Without these three points, what else is there to talk about intelligent manufacturing?

2021-01-07 09:00:22

Times

In the "Three Stages of Intelligent Manufacturing" series of "On Intelligent Manufacturing", the author discussed the basic understanding of the three stages. How to implement intelligent manufacturing, we need to consider the three fulcrum of intelligent manufacturing: product, equipment and process.

Three fulcrum points of intelligent manufacturing

The first thing to consider is what is the goal of driving smart manufacturing. Obviously, the enterprise is after the product, not to make the enterprise how fashionable. When a company sells a product, it is not about how beautiful or modern the product line is, but about what the value of the product is. Product is the performance of the enterprise facing the society. The goal of intelligent manufacturing is the product, not the intelligent manufacturing itself. Therefore, the intellectualization of products is one of the primary issues that enterprises must consider. Intelligent manufacturing if you can not produce intelligent products, intelligent manufacturing has lost the meaning of The Times. Moreover, if the products of enterprises are not intelligent, the possibility of products and enterprises being eliminated in the future is great.

The second fulcrum is equipment. The equipment in every key link of the production process (including research and development, design) must be intelligent. If this intelligence cannot be realized, labor productivity and labor efficiency cannot be greatly improved, and enterprises may not be competitive. If it is not digital, networked and intelligent production equipment, it is not the advanced manufacturing equipment of this era. Moreover, if the equipment is not intelligent, it may not be able to produce the intelligent products that the enterprise wants to produce.

The third fulcrum is the intelligent problem of enterprise production process. Intelligent equipment is to solve the intelligent problem of "point" in the production process; Only by realizing the intellectualization of the whole process of production, can enterprises realize the intellectualization of the whole enterprise and maximize the benefits of intellectualization.

Smart products are the first fulcrum

A machine tool factory, if the production equipment and process are intelligent, and it produces the machine tool is general machine tool, no intelligent elements, then the future of the machine tool factory is very worrying. Because, he will not go to buy such not enough intelligent machine tools. Therefore, when any enterprise considers how to develop its intelligent manufacturing, it should first think of how to achieve intelligent products. Even if the production process is not fully or partially intelligent, intelligent products can be made, then enterprises should first consider the problem of intelligent products. The intellectualization of the product is realized by the computer system, especially the embedded system, which contains various complex degrees in the product. Embedded system can not only become the most important and representative technology of intelligent manufacturing, but also form a huge industrial chain. Embedded systems in China are developing at a slower pace -- though not too late. The embedded system used in the product, the vast majority of the chip requirements are not necessarily particularly high, generally is tens of nanometers to hundreds of nanometers, or even a little lower grade, may be enough. Therefore, the technical difficulty is not very great.

Product intelligentization is a new and important trend in the development of computing technology. The original purpose of the invention of computing technology is for scientific calculation. Then, it develops to support the information processing and dissemination of all kinds of human business activities, namely business computing. The coverage of business computing is already much larger than that of scientific computing. After the 1990s, with the development of the Internet, QQ, WeChat, Facebook and so on began to rise, and computing technology penetrated into people's social life, greatly promoting the development of social computing, and the application coverage of computing technology was further expanded. Now, computing technology has begun to penetrate into a variety of product fields to enhance the level of intelligence of products. Intelligent products tens of billions, even hundreds of billions, the coverage of the product calculation can be said to be "far away from the session", will certainly bring great changes to the entire IT industry. Therefore, the next hot spot in the application of computing technology is product computing. All products are going to be smarter to varying degrees, and computing is likely to play a part. This point, with the rapid development of industrial Internet demand has a great relationship.

The functional requirements of current intelligent products are not exactly the same as those of the so-called embedded system in the past. The main functions are reflected in three aspects. The first is sensing. The product needs to be able to sense changes in the external environment, or be able to integrate data within the product. The second is computing, which includes the operating system of the product itself, as well as the various applications that the product uses. For example, from data analytics to advanced computing -- that is, artificial intelligence. The third is networking. With the development of the global Internet of Things, the product may have the function of connecting fog computing, edge computing and cloud computing. Therefore, the new generation of intelligent products, with the previous concept of embedded system has been very different.

Intelligent equipment is the biggest difficulty

Equipment is the biggest difficulty in intelligent manufacturing. Production equipment is generally more complex, and the volume may be small, the industrial software used is often very complex. This makes the production cost very high, the market is small, so there are not many enterprises willing or able to engage in intelligent equipment manufacturing. Moreover, because the development cycle of equipment is long, the risk of enterprise operation is great. In addition, the difficulty of equipment manufacturing is largely in the soft equipment, that is, industrial software as the representative of soft equipment, including CAD/CAE such software tools. Without soft equipment, there would be no "digitalization, networking and intelligentization". Take away the software, and all the achievements of information cease to exist. Industrial software is first of all an industrial product, and often is a high-end industrial product. This is the main difficulty of Made in China 2025, and the industry is far from fully aware of this point.

Process intelligence

The manufacturing industry of developed countries has taken the lead in the intellectualization of production equipment. Japan and Germany, in particular, have essentially cornered the global market for equipment for major manufacturing industries. The next step of intelligent manufacturing is to realize the intelligence of the process and complete the development from the "point" of equipment to the "line" of the process. The most typical representative of process intelligence is exactly the goal of Industry 4.0 and Industrial Internet. Industry 4.0 proposes that the information system of enterprises should be integrated, including vertical integration and horizontal integration. Vertical integration is the Intranet of an enterprise mentioned in one of the series of Three Essays on Intelligent Manufacturing, while horizontal integration is the extranet of an enterprise. Now, to fully integrate the Intranet and extranet together, the data completely open up.

