MPM201: Machine Learning Agent Best Practices for Critical Equipment

This course will help you prepare for the certification exam and the exam fee is waived with this course.


The Aspen Mtell application is a condition monitoring solution that uses automated Machine Learning to stop machines from breaking down, makes them last longer, reduces maintenance costs, and increases the net product output of any process.


In this training, users build and deploy Machine Learning Agents on few selected Critical Assets following best practices during the training session. During the training users will build Asset Hierarchy, Import Equipment Set, Import TDS, Audit Sensors data, clean and update the Equipment set, Deploy Hidden Failure Agents, Anomaly Agents, and Failure Agents. Review Anomaly Agents and Failure Agents. Review active alerts and Agent retuning. Transfer learning on similar Assets. Practice automatic Agent tuning for the best accuracy and earliest detection of excursions from normal behavior, and the precise patterns that match degrading conditions that lead to failure.

Audience:

Engineers, Consultants or Managers focused on Preventative Maintenance and System Reliability.

Training Details

  • Course Id:

    MPM201

  • Duration:

    5 day(s)

  • CEUs Awarded:

    3.5

  • Level:

    Advanced

Benefits

Build agents using your own data. Clean, prepare, and organize data for agent building. Learn advanced Agent building techniques such as transfer learning. Evaluate and retune live Agents based on changing process conditions.

Approach

  • Clear guidance on fundamental topics
  • Industry workflows
  • Hands-on workshops
  • Experienced instructor-guided demonstrations
  • Q&A on student-specific problem

Pre-requisites

Some knowledge of Aspen Mtell Prescriptive Maintenance concepts is expected. Prior completion of the MPM101 or MPM121 Learn to build Machine Learning Agents using Aspen Mtell Course will greatly enhance the value of this course.


To Build Machine Learning Agents on Critical Equipment, it's essential to complete the below prerequisites prior to the class.


  • Instructor Remote access to Application Server. The instructor will demonstrate the training on one of the customer assets.


  • Each user should have access to Aspen Mtell Client application either through Citrix or individual VDI or desktop installation with recommended hardware


  • A specific asset should be assigned to each user to use during the training. Asset Selected should be aligned with project value and MTell Capabilities. The instructor will be using a separate asset during the training to avoid conflict


  • Excel Templates should be completed before the training for each asset (number of users + Instructor).
    • Asset Hierarchy
    • Equipment Set Profile
    • Sensor Groups
    • Work Item history
    • Failure identified to work on during training
    • Offline condition for each asset

Agenda

Introduction/Overview
  • Explain the purpose of using Aspen Mtell
  • Define key terminology used in explaining the behavior of the application
  • Identify the main applications used to control Aspen Mtell
  • Describe the software architecture
  • Explain the workflow for the standard implementation strategy
Mtell SQL Server Database
  • Generate a new Aspen Mtell database
  • Switch between Aspen Mtell databases
  • Workshop: Create a new Mtell Database
Aspen Mtell System Manager
  • Open and Navigate the Aspen Mtell System Manager
  • Set up basic settings used by Aspen Mtell
  • Prepare CSV files for systems where live EAM and Historian connections are not available
  • Workshop: Configure Connections and Preferences
Aspen Mtell Agent Builder
  • Open and navigate the Aspen Mtell Agent Builder
  • Import and Update Work Order Details
  • Explain the Purpose of an Equipment Set Profile
  • Workshop: Import CSV Files
Equipment Set Profile
  • Understand the Purpose of an Equipment Set Profile
  • Build an Equipment Set Profile from Scratch
  • Import an Equipment Set Profile
  • Explain the Hierarchy below the Equipment Set Profile
  • Workshop: Generate Equipment Set Profile
  • Workshop: Import and Clean Data
Machine Learning (ML) Agents
  • Identify Machine Learning Agents
  • Understand the Purpose of each ML Agent
  • List the Expected Inputs and Outputs of Each ML Agent
  • Explain how the Different ML Agents work together
  • Optimize the performance of each Machine Learning Agent
  • Workshop: Generate and Analyze a Hidden Failure Agent
  • Workshop: Generate and Tune Anomaly Agents
  • Workshop: Generate and Tune Failure Agents
Rule Analysis
  • Analyze Excursions from previous agents
  • Run Excursion analysis to verify alarm condition of a potential agent
  • Generate rule-based agents
  • Deploy agents live
  • Workshop: Analyze a Univariate Excursion
  • Workshop: Deploy Agents Live
Mtell View
  • Understand the Purpose of and Use for Mtell View
  • Launch Mtell View and Navigate the User Interface
  • Acknowledge Alerts
  • Create and Use a Custom Process View
  • Workshop: Review Alerts in Mtell View
  • Workshop: Create a Custom Dashboard
Appendix
  • Key Definitions
  • Create Database by Running SQL Scripts
  • Using CSV Files in Lieu of Live Connections
  • Enabling the Aspen Mtell Log Manager

 

Register for a Class

Date Class Type Location Price Language

Aspen Technology, Inc. awards Continuing Education Units (CEUs) for training classes conducted by our organization. One CEU is granted for every 10 hours of class participation.