1. Project background
Xinyu Iron and Steel Group Co., Ltd. (hereinafter referred to as Xinyu Iron and Steel Group Co., Ltd.) is a large-scale provincial state-owned steel complex with a production capacity of 10 million tons.
After more than 60 years of development, Xingang has become an integrated company integrating ore mining, steel smelting, steel rolling and extended processing. It has more than 800 varieties of general steel, special steel, metal products, steel structures and chemical products. Manufacturing enterprises with more than 3,000 specifications. The company’s plate production capacity reaches 7 million tons/year, with thickness covering 0.28 mm-380 mm.
Among them, coking production mainly provides coke for the smelting workshop, and the coke oven production line mainly involves key links such as coal blending, coking, coke quenching, storage and transportation. The environment is relatively harsh, mainly manifested in high temperature and gas explosion environments. In accordance with the overall policy of building digital factories and smart factories at the group headquarters, we adhere to the overall philosophy of people-oriented and safe production. In order to solve many shortcomings such as untimely discovery of equipment hidden dangers, uncertainty of manual inspections, inaccurate equipment parameter data, and inadequate supervision, in order to eliminate safety hazards and prevent safety accidents, after research and investigation, it was proposed to establish an inspection robot system to replace Manually conduct coke oven inspections and solve the above problems. Inspection robots solve the problem of people in high-risk and harsh environments and special areas being unable to enter for on-site inspections. At the same time, they also provide a strong guarantee for production and provide strong support for enterprises to produce safely, reduce costs and increase efficiency.
2. Project scope
In order to meet production needs and improve the intelligence level of the overall coking process, a set of intelligent robots for coke oven flue and basement inspections were built.
The content of the inspection robot inspection and the overall structure of the basement are introduced: The coke oven is a 2*65-hole 7.6-meter tamping coke oven, which is divided into an above-ground layer and an underground layer (a coke oven below the two coke ovens is the example).
1) First floor above ground
It is air and exhaust gas exchange equipment, distributed on both sides of the ground floor. Every half hour, the on-site guide chain and the cover plate will be operated through oil cylinders and pull rods to switch the air intake and exhaust states. The guide chain and cover plate form a group, with 67 groups on one side and 86 groups on both sides. Each group controls the air intake and exhaust of an air port respectively. The status of each airport and the adjacent airport are in the opposite state. The robot patrols the ground floor. The inspection work mainly involves analyzing the position status of the equipment’s cover plate, guide chain, and chain gyroscope.
2) Basement level
The underground layer is equipped with gas pipelines and gas valve reversing equipment, which are linked with the above-ground layer. Every half hour on-site, the swing arm is driven to rotate 90 degrees through the oil cylinder pull rod according to the exchange procedure to reach the exchange limit position. There are 67 swings in a coke oven. Arm, the robot needs to detect whether each swing arm swings in place.
3. Project application advantages
(1) Technical aspects
The coke oven inspection robot improves the overall process chain by integrating multiple disciplines such as robotics, non-contact monitoring technology, multi-sensor fusion technology, image recognition technology, deep learning technology, navigation and positioning technology, and Internet of Things/industrial WIFI communication technology. Intelligent level.
The Coke oven inspection robot body adopts an explosion-proof design and is suitable for Class II explosion-proof environments except for Class I explosion-proof areas. It can be equipped with gas leak detection sensors as needed to collect on-site gas leaks and can realize online analysis and detection of different gases.
The coke oven inspection robot uses AI artificial intelligence technology, deep learning and video image analysis algorithms to conduct in-depth learning and analysis of on-site equipment, with a recognition accuracy of up to 98%. Global anomaly identification and alarm ensure that each inspection fault is accurately identified and processed. The analysis results are recorded in history and the storage time is not less than 30 days. Multiple methods of combined query retrieval, historical data statistical information graphic display, concise and intuitive, provide data support for production. At the same time, different algorithms can be used to identify and read data from different instruments, valves, local meters, etc.
The coke oven inspection robot uses an explosion-proof non-contact wireless charging system to charge the robot non-contactly, eliminating contacts and providing continuous battery life for the robot. The battery life is not less than 8 hours, and the charging time is 0.5-3 hours.
The coke oven inspection robot can be equipped with a two-way voice intercom system to achieve a two-way voice intercom between the scene and the dispatch room, providing a remote command system for dispatch.
The intelligent inspection robot comprehensive management platform software is based on the SUPOS industrial operating system and is a system platform dedicated to the comprehensive management of robot inspections. The platform is based on B/S architecture, providing users with a good human-computer interaction interface, and also providing standard APIs for access to other platforms. It can realize data docking with various systems such as on-site DCS, MES, and ERP. Realize data interoperability and data sharing for the production site.
(2) Management aspects
The coke oven inspection robot is designed to solve the hidden dangers of equipment failure during manual inspection of high-temperature, toxic, and explosive areas. The inspection is not timely, the hidden dangers are not easy to find, the personnel are not inspected in a timely manner, and the inspection data cannot be generated. Safety accidents or production ACCIDENT. Coke oven inspection robots effectively solve such safety problems, ensuring production while ensuring the safety of equipment and the safety of personnel and property.
The establishment of the coke oven inspection robot system can replace manual inspections in high-temperature and toxic explosion areas. It can operate 24/7, fundamentally freeing up personnel and providing a strong guarantee for enterprises to increase production efficiency and reduce costs.
The system automatically generates various data reports and historical analysis data ledgers, which can be used by upper-level managers to provide data support and assessment basis for production equipment and production processes and provide data guarantee for production decisions.
4. System architecture
Robot layer: visual analysis unit, power management BMS unit, drive management unit, sensor unit, main control unit, suspension support system, communication terminal module, etc.
Network layer: WIFI wireless communication terminal, point-to-point communication module, robot management server, data management server, engineer and operator station, etc.
Application layer: Intelligent robot inspection comprehensive management platform software basic platform and various modules.