BLASTDAR – A Large Radar Sensor Array System for Blast Furnace Burden Surface Imaging

In this paper, we present a radar sensor system for real-time blast furnace burden surface imaging inside a fully operative blast furnace, called BLASTDAR, the blast furnace radar. The designed frequency-modulated continuous-wave (FMCW) radar sensor array operates in the frequency band around 77 GHz and consists of several non-uniformly spaced receive and transmit antennas, making it a multiple-input multiple-output (MIMO) radar system with large aperture. Mechanical steering is replaced by digital array processing techniques. Off-the-shelf automotive-qualified multi-channel monolithic microwave integrated circuits (MMICs) are used. By means of this configuration a virtual antenna array with 256 elements was developed that guarantees the desired angular resolution of better than 3 deg., and a range resolution of about 15 cm. Based on the singlechannel FMCW signal model this paper will derive a multichannel signal model in combination with a digital beamforming approach and further advanced signal processing algorithms. The implementation of a simulation tool covering the whole design process is shown. Based on these simulation results a system configuration is chosen and the obtained setup is defined and presented. A description of the manufactured cost-efficient radio frequency (RF) and baseband boards together with the housing design shows the practical implementation of the sensor. For the system calibration two different methods are listed and compared regarding their performance. Verification measurements confirm the predicted performance of the developed sensor. Several measurements inside a fully operational blast furnace demonstrate the proper long-term functionality of the system, to the best knowledge of the authors, for the first time worldwide. It is in continuous operation since about two years in blast furnace #5 of voestalpine Stahl GmbH, Linz.