtime. All data can then be consulted in a uniform way by the IT system or the cloud via this OPC UA protocol.
This is just one of the simplest examples based on framework implementation, and it is also an example that is more understandable in the final implementation. After PLC virtualization and software definition, the decoupling of the lower-level IO will realize the maximum flexibility and scalability of the industrial control system. In addition, there are many benefits. In the diagram to achieve this, the existing industrial cloud platforms are all running in a mode compatible with the traditional PLC architecture. For example, the MindSphere industrial cloud platform of Siemens, which adopts a cloud-based open IoT architecture, can transmit the industrial field device data collected by sensors, controllers and various information systems to the cloud in real time through a secure channel. Provide enterprises with big data analysis and mining, industrial APP development, and intelligent application value-added services in the cloud. Its architecture is shown in the following figure:
Figure 6 MindSphere Industrial Cloud Platform Architecture
The MindSphere platform includes three layers: the edge connection layer, the development and operation layer, and the application service layer. It mainly includes three core elements: MindConnect, MindClound, and MindApps. Among them, MindConnect is responsible for transferring data to the cloud platform, MindClound provides users with data analysis, application development environment and application development tools, and MindApps provides users with integrated industry experience and data analysis results. industrial intelligence applications.
In the MindConnect layer, currently the main compatibility is to collect data from the on-site PLC or historical database, directly connect to the PLC or historical database for data acquisition, or perform data acquisition through a data acquisition gateway, all of which are compatible with traditional architectures.
This model or technical solution can be called the first stage of the Industrial Internet. To realize the blueprint built by the Industrial Internet or Industry 4.0 in the true sense, it is not enough to collect data for big data analysis, prediction, optimization, etc. We also need the device itself to have intelligent computing and intelligent processing capabilities. For the terminal, these optimization and computing capabilities may be satisfied by the computing capabilities of the smart device itself, but in the real industrial production process, the final control process also needs PLC to control, the intelligent computing, intelligent processing of PLC It must have the ability of software definition. Only through software-defined PLC, can the PLC control program be optimized and adjusted according to the application function after data analysis, prediction and optimization of the upper layer, so as to improve the efficiency of industrial production and avoid the need for personnel. Cumbersome debugging, etc., increase operating costs. However, the existing cloud platform is still unable to achieve the optimal definition of the bottom layer. Therefore, only after the key core control equipment PLC can be opened, the second stage of the Industrial Internet can be opened.
The essence of Industrial Internet, Industry 4.0, and Made in China 2025 is that the Internet is deeply involved in industrial production, thereby raising productivity to a whole new level. The past 10 years have been the 10 years of the consumer Internet. In the last wave of the Internet, the Internet connected billions of people around the world, and also created Internet giants such as Google and Amazon. The climax of the Internet in the circulation, consumption, retail, and communication industries has been obvious to all, and the Internet is in the field of industrial production. According to estimates by the communications giant Cisco, by 2020, the Internet will create 15 billion to 50 billion connected devices, including the connection between people and things and things, which is several times or more than the number of people connected in the traditional Internet era.
The future has come, may the day live forever.