【Ten Years of Rapid Development in Chemical Equipment】2647-2025: Zhejiang Zhongkong’s world’s first large-scale temporal model for the process industry is now available

Thread Content

Give praise and encouragement to the achievements made in China’s chemical engineering technology and equipment; your participation in discussions is the greatest encouragement. **********************【Ten Years of Great Development in Chemical Engineering Equipment】A continuously updated summary post – feel free to join the discussions: https://bbs.hcbbs.com/thread-3576046-1-1.html ***************** On August 28th, Zhongkong Technology Co., Ltd. (hereinafter referred to as Zhongkong Technology) unveiled the time-series large model TPT 2 (Time-series Pre-trained Transformer 2) in Hangzhou, for use across the world. Building on TPT 1, the world’s first large-scale temporal model for the process industry developed by this company, TPT 2 will focus on addressing long-standing issues in this sector such as difficulties in energy consumption optimization, frequent failures and accidents, and large fluctuations in product quality. By leveraging capabilities in temporal prediction and intelligent diagnosis, it enables precise monitoring and dynamic optimization of production processes. It is intended to be deployed on a large scale in industries such as petrochemicals and chemicals, thereby contributing quantifiable benefits in terms of cost reduction and efficiency improvement for these sectors. 　　“TPT 1 has proven its usability and effectiveness in over 110 cases. The newly launched TPT 2 further advances the intelligent application in the process industry with an updated architecture and more powerful models. ”Chu Jian, founder of Zhongkong, said that TPT 2 is not a general-purpose large model for simple applications; rather, it is deeply integrated into industrial scenarios and aims to address core issues such as safety, quality, cost, efficiency, and low carbon emissions that companies are concerned about, thereby helping them enhance their competitiveness in the current economic environment. 　　“Energy conservation, reduction of consumption, and improvement of safety in the chemical industry are all critical issues, which place strict demands on industrial AI; TPT was created precisely for this purpose. ”In an interview with reporters, Chu Jian said that Zhongkong Technology will continue to advance product iterations, strengthen cooperation with users, and help boost the global influence of industrial AI in China, making it a key tool to support the intelligent transformation of industry. 　　Zhang Huize, vice president of Zhongkong International Operations and the main publisher of TPT 2, said that TPT 2 is a revolutionary industrial AI tool designed to meet the core needs of process industries such as oil and chemicals. Based on industrial time-series data to develop AI algorithms, this model is the world’s first reliable model that fully aligns with the first principles of process industry. It can be adapted to a wide range of production processes and utilized in closed-loop systems; it enables the precise extraction of value from industrial data. It plays a key role in critical scenarios such as ensuring production safety, improving product quality, reducing energy and material consumption, optimizing equipment maintenance efficiency, and facilitating autonomous factory operation, providing enterprises with integrated support ranging from solutions to industrial intelligents. 　　It is reported that TPT has been applied in over 100 units at companies such as China National Petroleum Corporation, Sinopec, Wanhua Chemical, and Datang Dolun Coal Chemical Industry, with significant results. Li Wei, General Manager and Chief Engineer of China National Petroleum Corporation Lanzhou Petrochemical Yulin Chemical Co., Ltd., said that TPT 2 helps Lanzhou Petrochemical Yulin Chemical to effectively address industry challenges such as weak sensing capabilities of ethylene production units, difficulties in achieving operational efficiency balance, and slow adjustment processes. The accuracy of anomaly prediction reaches 99.79%, and the yield of ethylene per reactor increases by 0.373%. 　　Nearly 700 representatives from leading companies, industry organizations, and research and investment institutions across dozens of ** and regions around the world attended this press conference. As of that day, there were 518 companies with pre-orders for TPT, 112 units were sold, and 3,268 users had registered online. 　　“TPT 2 is the trend of the future; in the future, truly unmanned factories may be built using it. ”Wang Cong, deputy general manager and chief engineer of China Tianchen Engineering Co., Ltd., told the reporter.