M. Neumayer

Accelerated Bayesian Inference for the Estimation of Spatially Varying Heat Flux in a Heat Conduction Problem

This article aims at the acceleration of an inverse heat transfer problem solution within the Bayesian framework. The physical problem involves a spatially varying heat flux, which can reach very large magnitudes in small regions, such as in the heating imposed by high-power lasers. The inverse problem of estimating the imposed heat flux is solved by using the Markov chain Monte Carlo method, with simulated transient temperature measurements. The solution of the inverse problem is based on a reduced model, which consists of an improved lumped formulation of a linearized version of the original nonlinear problem. Two different priors are considered for the sought heat flux, including a total variation density and a Gaussian density. The Gaussian prior is based on the physics of the heat conduction problem. Parameters appearing in both priors are also estimated as part of the inference problem in hyperprior models. The Delayed Acceptance Metropolis-Hastings (DAMH) Algorithm and the Enhanced Approximation Error Model (AEM) are applied with the objective to improve the accuracy of the inverse problem solution.

Combined Capacitive and Ultrasonic Distance Measurement for Automotive Applications

This work presents a sensor fusion concept based on capacitive and ultrasonic techniques for distance measurement in automotive applications. Although ultrasonic sensors are a well accepted technology for distance sensing applications, they reveal drawbacks in the immediate vicinity of a vehicle. Capacitive sensors are suited for distance measurements of up to 0.3 m and may also provide information about the approaching object itself. The measurement range of the proposed fusion concept reaches up to 2 m whereby blind spots are avoided and means for object classification are provided. The feasibility of the approach and its robustness against environmental influences are demonstrated by means of experimental investigations.

Current Reconstruction Algorithms in Electrical Capacitance Tomography

Figure 1 depicts a scheme of an electrical capacitance tomography (ECT) sensor. A number of electrodes are mounted on the exterior of a nonconductive process pipe. By measurements of the capacitances between certain electrode, it is aim to compute an image of the material distribution.

Robust sensing of human proximity for safety applications

This paper presents a robust capacitive proximity sensor that facilitates the safety features of power tools. Power tools (e.g. pneumatic metal forming machines) enable easier and faster work but they can also cause severe injuries, if not operated carefully. However, accidents and related injuries can be prevented to a large extent by providing an automatic shut-off functionality. In order to realize such a functionality, a sensing mechanism which detects the presence of a human in the dangerous working areas of the power tool is necessary. A capacitive sensor which is suitable for this application has been developed. In most of the cases, the tools are made of metallic parts and have good electrical conductivity. Typically, the tools have a fixed part and a movable part. In the proposed scheme, the proximity sensor is attached to the movable part of the tool. Depending on the position of the movable part, changes of the electric field distribution across the electrodes will occur. Hence, comparatively large changes of the sensor capacitance are introduced compared to a capacitance change for the presence of a human hand. The current position of the movable part and the prior knowledge of the capacitance values for a vacant condition are used to solve this issue. The practicality of the proposed sensor system is demonstrated with detailed simulation studies and results from the measurements conducted with a prototype sensor installed on a typical metal forming machine.

A Pretouch Sensing System for a Robot Grasper Using Magnetic and Capacitive Sensors

Pretouch sensors are capable to classify objects and estimate their position prior to touching and thus close the gap between vision- and contact-based sensing. This will be particularly useful for robotics applications not only just for manipulation of objects but also with respect to safety. As robots will more and more operate in “open environments” where there is little prior knowledge, it will be important to gather as much information on the environment as possible. However, although there are many measurement principles that might be applied, only a few can cope with the requirements, e.g., limitations with respect to spatial dimensions, weight, and power consumption. In this paper, we investigate a measurement system for two types of materials. Dielectric and ferromagnetic materials, which are common in many industrial applications, can be located and distinguished in the vicinity of a robot grasper. Inspired by magnetic field tomography, we use a permanent magnet and apply giant magnetic resistor sensors to measure the magnetic field deformation caused by ferromagnetic objects. Furthermore, we use an electrical capacitance tomography approach to measure the change of the electric field by dielectric objects. Based on the measurement results, we solve an inverse problem with respect to the object position and spatial permittivity distribution. We present experimental results for a prototype implementation and provide a description of the calibration method.

A novel sensor fusion concept for distance measurement in automotive applications

This paper presents a sensor fusion concept based on capacitive and ultrasonic techniques for distance measurement in automotive applications. Although ultrasonic sensors are a well accepted technology for distance sensing applications, they reveal drawbacks in the immediate vicinity of a vehicle. Capacitive sensors are suited for distance measurements of up to 0.3 m and may also provide information about the approaching object itself. The measurement range of the proposed fusion concept reaches up to 2 m whereby blind spots are avoided and means for object classification are provided. The feasibility of the approach and its robustness against environmental influences are demonstrated by means of experimental investigations.

Model based monitoring of ice accretion on overhead power lines

Atmospheric ice accretion is a natural phenomenon in cold climate regions. On overhead transmission lines, the ice load can cause severe damage and can cause critical situations for those relying on this infrastructure. Even fatal casualties have been documented due to the lapse of transmission lines. Capacitive sensing has been reported a feasible approach for detecting ice accretion events and autonomous sensor systems have been presented. The automated extracting of information about the state of ice accretion, e.g. the layer thickness, from capacitive measurement readings is a vital part for the usability of such systems. In this paper we report on a model based approach for estimating ice accretion parameters from capacitive sensor readings. We identify critical input parameters and present the design of a surrogate model for a capacitive ice sensor. For this we present a test rig to generate ice in a climate chamber. We finally demonstrate the estimation of ice thickness from measurements.

De-icing system with integrated ice detection and temperature sensing for meteorological devices

We present a de-icing system indented for application with meteorological sensors in harsh environments. The heating unit is implemented as a double-meander structure on a printed circuit board. The double-meander structure allows for measuring capacitance in the differential mode. The presence of ice (and water) leads to an increase of the capacitance and can thus be detected. Additionally, the temperature of the heating unit can be obtained from the temperature dependent variation of the ohmic resistance of the heating structure. With this information the heating unit can be controlled efficiently as excessive heating or heating in the absence of ice can be avoided.

Estimation of a 4-port scatter matrix from 2-port measurements

Testing of electrical components is of vital interest for the industry and has therefore become a main research area in instrumentation and measurement. The characterization spreads from DC measurements up to measurements at RF frequencies. For the higher frequency range scatter parameter measurements using network analyzers (NA) are widely used to characterize the components. However, most NAs are 2-port NAs making the characterization of a n-port time consuming as all ports have to be connected in the course of the measurement process. In this paper we present a methodology to estimate the full scatter matrix of a passive 4-port by means of local 2-port measurements only.

Electrical Capacitance Tomography: Current sensors/algorithms and future advances

Electrical Capacitance Tomography (ECT) forms an inverse problem aiming on the determination of material specifics x in the domain ΩROI from capacitance measurements d̃ taken at the boundary ∂ΩROI of the region of interest (ROI). In this paper we give an overview about the current state of the art of ECT for process tomography. This includes aspects about instrumentation and reconstruction algorithms. Further we will give an outlook about current developments, future advances and applications for ECT techniques.