Multi-field bus for big data applications brought to dairy plants

This article introduces the design and application of dairy product bus, the application of big data and the design of intelligent factory - to reduce the number of factory bridges and cable laying, reduce the number of clean board openings in clean areas, meet the requirements of on-site cleaning, To meet the collection and transmission of a large amount of data, so that the control system and cable laying does not become a bottleneck restricting the arrangement of equipment, and reduce the investment cost of the factory construction, it is necessary to adopt a multi-bus network architecture to achieve distributed control. Wen / Sun Zhipeng, Wang Dongmei

The processing technology of dairy products is characterized by short process, strong batching, many detection links, strict quality control and large data collection. Moreover, the layout of dairy products is more concentrated, and there are many functions between functions. The equipment is distributed among different functions. Each function has a different level of cleanliness. In order to reduce the number of factory bridges and cable laying, reduce the number of clean board openings in clean areas, meet the requirements of on-site cleaning, and meet the collection and transmission of a large amount of data, so that the control system and cable laying does not become a bottleneck restricting equipment layout. And reduce the construction investment cost of the factory, the single use of traditional DCS and PLC can not meet the actual use needs, so the need to adopt a multi-bus network architecture to achieve distributed control.

Taking a liquid milk factory as an example, the main process of liquid milk includes pre-treatment (milking, pasteurization, batching, ultra-high temperature), filling, packaging, CIP; public works include refrigeration, boiler, water treatment, air pressure, sewage, air conditioning. The utility part is the traditional process technology. There is not much introduction here, and the automation of the pre-processing part mainly implements the ISA S88 (IEC 61512) and ISA S95 (IEC 62264) standards, and the whole system network architecture is more complicated. Through years of exploration and continuous optimization, the industrial Ethernet protocols currently used in our single factory are Profinet, AS-I, I/O Link and Modbus TCP.

The following is a detailed description of each bus design and illustrates the application of big data in dairy plants.

Profinet bus

The pre-processing process requires two sets of PLCs to complete the main process control. Then the main PLC generally selects a large CPU, because its computing power and storage capacity can meet the processing requirements of about 4 000 I/O points and 20,000 communication points. . The CPU of the main PLC is installed in the control cabinet between the pre-processing MCCs. At the same time, an industrial-grade core switch is configured to connect the industrial gateways and the PLCs of the stand-alone devices through the Profinet bus. The reason why only choose the CPU of the PLC without configuring the I/O card is mainly based on cost, installation method and configuration flexibility.

At present, the mainstream valve island also supports the Profinet communication protocol, which can directly connect the valve terminal to the access layer switch in its process area. With this design, only one communication cable (Profinet) and one 24V are used in one valve island. DC power supply, a main air supply pipe and air supply pipe to each pneumatic switch valve, which greatly saves the cable laying amount and the bridge quantity. The site is very simple, because the bus communication mode is adopted, the field valve island controller is used. More data is acquired, such as fault information and communication bus information, which meets the requirements of intelligent control and achieves refined management of preventive maintenance of equipment.

AS-I bus

In the pre-treatment site, a large number of valve arrays are used for the control of the inlet and outlet of the storage tank, such as the original milk warehouse valve array, the pasteurized milk bar valve array, and the tank valve array to be filled. The valve array is composed of a large number of switch-type sanitary valves. These valves are driven by compressed air and have valve position feedback. If traditional DI and DO are used for control, a large number of I/O cards are required, and a large number of cables are required. It is very difficult to lay a large number of cables in the already crowded valve array, and it is extremely inconvenient for later maintenance.

At present, most of the valve arrays we use are Alfa Laval and Tuchenhagen. The drive parts of these valve arrays support the AS-I bus. The reason why the AS-I bus is selected is mainly based on its digital response speed (bus scan period is less than 5 ms). ), bus anti-interference ability, easy installation (flat cable piercing connection), strong bus load capacity (single bus 496 I / O), high protection level (IP69K). After adopting the AS-I bus, there are only two cables for each valve in the access valve array. One is the power bus and the other is the communication bus. This solution reduces at least one cable (valve feedback), but A valve array is composed of a large number of sanitary valves. Each valve reduces one cable. The amount of cable reduced by one valve array is very large, the overall calculation cost is reduced, and the site is very simple. By using the AS-I bus, we have also obtained more valve diagnostic information. Compared with the traditional scheme, the loop data collection is greatly improved. These data are used for valve operation analysis, guiding maintenance and repair, and assisting in the development of spare parts planning. .

I/O Link Bus

In addition to the digital quantity, we also need to deal with a large number of analog quantities, such as temperature, pressure, liquid level, flow rate and control of the regulating valve. The traditional processing method is processed by AI and AO card, but there is also a line in this way. The problem of large cable laying and low information collection is that I use I/O Link communication to collect data from the field instrument. This communication method is similar to the AS-I bus.

