下载PDF
实例探究 > Leading Pulp & Paper Manufacturer Detects and Avoids Major Fire Using Aspen Mtell

Leading Pulp & Paper Manufacturer Detects and Avoids Major Fire Using Aspen Mtell

技术
  • 分析与建模 - 机器学习
适用行业
  • 造纸
适用功能
  • 流程制造
用例
  • 预测性维护
服务
  • 数据科学服务
挑战
The company faced process and mechanical events at one of its wood products processing plants that had created product quality and throughput interruptions, causing product losses. The challenge was to identify and prevent these events to avoid operational shutdowns and maintain product quality.
关于客户
The customer in this case study is a leading manufacturer in the Pulp & Paper industry. The company operates a wood products processing plant where it faced challenges with process and mechanical events that were causing product quality and throughput interruptions. These interruptions were leading to product losses and potential operational shutdowns. The company needed a solution that could help it detect and prevent these events to maintain product quality and avoid costly shutdowns.
解决方案
The company implemented Aspen Mtell, a machine learning solution that scans archived data in the plant historian and correlates it with posted failure incidents in its asset management system. Aspen Mtell was able to identify the failure signature for an overheating situation in a kiln. The solution's industrial machine learning found this same pattern among dozens of sensor signal data streams, where no single sensor could provide a clear indication of impending failure. Live autonomous agents then constantly monitored incoming signals for this failure signature.
运营影响
  • Aspen Mtell provided a nine-day advance warning of imminent overheating, allowing the plant to change operating conditions and avoid an operational shutdown.
  • Autonomous agents successfully detected other impending equipment failures weeks in advance, avoiding costly shutdowns.
数量效益
  • 9 days advance warning of imminent overheating
  • Avoided operational shutdown

相关案例.

联系我们

欢迎与我们交流!

* Required
* Required
* Required
* Invalid email address
提交此表单,即表示您同意 IoT ONE 可以与您联系并分享洞察和营销信息。
不,谢谢,我不想收到来自 IoT ONE 的任何营销电子邮件。
提交

Thank you for your message!
We will contact you soon.