MATLAB is a language for technical computing that integrates computation, visualization, and programming. It is provided with a graphical environment where one can express problems and get solutions in familiar mathematical notation. The PI System and MATLAB are in several ways complementary. For instance, PI System is meant to effectively collect, store, and serve masses of time series data in real time, whereas MATLAB is in no way a data store. MATLAB and its convenient ad-hoc computing interface extend the computing ability of the PI System by adding its library of advanced functions. In other words, combining the "real-time data infrastructure" capabilities of the PI System and the advanced mathematical and computing functions of MATLAB is a logical choice in many applications.

In this white paper, we will discuss various ways to bring PI data into MATLAB:

- Using PI AF SDK
- Using PI Web API
- Programmatically, using a mix of PI and MATLAB functions
- Using MATLAB's Database toolbox and the PI JDBC Driver
- Using comma-separated values (CSV) files
- Using ADO and the PI OLEDB Provider
- Using MATLAB's OPC Toolbox and the PI OPC DA Server
- Using Integration Objects' OPC HDA Toolbox and the PI OPC HDA Server

The code and supporting files for this white paper can be found on GitHub.

Thanks for this helpful White Paper. I wanted to point out that in 2014b MatLab has introduced some additional functions that make reading JSON from a restful service even easier and eliminate the need to use an additional json parser.

If you are updating your Whitepaper it might be good to demonstrate how to use basic authentication since hopefully most people aren't running their PI Web API without any security.

>> options = weboptions('Username','*******','Password','********')

>> data=webread(web_url,options)

data =

UnitsAbbreviation: 'W'

Links: [1x1 struct]

Items: [44092x1 struct]