ArtificialIntelligence-BasedDetectionSystemForHazardousLiquidMetal.pptx

AI based liquid detection

1. ARTIFICIAL INTELLIGENCE-BASED DETECTION SYSTEM FOR HAZARDOUS LIQUID METAL FIRE Praveen Sankarasubramanian Research Scholar, Computer Science, and Engineering Vels Institute of Science, Technology & Advanced Studies Chennai, India [email protected] Dr. E.N. Ganesh DEAN, School of Engineering Vels Institute of Science, Technology & Advanced Studies Chennai, India [email protected]

2. INTRODUCTION USE OF LIQUID METALS, CHEMICALS, TOXIC GASSES ARE COMMON IN INDUSTRIES. CORROSION AND PRESSURE CAN DETERIORATE THE STRUCTURE HANDLING LIQUID METALS, CHEMICAL CARRYING STRUCTURE. LEAKAGE CAUSES ECOLOGICAL DISASTER. CONTINUOUS MONITORING AND TIMELY DETECTION OF RISK REDUCES THE IMPACT.

3. INTRODUCTION IN THE YEAR 1995 , AT THE MONJU NUCLEAR POWER PLANT, FIG. 1 A THERMOWELL INSIDE THE PIPE CARRYING SODIUM COOLANT BROKE DUE TO THE INTENSE VIBRATION. THE ROOT CAUSE OF THE INCIDENT WAS A DEFECTIVE WELD POINT. SEVERAL KILOGRAMS OF SODIUM LIQUID LEAKED. L IQUID SODIUM REACTED WITH AIR AND MOISTURE RAPIDLY . IT PRODUCED AN ENORMOUS AMOUNT OF HEAT AND FILLED THE ROOM WITH CAUSTIC FUMES. DUE TO THIS, THE REACTOR STOP PED FUNCTIONING.

4. SYSTEM OVERVIEW

5. SYSTEM OVERVIEW- FIXED UNIT

6. SYSTEM OVERVIEW – MOVING UNIT

7. SYSTEM OVERVIEW- SENSOR MANAGEMENT MODULE

8. SYSTEM OVERVIEW – DATA ANALYTICS MODULE

9. SYSTEM OVERVIEW – PROCESSING MODULE

10. PROPOSED METHODOLOGY

11. RESULTS SUCCESS IN DETECTING AREA AND MOVEMENT OF FIRE. THE FRAMEWORK DETECTED 98.046875% OF FIRE IMAGE AND 95.6687898089172% ACCURACY IN DETECTING THE SMOKE IMAGES.

12. CONCLUSION