Musings on Automation: Happy New Year!
Published on : Tuesday 03-01-2023
Many new things are happening in the field of automation, says PV Sivaram.
Anand comes bouncing with joy – “Happy New Year”! I reciprocate in a more sober manner. Magazines are full of numbers – achievements of the old year, and targets for the New Year. Alongside, there are also predictions. Predictions based on knowledge, predictions which are guesswork. Astrologers too get a premium position for their type of predictions.
What will be the important topic of the coming year, Anand? He is a little stuck, but naturally. His mind is going along the predictions of the journalists. Anand, I pull him back out of his reflections. We are not going to talk about politics or economics. We are going to talk about automation.
Many new things are happening in the field of automation. Hardware and the controllers are always gaining in speed, memory capacity, and power of computation, communication bandwidth and speed. One thing that has long term implications is how algorithms or codes get generated. Is there any new programming language or system going to be launched, Anand asks a little breathlessly. A new programming language or coding platform will not create a major change. What is going to gain in strength is a completely new approach.
What is this new approach? Well, as you know, all programs, including programs for PLCs and such controllers, start with a method. This method may also be called an algorithm. This algorithm is a way to solve a problem and yield a result. The method is codified in a language for the computer or processor. Then we provide inputs, and the program yields a result.
In the real time world of automation and control, it works in a similar way. There is a desired process output. The deviation from the desired value to the actual achieved value is the error. Effort of the control loop is to bring the error or deviation to a minimum. So the error value is fed to the controller, who tries to adjust the process in a way to reduce the error. The way in which the controller adjusts the process is the control algorithm or control program. This program is created by a programmer at his desk, using his knowledge and experience and intelligence. This is an endless loop.
The new way is that instead of the human programmer, Artificial Intelligence can generate the algorithm. How does this work? We use a specialised part of Artificial Intelligence (AI) called Machine Learning (ML).
Let us take an easy example of oranges and apples. The control problem is to sort fruit into three buckets – apples/oranges/neither. In the technique of ML, we present the machine (the computer/the algorithm) with a large number of results in three buckets. The buckets are labelled as apple/orange/neither. The processor/controller/algorithm studies each of these samples in each bucket and extracts some defining traits or features which distinguish an apple from an orange. By inspecting a random fruit coming in, and measuring for the traits, the algorithm can put the fruit into the appropriate bucket.
Anand has lost the thread. How does all this connect with the famous PID loop? That is the core of all control processes! Wait, Anand, wait. You surely know that the PID controller makes use of three significant parameters such as kP, kI and kD to drive the controller. The choice of these three parameters can seriously influence efficiency of the process. The indications are obtained from response to a step change in the set-point or response to process disturbance.
The classical way of going about discovering these parameters is to first make an intelligent guess. As the process runs in a closed loop, to introduce a set-point change and check the behaviour. Based on the behaviour, one can change the parameters. An improvement on this strategy is to allow auto-tuning. In auto-tuning a set-point change is imposed and the system calculates the parameters. Using these parameters, the system is switched to online closed-loop control mode.
The advanced mode is self-tuning. Here the controller is continuously monitoring the system behaviour and adjusts its parameters on an ongoing basis. This yields much higher efficiency. Do you now see the connection to ML, Anand?
Wishing you all a Happy, Healthy and Peaceful New Year.
PV Sivaram, Evangelist for Digital Transformation and Industrial Automation, is mentor and member of steering committee at C4i4. He retired as the Non-Executive Chairman of B&R Industrial Automation and earlier the Managing Director. He is a past President of the Automation Industries Association (AIA). After his graduation in Electronics Engineering from IIT-Madras in 1976, Sivaram began his career at BARC. He shifted to Siemens Ltd and has considerable experience in Distributed Systems, SCADA, DCS, and microcontroller applications.
Sivaram believes strongly that digitalisation and adoption of the technology and practices of Industry4.0 is essential for MSME of India. He works to bring these concepts clearer to the people for whom it is important. He believes SAMARTH UDYOG is nearer to the needs of India, and we must strike our own path to Digital Transformation. Foremost task ahead is to prepare people for living in a digital world. He is convinced that the new technologies need to be explored and driven into shop floor applications by young people. We need a set of people to work as Digital Champions in every organisation.
