ADAPTIVE SIGNAL DISCRIMINATION AS APPLIED TO COAL INTERFACE DETECTION.pdf

ADAPTIVE SIGNAL DISCRIMINATION AS APPLIED TO COAL INTERFACE DETECTION D R . G A R Y L. MOWREY Pittsburgh Research Center Bureau of Mines U.S. Department of the In te r ior Pittsburgh, PA Abstract: The Bureau of Mines is becoming increasfngly involved i n the areas of automation and robot ics for the U.S. mining i n d u s t r y t o increase both mining productivity and safety. Both robot izat ion of ex is t ing mining equipment and fu ture integrated robot ic and/or automated mining systems a r e being invest igated. One of the primary concerns associated with automated underground coal mining is the development of an appropriate coal-rock in te r face detect ion ( C I D ) sensing system t o determine when a mining machine is mining coal or s t a r t i n g t o mine adjacent s t r a t a (e .g . , roof or f l o o r ) . Several promising C I D techniques a re current ly being investigated by the Bureau including strata-machine vibrat ion, natural gamma rad ia t ion , x-ray fluorescence, ground-probing radar , and thermography. This paper describes two C I D methods associated with vibrat ions generated by mining machines and how adaptive s ignal discrimination (ASD) technology is being used t o process these s ignals . I n these C I D methods, s t r a t a ( roof , coal seam. f l o o r ) o r mining machine vibrat ions a r e monitored for complex s igna ls generated, which vary according t o the type of geologic material being cut and the type of mining machine being used. These s igna ls a re then analyzed using sophis t icated s ta te-of- the-ar t ASD technology. I t is t h i s advanced s ignal analysis tha t dis t inguishes t h i s approach from those of pr ior research. The ASD system is i n i t i a l l y trainedT7 using a database of fea tures extracted from known s igna ls measured under conditions of i n t e r e s t (e .g . , cut t ing coal , I 1 s ta r t ing t o cut