The hottest intelligent manufacturing the next tuy

  • Detail

The next outlet of Intelligent Manufacturing: industrial intelligence

intelligence, which can be understood as data and AI based on it. Starting with production line automation, multi-source and heterogeneous industrial data are collected, transferred, analyzed and helped to form decision-making and control. End-to-end solutions form a typical portrait of current industry players

industry is generally divided into process industry and discrete industry. The biggest difference between the two lies in the degree of automation of production, the availability of data and the complexity of industry. The biggest commonality is that each scenario has different needs. Entering any subdivision requires a deep enough industry knowhow and upstream and downstream resource integration capabilities

intelligence can be understood as data and AI based on it. Starting with production line automation, multi-source and heterogeneous industrial data are collected, transferred, analyzed and helped to form decision-making and control. End-to-end solutions form a typical portrait of current industry players

why industrial intelligence

blue ocean

the total GDP of industry, especially manufacturing, is much higher than that of retail, finance, construction and other industries. In fact, the amount of effective data generated in the industrial field every day is no less than that of bat and other Internet companies. The amount of data generated by a large-scale factory every day can even reach billions to tens of billions


although industrial scenes generate high-frequency and massive data every day, a large amount of raw data itself has no direct significance, and may cause large-scale delay and occupy a large amount of bandwidth. We not only need to do real-time monitoring and analysis in some scenarios, but also need to collect more data to the cloud for more dimensional and longer-term economic benefit and value analysis, which is the value of cloud computing. Cloud computing + edge computing is a finer granularity and more complex architecture than traditional consumer interconnection, which also means higher barriers

inflection point

a logic of interconnection is called "copy to China", and "copy to industry" is the same. The large-scale data application and platform architecture have experienced full verification and evolution in the finance, telecommunications and other industries, coupled with the catalytic effect of made in China 2025 on the policy side, constitute the prerequisite for the establishment of the inflection point

player portrait of industrial intelligence

what users need at this stage is not a single product, but an end-to-end overall solution. A qualified industrial intelligence company should have the ability to construct overall solutions

first of all, user needs always come first, and technologies that do not meet their needs are false propositions. In addition, a good solution starts with a perfect architecture. For industrial scenarios, from the integration of internal and external multi-source data, to the platform architecture of cloud + end, the establishment of knowledge base, the selection of appropriate models, and then to reverse decision-making and control, only through complete connection can a closed loop be formed

on the whole, industrial intelligence presents a horizontal (overall architecture) +n vertical (multiple sub industries) pattern

path selection of industrial intelligence

for big B customers in the industrial field, what they need at this stage is not a single product, but an end-to-end overall solution. Although this is the current situation, it is also the ultimate goal of industrial entrepreneurs. However, path selection is very important

as for the development path, the mainstream in the industry believes that automation - (digitalization) - informatization - intellectualization is a reasonable order for industrial users to advance, and the previous stage is a necessary condition for the start of the next stage. Therefore, domestic enterprises in the field of industrial intelligence only pay attention to the opportunities in the field of automation for a long time, and even equate industrial intelligence with "robot" or "industrial automation". Judging from a large number of practices on the user site, these stages have a significant sequence, but at the same time, they cross penetrate and iterate

specifically, Professor Bernd Meyer, head of Fraunhofer Institute of production technology and materials utilization, revealed that most customers in the manufacturing industry are not automated enough, so priority should be given to completing production line automation. Some manufacturers use Taihe board to realize equipment interconnection in industry, get through equipment level data, feed back to the platform layer through MES, and realize preliminary IOT without replacing the original industrial control equipment. The user acceptance is very high, and the performance increases several times every year, with a very obvious trend. This kind of mode can be called "system integration with M2M device IOT as the core"

further demand comes from super large head customers in discrete manufacturing industry and the vast majority of customers in process manufacturing industry. Due to the high degree of automation of production line, we observed that such customers also have a high degree of acceptance of informatization

