Artificial intelligence and artificial self-healing can reduce chemical accidents

Thread Content

Xinhuanet, Nanjing, September 22 (Yu Qizhong and Dong Xian) "We did some big data analysis on chemical safety accidents and found that human errors and equipment problems each accounted for about 40%. The combined proportion of the two accounted for more than 80% of chemical safety accidents, and human errors were ultimately the main cause of equipment problems. Therefore, we have conducted long-term follow-up research and used vibration principles to apply artificial intelligence monitoring in chemical systems to provide early warning of equipment failures. We have also performed artificial self-healing to repair chemical equipment failures through multiple experiments. Current statistics show that artificial intelligence and artificial self-healing can effectively reduce chemical accidents. ”On the morning of September 21, the 3rd China International Chemical Process Safety Symposium was held in Nanjing. Gao Jinji, academician of the Chinese Academy of Engineering and professor of Beijing University of Chemical Technology, gave a keynote speech on "Artificial Intelligence and Artificial Self-Healing Drive the Intrinsic Safety of Chemical Industry" at the symposium, sharing the research results and latest opinions of him and his team. Gao Jinji believes that "no action means safety, no damage means safety." The so-called intrinsic safety of chemical industry should be that there is neither harm nor damage, that is, the possibility of danger evolving into an accident is zero. How to reduce or even eliminate the occurrence of chemical machine failures? He elaborated on the application of "artificial intelligence" in chemical equipment and the concept innovation of "artificial self-healing". Intelligent monitoring prescribes the right medicine “Since 1991, we have independently developed a monitoring network and diagnostic system for machine and pump groups, which have been successfully used in hydrocracking and other equipment, thus avoiding many major equipment accidents. ”Gao Jinji said that the causes of accidents in the chemical industry are mostly manifested in the failure of production equipment, and even manual operation errors are often reflected in the equipment. How to ensure the safety of equipment has become the key to ensuring the safety of the chemical process. In recent years, Gao Jinji and his team have independently developed a data collection system and automatic fault diagnosis expert system based on the industrial Internet, and developed and established wired and wireless remote monitoring systems for rotating machinery, reciprocating machinery, etc., using remote databases and historical databases to provide samples for fault diagnosis. “We have classified equipment faults into 10 categories, a total of 56 types, and analyzed the characteristics and changes of each fault signal. ”Gao Jinji said that the sensor is then connected to the equipment to remotely monitor the equipment and collect the temperature, vibration and other information of the equipment in real time, so that the status of the equipment can be understood in a timely manner, and faults within the machine can be discovered immediately and measures can be taken. “The remote monitoring of artificial intelligence on chemical equipment can provide early warning of faulty equipment, which can not only prevent accidents, but also extend the operating cycle and reduce maintenance costs. In the next step, the team will study higher-level artificial intelligence applications to more accurately judge equipment failures and directly issue equipment shutdown instructions to avoid major safety accidents. ” Automatic regulation and self-repair "Remote monitoring of artificial intelligence can detect the 'disease' of chemical equipment in time. This method is to manually inspect and maintain the equipment after early warning. It is a passive protection of the machine. For large chemical companies, parking sometimes causes greater losses. ”Gao Jinji said that adhering to a problem-oriented approach, he and his team proposed the concept of "artificial self-healing" and conducted repeated tests in the laboratory. “We hope that chemical equipment should not only be 'smart', but also 'healthy' and be able to automatically repair themselves after encountering a fault, just like the human body can heal itself after a cold. ” Gao Jinji believes that only by realizing artificial self-healing of equipment can we truly realize active protection of machines and equipment and improve safety values. “Machine fault self-healing technology is still based on the principle of vibration. It is installed into the data acquisition system at the beginning of the equipment manufacturing. When an abnormal vibration wave occurs in the machine, the fault can be automatically discovered and control measures can be taken to realize the self-healing of the machine during operation and eliminate the fault in the bud. ”Gao Jinji said that this is a closed-loop system, and testing and regulation are all proactive, and everything is completed inside the equipment. Compared with passively controlled emergency systems, self-healing systems eliminate manual operations. “After artificial intelligence in the manufacturing field, it must be artificial self-healing. ”Gao Jinji said that although the technology is still in the laboratory stage, the concept of artificial self-healing has attracted great attention in the field of mechanical equipment at home and abroad. Artificial intelligence is currently in the ascendant, and artificial self-healing has broad development prospects, and will be promising in chemical equipment safety and other fields.