Burkhard Hensel

BASont - A modular, adaptive building automation system ontology

Several ontologies exist that model aspects of home or building automation systems for specific use cases. However, no comprehensive approach exists, that models building automation systems in a modular way to be usable in various use cases and tools. The paper proposes the BASont that addresses various use cases over the life cycle of a building automation system from design, to commissioning, to operation, and refurbishment. The use of the BASont is demonstrated for a data retrieval and self-commissioning use case.

An adaptive PI controller for room temperature control with level-crossing sampling

Event-based sampling allows saving energy in the sensor transmitter by avoiding unnecessary messages. One important application is room temperature control with wireless sensors. Optimizing the controller parameters of a PI controller for this application is a difficult task, because usually no process model is available and challenging issues like actuator saturation have to be taken into account. Adaptive controllers offer the possibility to tune themselves automatically. In this paper, an adaptive PI controller based on pattern recognition is proposed, designed for room temperature control, sensor energy efficiency, and level-crossing sampling. The implementation is much easier than that of most other adaptive controllers and robustness to disturbances and noise is high. The focus of this paper lies rather on the basic idea, simulations and practical issues than on theoretical investigations.

A simple PI controller tuning rule for sensor energy efficiency with level-crossing sampling

Event-based sampling is currently an important topic in research and practice, because it allows saving energy in the sensor transmitter by avoiding unnecessary messages. The main contribution of this paper is a simple PI controller tuning rule adapted to the usage of level-crossing sampling with special focus on energy efficiency. The tuning rule uses a FOLPD (first order lag plus time delay) process model. It is shown that it is advantageous for sensor energy efficiency to tune the controller in a way which results in less overshoot than in periodic or continuous-time control. This result can also be applied to other types of controllers.

Wireless, collaborative virtual sensors for thermal comfort

Monitoring building performance data for energy-efficiency is an increasing market in which especially wireless sensor networks benefit from their easy installation in existing buildings. Nonetheless, large scale installations are still hampered by sheer installation costs. This paper introduces a virtual sensor approach that utilizes available information in a collaborative network to compute new sensor data. An example is provided in which the approach estimates the thermal comfort with only low-cost temperature sensors in most rooms.

Adaptive controllers for level-crossing sampling: Conditions and comparison

When wireless sensors are used for control purposes (e. g. room temperature control), energy efficiency of the sensors is of great interest, because the energy must come from a (limited) battery or from energy harvesting. Since message transmission is the most critical point regarding energy efficiency, event-based sampling is a typical method for reducing the message rate. Levelcrossing sampling, also known as send-on-delta sampling, is the most popular type of event-based sampling. Some technical processes are time-variant. The controller parameters should then be updated regularly to reach good control performance. Adaptive controllers are able to estimate the current process parameters and optimize the controller parameters automatically. Adaptive control and event-based sampling have hardly been examined together. One focus of this paper is the analysis of preconditions which must be fulfilled to apply adaptive controllers in conjunction with level-crossing (or event-based) sampling sampling. Further, performance measures are discussed which are the basis for a qualitative comparison of such controllers. This paper finally compares two adaptive control strategies with respect to level-crossing sampling.

Control of a Vacuum Coating Process with Long Dead-Time and an Integrator: a Case Study

Abstract This paper presents the comparative analysis of different methods for the control of a characteristical industrial process with a long dead-time. Not only traditional control strategies (Smith predictor, Model-Following Controller and Generalized Predictive Control) are considered, but also an alternative approach - the use of virtual sensors.

Generator for modular virtual sensors

Virtual sensors are “software simulations” of real sensors based on measurements of other sensors and a model. This paper presents a new design software which supports the creation and parametrization of modular virtual sensors, i.e. virtual sensors consisting of several “subsystems”. The workflow, the core concepts, the plug-in interfaces for user-specific extensions, and an example are presented. The main highlights of the proposed software are the possibilities to automatically invert (sub)systems-allowing efficient parameter estimation-, and to automatically generate several proposals for virtual sensors out of a larger physical model.

Holistic Ontology-Based Assistance System for Efficient Process Model Parameter Identification

Accurate models of technical systems are the basis for many tasks like system analysis, predictions, or controller design. Usually, the values of several important parameters cannot be determined by theoretical analysis only; instead, process identification is necessary. For several applications, the efficiency of the identification procedure is very important, for example, for the creation of thermal models of machine tools, because of the large time constants and the expensive machine time. The goal of the authors is the support of this task as far as possible by software. This paper contributes to that goal twofold: on the one hand, it provides a collection of influences which have to be considered for supporting the identification procedure. On the other hand, concepts for computer-based support are presented—ontologies and automatic design methods based on evolutionary algorithms.

Virtual sensor for calibration of thermal models of machine tools

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.

Economic and technical influences on feedback controller design

There are many influences on practical feedback controller design, ranging from control performance requirements over an economic view on hardware, software, and personnel costs to technical conditions like limits of the actuators. This article gives a broad overview on these influences and discusses approaches to deal with these problems. The goal of this paper is to be a help when analyzing the variety of requirements for a concrete application and to appoint available resources (money and personnel) more efficiently.