The reason why the I/O Link bus is used in analog signal processing is mainly based on the design characteristics of the bus for the field instrument. In addition to collecting process variables, it can also collect diagnostic information in the I/O Link instrument. Meets the smart factory's collection requirements for equipment and process variables.

Modbus TCP bus

The Modbus bus has a long history. It was first introduced by Schneider-owned Modicon. The Modbus bus is used. It is mainly based on a large number of multi-function power meters in the factory and a communication protocol reserved at the periphery of the plant. . In order to realize the function of the energy management system, we need to collect a large amount of meter data. In addition to the flow, we also need to collect electricity. Then, when collecting power, we need to use the Modbus RTU protocol. We design a bus to connect up to 15 power meters to ensure that the data delay is within the acceptable range. Modbus RTU passes through the switch after being converted to Modbus TCP by the gateway. It is sent to the energy management system server for processing, which realizes the data collection of the electricity and the peripheral flow meter of the plant, and it is also the simplest and most cost-effective solution.

The above describes the bus design of the whole plant. The following describes the application and processing of these data on the bus, and how to implement big data and intelligent applications in a dairy factory. First, let's look at the network architecture of a factory, see dairy products. Factory network architecture diagram.

We designed a MES system at the factory that includes functions such as energy management, quality management, traceability management, production scheduling, equipment management, CIP cleaning reports, asset management, and personnel management. In addition to the conventional process equipment control, the energy management function is realized through a large amount of data collected on the bus. The water, electricity, steam, gas and cold of the whole plant all enter the energy management function module, and these data are used to analyze various processes of the factory. The energy consumption status of the segment, as well as the operating status of the equipment and the batch of the product, so that we can clearly understand the actual energy consumption values ​​of different batches, different production volumes, different items and different shifts of personnel, through accurate The data calculation, the final analysis results will be used in the factory's KPI index formulation, as well as energy-saving and consumption reduction programs, production process optimization, scheduling optimization, energy purchase program development, etc., greatly improving the efficiency of energy management.

The laboratory data is also communicated via Ethernet to the MES system server via the Lims system. The indicators are directly related to the process, the test data is interlocked for production release, and is associated with the batch information to generate a report for traceability. Through the instrument information collected by the bus, the maintenance and repair plan of the instrument can be directly formulated, the instrument fault can be accurately located, the asset management of the instrument and equipment can be realized, the automated preparation of the spare parts plan can be realized, the inventory can be reduced, and the use of funds can be optimized. Through the wireless AP, the whole plant's mobile operation and mobile inspection function are realized. Using the mobile terminal to scan the device RFID tag, it can be automatically linked to the inspection report, and can trigger the maintenance process, fully automated management equipment operation, greatly reducing the amount of labor and reducing the operating costs of the plant.

Dairy Factory Network Architecture

The OEE analysis of the equipment calculates the real-time OEE through the actual operating data of the equipment. When the OEE is reduced to the set value, the system automatically alarms, and the equipment that directly causes the OEE of the entire production line to be lowered is displayed on the screen, by further clicking on the OEE of the equipment. The three component parameters can deeply drill the equipment failure that causes the efficiency of the equipment to be reduced, call the equipment fault, and further relate to the inspection report, equipment maintenance manual, equipment spare parts inventory and other information, and realize the intelligent OEE and equipment management functions. . The production scheduling function of the MES system is to obtain the order information within the specified date range through ERP, query the inventory information according to the order information, and automatically optimize the combination of equipment according to the cleaning condition of the plant equipment and the equipment capacity, and automatically carry out the production task. When the equipment is reached, the equipment prompts the operator to start the feeding production. All the links in the whole process are automatically associated and released, and through the traceability management function, the raw material information corresponding to each work order, the batching information of the batch products, The information of the staff on duty, the equipment information passed by the materials, etc., realize the traceability management of the whole process. The MES system simultaneously interacts with the milk collection system and the finished product library to ensure the integrity of the traceability. Finally, all business units (production, quality, safety, logistics, equipment, human resources, procurement, finance, logistics) will be intelligently operated by ERP and MES systems throughout the country, greatly improving the utilization rate of the factory production line and realizing The load balance of the production line.

All control and management of the factory relies on a large amount of data collection and analysis on the site. With this data, managers can understand the actual situation of the production line. Through the support of big data and intelligent analysis, the production plan can be more accurately determined.

In the next step, we will correlate market data, correlate consumption trend data with production and logistics data of factories across the country, and achieve balanced intelligent analysis from the consumer end to the production end to better adjust the product quality and output. With the continuous introduction of various types of buses, a large amount of data has been discovered and applied. The future development trend of the dairy industry will be the artificial intelligence analysis to dominate the production and management, and a large number of artificial intelligence analysis to achieve flexible production processes. .

(+)Sparteine

Natural Sparteine,Plant Therapy Sparteine,Sparteine Cas 485-35-8,Sparteine Extract Powder

Shaanxi The River Pharmaceutical Co., Ltd. , http://www.dahepharma.com