Casosprácticos

Mondi通过机器学习实施基于统计的健康监测和用于制造过程的预测性维护

挑战

减少塑料制造工厂中的浪费和机器停机时间

解决方案

Use MATLAB to develop and deploy monitoring and predictive maintenance software that uses machine learning algorithms to predict machine failures

结果

  • More than 50,000 euros saved per year
  • 原型在六个月内完成
  • Production software run 24/7

“ Mathworks Consulting的支万博1manbetx持是我见过的最好的；顾问快速且知识渊博。我们已经看到了节省成本的积极投资回报，现在我们有更多的预算和时间来完成更多的机器学习项目，这些项目将提供类似的收益。”

Dr. Michael Kohlert, Mondi

Mondi Gronau是包装和纸产品的国际领先制造商。s manbetx 845该公司的塑料生产工厂每年提供约1800万吨塑料和薄膜产品。s manbetx 845该工厂的900名工人每天24小时365天，每天24小时运营大约60个塑料挤出，印刷，粘合和缠绕机。

Machine failures that result in downtime and wasted raw materials cost Mondi millions of euros each month. To minimize these costs and maximize plant efficiency, Mondi developed a health monitoring and predictive maintenance application. The application uses advanced statistics and machine learning algorithms to identify potential issues with the machines, enabling workers to take corrective action and prevent serious problems.

Mondi在MATLAB中开发了该应用程序®教授的支持万博1manbetx下MathWorks咨询和Dr.-Ing. Andreas König, holder of the Chair for Integrated Sensor Systems (Department of Electrical Engineering and Computer Engineering at the Technical University of Kaiserslautern).

“As a manufacturing company we don’t have data scientists with machine learning expertise, but MathWorks provided the tools and technical knowhow that enabled us to develop a production preventative maintenance system in a matter of months,” says Dr. Michael Kohlert, head of information management and process automation at Mondi.

Mondi Gronau的塑料生产机器之一，每年提供约1800万吨塑料和薄膜产品。s manbetx 845

挑战

蒙迪工厂的挤出和其他机器既大又复杂，长达50米，高15米。每台机器最多都由五个可编程逻辑控制器（PLC）控制，这些逻辑控制器（PLC）来自机器传感器的日志温度，压力，速度和其他性能参数。每台机器每分钟记录300–400个参数值，每天生成7 GB的数据。

Mondi在使用此数据进行预测维护方面面临一些挑战。首先，植物人员在统计分析和机器学习方面的经验有限。他们需要评估各种机器学习方法，以确定哪些方法为其数据产生了最准确的结果。他们还需要开发一个应用程序，该应用程序清楚地将结果呈现给机器运营商。最后，他们需要包装此应用程序以在生产环境中连续使用。

解决方案

Mondi与Mathworks Consulting和-Ing教授合作。AndreasKönig将在MATLAB中开发和部署健康监测和预测维护软件。

The Mondi team had previously set up an Oracle®数据库通过以太网网络从工厂中的所有机器中收集数据。他们使用数据库Toolbox™从MATLAB内部访问此数据库。

接下来，团队开发了MATLAB脚本来通过删除异常值和无效值来清洁数据。

They developed an application in MATLAB to query the database and present the results graphically. For example, an operator can use the application interface to plot the pressure measured by a particular sensor over a period of minutes, hours, or weeks.

To enhance the application, they added statistical process control (SPC) capabilities that alert operators to sensor values that are outside normal operating ranges.

Mondi和Mathworks Consultants使用统计和机器学习Toolbox™和Deep Learning Toolbox™，评估了几种机器学习技术，包括神经网络，K-Nearest邻居，行李袋的决策树和支持向量机（SVMS）。万博1manbetx

For each technique, they trained a classification model using logged machine data and then tested the model’s ability to predict machine problems. The tests showed that an ensemble of bagged decision trees was the most accurate model for their data.

The team further enhanced the MATLAB application by updating the interface to incorporate predictions from the machine learning model. These predictions enable equipment operators to receive warnings about potential failures before they occur. Mondi then used MATLAB Compiler™ to create a standalone executable version of the application, which is now used in production at the plant.

A MATLAB based HMI that enables equipment operators to receive warnings about potential failures before they occur.

结果

  • 每年节省超过50,000欧元。“Our financial control department determined that we are saving more than 50,000 euros per year by using MATLAB for predictive maintenance,” says Dr. Kohlert. “That total is based on just eight machines. We expect that to increase at least fourfold as we analyze the data from more of our machines.”
  • 原型在六个月内完成。“有很多顾问，有很多讨论，但没有行动，”科勒特博士指出。“ Mathworks顾问直接开始。我们在两个月内进行了第一个测试，六个测试在六个月内进行了工作原型。MATLAB代码易于理解，因此我们可以在需要时迅速进行更改。”
  • 生产软件运行24/7。Kohlert博士说：“有人误解MATLAB仅用于研究或发展。”“即使是在圣诞节，我们也不停地操作机器，并且我们依靠基于MATLAB的监视和预测维护软件来连续而可靠地运行。”

s manbetx 845使用的产品

Learn more about Mondi