another kind of manufacturers can directly start from the top-level design, serve users with industrial big data platform or scenario AI model at the platform level, and solve business problems in real time. On the other hand, in the data acquisition layer, sensors are installed in some parts with incomplete data, intelligent detection equipment is installed, and even small-scale production line integration is done. This kind of mode tends to be more acceptable to users, which means that the premium of the project is often higher. We can call it "system integration with data application as the core"

therefore, we can see three development paths, facing different customers, different scenarios, different development stages, there are different path choices:

first, system integration with production line automation as the core

second, system integration with M2M device IOT as the core

third, system integration with data application as the core

of course, different paths lead to the same goal. Ultimately, it is to provide users with overall solutions, with meeting user needs as the core

industrial big data of industrial intelligence

first of all, where is the data

one is management data: structured SQL data, such as product attribute, process, production, purchase, order, service and other data. This kind of data generally comes from ERP, SCM, PLM and even MES systems of enterprises. The amount of data itself is small, but it has great mining value

the other is the data of machine operation and IOT: most of them are unstructured and streaming data, such as equipment working conditions (pressure, temperature, vibration, stress, etc.), audio and video, log text and other data. This kind of data is generally collected from equipment PLC, SCADA and some external sensors. The amount of data is large and the collection frequency is high. It is necessary to share the experience after the pilot to cement, electrolytic aluminum Do some pretreatment for other industries with excess capacity such as flat glass

in general, due to the fragmentation and dispersion of the scene, industrial data itself has the characteristics of large volume, multi-source, heterogeneous and high real-time requirements. With the gradual access of 28billion devices in the future, these characteristics will be further strengthened. This is one of the core difficulties in doing industrial big data services. It is not only different in magnitude, structure and application from interconnected big data

secondly, based on these industrial data, what services should the platform layer provide

complete protocol analysis: data collection must first complete the opening of industrial protocols. Taking the application layer protocol as an example, ethernet/ip and PROFINET have the largest market share, followed by EtherCAT, Modbus TCP and ethernetpowerlink

standardized data integration: the collected data should be subject to unified master data management. The first step is to establish standards. Generally speaking, we need to use ISO or other industry standards to formulate unified coding, structure, circulation mode and attributes to ensure the consistency of data, which is very important

in the process of project implementation, it is also important to gradually accumulate industry knowledge base, appropriate algorithm components and related mechanism models. This is a key step from data standard evolution to business standardization, and lays the foundation for the realization of real product level microservicing

strong PAAS support: due to the particularity of industrial data itself, the platform must have strong middle-level support ability. We take the time series database as an example, which is a typical variety of equipment working conditions and sensor data. This kind of data has high frequency and large volume. It needs to pull out all values for calculation every time when it is processed with traditional relational database. The throughput is great and the performance is very poor. Therefore, a highly compressed and high-performance temporal database is one of the necessary capabilities of the platform layer

finally, what applications should we do

equipment level: quality control. In the era of industrial intelligence, if we can collect the appropriate real-time data of Shanying paper () in the first quarter of 2019, combined with the mechanism model applicable to the equipment, it is possible to mine the correlation or causal relationship between product quality and key data with machine learning method, and it is also possible to realize real-time quality control and fault early warning. If the data frequency can form a perfect envelope for the process flow, We may also achieve maximum efficiency improvement

plant level: planned production scheduling. The ultimate goal of industrial intelligence is to achieve large-scale personalized customization, that is, c2m. The goal of this problem is to achieve the optimal local production capacity at that time. The constraints come from the production line equipment, personnel, product attributes, supply chain data, etc. through the learning and training of historical data, it is not difficult to form a better prediction model

this model can dynamically analyze and adjust according to the real-time data of production lines and factories, so as to help enterprises achieve accurate control and maximize economic benefits

in the foreseeable future, as the integrity and reliability of data become higher and higher, and the scenarios become richer and richer, a considerable number of priority enterprises will be born at the data application level. They help industrial users reduce costs, improve efficiency, and solve real business problems

Copyright © 2011 JIN SHI