1+1>2 benefits? How traditional industries use AI to drive ESG Chimes AI CEO Xie Zongzhen has cooperated with many traditional companies such as Formosa Plastics and Taipower in recent years to provide low-threshold AI for ESG solutions through disruptive innovations in technology, helping companies in energy saving, carbon reduction, industrial safety and environmental protection, In the application scenarios such as circular economy, practice the sustainable development of enterprises. How can artificial intelligence technology achieve corporate ESG goals? Xie Zongzhen said that the current main application of artificial intelligence technology includes the assistance of intelligent monitoring, such as real-time monitoring of energy use, and conditional control, including report management and alarm event management. If it is further expanded, it can be divided into several categories, including energy management applications, such as electricity consumption statistics and KPI management, contract capacity optimization, and then switching from electricity consumption to carbon inventory and carbon emission statistics. After knowing the energy management, there will be corresponding energy saving on the factory side. Based on the production quality, yield or reliability, it will be extended to the optimization control of heating system energy saving, air conditioning water system optimization control, and air pressure. Optimization of system or wastewater treatment performance. In addition, whether the reliability of the operation of the equipment in the plant can continue to operate, including the remaining life prediction of the rotating equipment, the abnormal diagnosis of the ice machine equipment, the performance diagnosis of the air-conditioning water system, and the preventive maintenance such as SPC statistical control. Extensible applications in process optimization include process formula simulation optimization, product process abnormality monitoring, optimal timely control of process, and optimal energy-saving operation of process. How can enterprises use the services provided by new innovations to solve problems? Xie Zongzhen said that traditionally, the AI application development process starts from data collection, data import and collation, analysis and modeling, application deployment and online. After the model is launched, different user interfaces will be available depending on the application product. When the operating parameters of the equipment need to be changed, the user can make adjustments through automatic control or manual control, and finally get an early warning of energy saving benefits or remaining life of the equipment. However, once the equipment is replaced and the production parameters are changed, the original model may become invalid, and the entire modeling process needs to be re-operated. To this end, Chimes AI hopes to help manufacturers build their own No-Code AI platform through a friendly graphical interface, so that the operation team can effectively copy and spread the model and provide sufficient support. Taiwan's leading petrochemical company, Formosa Plastics, is a long-term customer of Chimes AI. In recent years, it has continued to research innovative concepts, use new technologies and tools, and fully promote digital transformation to enhance industrial competitiveness. (Extended reading: How can high-energy-consuming petrochemical plants promote industrial sustainability through AI? ) Zheng Qicong, the team leader of the Formosa Plastics Maintenance Center, said that the traditional monitoring system equipment has obvious signs of abnormality before issuing an alarm, but in actual operation, it cannot wait until the equipment fails to deal with it. This approach no longer meets the needs of the industry, so it is necessary to develop extremely Early equipment pre-diagnosis system. On the other hand, the maintenance personnel of each unit need to be responsible for a large number of equipment and different types of different brands. Planning and execution are arduous tasks, and there is a lack of corresponding talents and technologies. The initial team of the system also developed the AI model by itself, but due to the lack of maintenance and operation integration interface and efficient management mechanism, it was impossible to update iteratively in a timely manner when the model deviated.
Forum Role: Participant
Topics Started: 0
Replies Created: 0