Black box modeling of PIDs implemented in PLCs without structural information: a support vector regression approach

Abstract

In this report, the parameters identification of a proportional–integral–derivative (PID) algorithm implemented in a programmable logic controller (PLC) using support vector regression (SVR) is presented. This report focuses on a black box model of the PID with additional functions and modifications provided by the manufacturers and without information on the exact structure. The process of feature selection and its impact on the training and testing abilities are emphasized. The method was tested on a real PLC (Siemens and General Electric) with the implemented PID. The results show that the SVR maps the function of the PID algorithms and the modifications introduced by the manufacturer of the PLC with high accuracy. With this approach, the simulation results can be directly used to tune the PID algorithms in the PLC. The method is sufficiently universal in that it can be applied to any PI or PID algorithm implemented in the PLC with additional functions and modifications that were previously considered to be trade secrets. This method can also be an alternative for engineers who need to tune the PID and do not have any such information on the structure and cannot use the default settings for the known structures.

Wide-Range Measurement of Thermal Preference—A Novel Method for Detecting Analgesics Reducing Thermally-Evoked Pain in Mice

“…The two-plate analgesiameter for freely moving mice was used in this study ( Figure 2 ). This prototype device is our own solution, which has several advantages as compared to classical analgesiameters, such as: (1) measuring only thermally-evoked pain, (2) controlling current temperature on each plate of the device using advanced proportional-integral-derivative (PID) algorithms implemented in PLC S7-1200 controller (Siemens, Germany) [ 32 ], which allows the mouse to choose the preferred thermal zone, without being forced by the researcher, (3) rapid and precise regulation of the temperature set-point on each of the two plates of this device enables to establish separate “thermal zones” randomly in each testing session, which significantly limits potential false positive results due to animals’ ability to learn choosing the preferred thermal zone of the device (i.e., a preferred plate) based on external visual cues, (4) the possibility of setting narrow-range thermal zones on each plate, which increases sensitivity of the method; at the same time this allows to measure the effect of a drug on hyperalgesia and allodynia, (5) the measurements are collected from unrestrained animals and therefore this method is less stressful for animals as compared to some standard thermal pain tests (e.g., cold water test and cold plate test), thus offering more reliable data due to a reduction of stress-related behavioral reflex responses of experimental animals (e.g., stress-induced analgesia) [ 33 ] and (6) higher objectivity of measurements (i.e., observer-independent assessment) because time spent by the animal in a given thermal zone and trajectories are recorded by the video camera (GoPro Hero7 Black, San Mateo, CA, USA) and subsequently they are analyzed with the use of mathematical methods of image analysis (deep learning and machine learning) [ 34 , 35 ]. Of note, the camera used for the detection of animals’ movements enables one to assess the effect of test drugs on locomotor activity of mice and this allows one to exclude potential false positive results in the pain test that are due to decreased locomotor activity [ 31 ].…”

Impact of feature selection on system identification by means of NARX-SVM

“…In our research we used the methods of regressor selection described in [5,7] together with the modifications shown in this paper. The main task of the system is to maintain the temperature of the aluminium plate within the range of 0-50°C.…”

“…We have optimised the SVM parameters (i.e. C and γ) according to the procedure described in detail in [7]. The evaluation of model fitness to data measured was performed using the fit index, calculated pursuant to the formula (a higher value means a better model quality)…”

Industrial Application of Deep Neural Network for Aluminum Casting Defect Detection in Case of Unbalanced Dataset

“…The use of artificial intelligence methods finds practical application in many different scientific and industrial fields such as electrical engineering [1], mechanical engineering [2,3], materials engineering [4], environmental engineering [5], and many more. Artificial intelligence methods, and machine learning in particular, are also indicated as an important element of the popular Industry 4.0 concept [6].…”