MAY-JUN 2017

Issue link:

Contents of this Issue


Page 17 of 53

18 INTECH MAY/JUNE 2017 WWW.ISA.ORG PROCESS AUTOMATION focus, an embedded machine-learning capability can eliminate the need for data science expertise or for custom software development. A specific example of an embedded ML capability is an engine that performs pattern recognition and classification on multivariate time series data, which includes continuously recorded sensor and parametric data, as well as inter- mittently collected inspection mea- surements. Process data is largely time series data, and a real-time pattern- recognition engine is a practical way for operations teams to better understand the state of process machinery or the process itself from process data streams. Pattern recognition and classification could, for example, be used to: l identify a quality issue with raw materials l identify process configuration issues l provide early warning of a maintenance need or impending machine failure Figure 3 shows the use of a pattern- recognition engine with a process histori- an. Key elements of this approach include: l easy deployment and integration to existing data store l augmentation of the existing data stream (i.e., output feeds back into the data store) l usable by the existing business or operations team No data science requirements, such as feature engineering, algorithm selection, or hyperparameter tuning Use of a pattern-recognition engine can be simple and straightforward. A user needs to: l Configure the engine for a particular need Select a type of entity of interest (e.g., a machine, a line, a process step or phase) Select a set of signals that could con- tain patterns that reveal the state of each entity l Validate the model produced by the engine for the desired need l Manage the execution of the engine Tell it when to update models Control when it is applied to live (re al-time) data l Provide labels of known conditions against occurrence time, where available The engine will find patterns on its own, but it needs the subject- matter experts to give names and contextualization to conditions. A properly embedded pattern-recog- nition engine can be used in the same way that historian features, such as cal- culated fields or attributes, are used to augment raw data streams. For example, the output from the engine can be: l fed to a control system, making it a form of soft sensor l used in dashboards, rules, thresholds, and alerts—providing a digested mea- surement of state l used in historical data analyses One advantage of a pattern-recogni- tion-based approach is the flexibility. The models produced are purely data driven, and do not require an under- standing of the causal relationships, or a detailed understanding of the signal origin. Large numbers of sig- nals from disparate sources could be speculatively thrown into a pattern- recognition engine to identify condi- tions. New sensors could be added to attempt to capture phenomena of in ter- est. A simple example is the combina - tion of process execution and quality data with machine trace data from a manufacturing execution system. As Figure 2. Time series data patterns Figure 3. Typical pattern-recognition engine

Articles in this issue

Links on this page

Archives of this issue

view archives of InTech - MAY-JUN 2017