Integration of Intranet and Extranet

In addition, the integrated System should be built into a Cyber-Physical System (CPS). "Cyber" here refers to computers or computer networks. In a lot of modern enterprises, no matter Intranet or extranet, it is still just an independent computer network or system, or realized a preliminary integration. How to integrate with the physical entity of the enterprise and operate effectively is a university question. A 2006 report by the National Science Foundation (NSF) pointed out that current systems science, including systems engineering theory, developed in the industrial age, does not answer such questions well. They think that the physical entity and its embedded computer and network system how to concerted, efficient work truly, how to enhance the adaptability of such systems, autonomy, functionality, reliability, security, availability and efficiency, and will develop into a new system engineering, is the need to focus on the development of frontier proposition. In fact, there are a lot of research reports on CPS in the United States, which pay great attention to this topic.

The realization of intelligent process

The goal of Industry 4.0, or Industrial Internet, is not only to connect the Intranet and Extranet, but also to turn it into an intelligent physical system (CPS). Both can be represented by a "5C" (five layers) structure.

Industrial Internet and Industry 4.0 "5C (five tiers)" architecture

The bottom layer is the smart connectivity layer, the second layer is the data into information, and the third layer is the Cyber layer, which is the cloud computing data center of the enterprise. Here, it is necessary to compare the effective data obtained from the second layer processing with the expected value corresponding to the enterprise computer system. The fourth layer is the cognitive layer, according to the comparison of differences, find out where the problem is and how to solve the problem. Therefore, this layer is really a decision layer. The fifth layer is the configuration layer, can according to the decision requirements, through the computer network, to the person, to the thing, to the computer to reconfigure or change. Such a five-layer structure constitutes a standard feedback control system, which can carry out real-time feedback and control on the control objects of the enterprise, namely: people (employees), machines, computer systems, various physical entities, etc. For such a feedback system, the corresponding technical support at each level is shown in Fig. 5. It is the use of the most modern advanced technology, the industrial Internet to achieve the whole process of enterprise business activities of the intelligent control.

According to this idea, the system architecture design of Industry 4.0 and Industrial Internet was completed respectively in 2015. The reference architecture for the Industrial Internet clearly states the elements of the system and their relationships with each other, and provides an open "Industrial Internet System Design Guide". It should be emphasized that what we are talking about here is a guide, which is a direction for everyone to work together in the same direction, not a standard. This architecture design describes the internal and external three-tier structure of the industrial Internet system. From the edge layer, to the platform layer, to the enterprise layer, if we think of it as a sphere, the outside is the edge layer on the device side, and in the middle is the platform layer. Of course, there is now a trend to generalize the industrial Internet platform), the innermost layer is the enterprise layer. In the edge layer is mainly the gateway of the edge, collect a variety of data; After sending the data to the platform layer, the platform layer does the necessary processing and analysis of the data; After the analysis is done, it is sent to the enterprise layer and sent to the enterprise application system. The enterprise will make different analysis according to different applications, make judgments and decisions, and then send the data back to the platform layer and the edge layer, until it reaches the various departments and units connected inside and outside the enterprise.

The internal and external three-tier structure of the industrial Internet architecture

Obviously, data analysis and processing in the industrial Internet system is extremely important, including: the acquisition of endpoint data, advanced data processing technology to extract information from the data, the analysis and calculation of various decision-making models, as well as the output of system results. Among them, a large number of methods are used in computational science: need modeling, need algorithms, need data, and so on, and finally produce decision data. Of course, security, trust, privacy, etc., are also considered in detail in the structure.

Intelligent Manufacturing and the Industrial Internet

Now, the concept of the domestic industrial Internet platform to discuss a lot. Industrial Internet platform, is an enterprise as the center of the platform, rather than say in the entire industrial industry to build a large so-called "industrial Internet platform". The so-called platformization is the trend of development, in fact, it refers to the platformization of enterprises. Every large enterprise will have its own enterprise platform, and will not move its business to the platform of other enterprises. Boeing's platform will not go to AVIC's platform, and Airbus's platform will not go to Boeing's platform. If there must be a shared platform between industry and industry, it is the global Internet of Things (IoT) platform. It is not designed for any industry or any department, but for all walks of life and even individuals all over the world.

Industrial Internet platform is an ideal intelligent platform of "process". The vision is perfect, but the system is complex. In the process of implementation, there are many unknowns, and the differences between enterprise platforms of different industry categories are also great. AVIC's platforms, for example, are unlikely to be used by PetroChina and will basically have to be torn down and rebuilt. Therefore, each enterprise must proceed from its own urgent needs and actual benefits, step by step, absolutely cannot blindly follow, especially considering that the current level of China's manufacturing industry development and the level of information technology are still far from the international advanced level, the road of "process" intelligence is still far away. If all the resources and energy of intelligent manufacturing are invested in the industrial Internet platform, and the platform is understood as the platform of the industry, it may misjudge the development direction of intelligent manufacturing. The most pressing task is to make our products and equipment intelligent, which is the key direction of intelligent manufacturing for China at present.

NewsNews

Popular productsproduct

Without these three points, what else is there to talk about intelligent manufacturing?

Tags