With the trend of increasing sensors implementation in production systems and comprehensive networking, essential preconditions are becoming required to be established for the successful application of data-driven methods of equipment monitoring, process optimization, and other relevant automation tasks. As a protocol, these tasks should be performed by engineers. Engineers usually do not have enough experience with data mining or machine learning techniques and are often skeptical about the world of artificial intelligence (AI). Quality assurance of AI results and transparency throughout the IT chain are essential for the acceptance and low-risk dissemination of AI applications in production and automation technology. This article presents a conceptual method of the stepwise and level-wise control and improvement of data quality as one of the most important sources of AI failures. The appropriate process model (V-model for quality assurance) forms the basis for this.

Process identification, i.e. finding static, dynamic or signal models, requires good knowledge in the application of statistical analysis, signal and system theory. The main goal of this article is to prepare the process identification technology for the process analyst in a clear and understandable way. A new philosophy of usability, process analysis, visualization, consulting and memory function at the example of the ADM tool, which shows a method-integrated approach, is introduced.

Machine tools are important parts of high-complex industrial manufacturing. Thus, the end product quality strictly depends on the accuracy of these machines, but they are prone to deformation caused by their own heat. The deformation needs to be compensated in order to assure accurate production. So an adequate model of the high-dimensional thermal deformation process must be created and parameters of this model must be evaluated. Unfortunately, such parameters are often unknown and cannot be calculated a priori. Parameter identification during real experiments is not an option for these models because of its high engineering and machine time effort. The installation of additional sensors to measure these parameters directly is uneconomical. Instead, an effective calibration of thermal models can be reached by combining real and virtual measurements on a machine tool during its real operation, without additional sensors installation. In this paper, a new approach for thermal model calibration is presented. The expected results are very promising and can be recommended as an effective solution for this class of problems.

In this paper, a Smith predictor control scheme is compared with a solution on the basis of a simple PI controller with a virtual sensor. Virtual sensor eliminates the technology-related dead-time and makes the control of dead-time processes possible without high development costs. The effectiveness of the proposed scheme will be demonstrated by simulations of real process with an integrator and long dead-time.

This paper presents a new method of virtual measurement of process parameters in the dead-time systems using the virtual adaptive sensor. The method's goal is elimination of the technology-related signal dead-time in automation systems for optimal process control. The additional goal is to make the real-time control possible also in complex processes with dead-time behavior without high development costs. A vacuum coating equipment is considered as an example where the coat thickness is measured by the X-ray fluorescent (XRF) method. It is characterized by significant dead-time that doesn't allow to realize qualitative process control. In order to choose a better strategy to develop the virtual adaptive sensor, the automation process was identified. The simulation of the virtual adaptive sensor on the basis of Artificial Neural Network has shown promising results.

Model identification and controller design for processes with long dead times requires the availability of tools which support static, dynamic and time series analysis as well as data preparation methods (e.g. dead time correction, filtering algorithms, etc.). ADM, a tool for process identification yields this functionality introducing a new philosophy of usability. In this article this philosophy is described using the example of identifying a process model for an electron beam physical vapor deposition (EB-PVD) coating process of metall stripes with a long dead time. Furthermore a controller design for this process is presented.