دانلود مقاله FLOW MEASUREMENT AND THE SPRT STATISTICAL LEAK DETECTION FINDS SMALLER LEAKS WITHOUT FALSE ALARMS
سال انتشار: ۱۳۸۶
محل انتشار: اولین کنفرانس لوله و صنایع وابسته
تعداد صفحات: ۷
Brian Twomey – ATOMS Contracts L.td 57/63 Live wall road gibratar
Michael Twomey –
John Lewis –
The online-learning capabilities of ATMOS LDS make it the most successful and reliable leak detection system working with all varieties of flow meters; even clamp-on ultrasonic flow meters (sometimes such meters experience repeatability issues when product density changes). The unique online-learning capabilities of the ATMOS LDS allow it to handle the flow difference discrepancies sometimes experienced by certain flow meters, e.g. clamp-on ultrasonic flow meters.There are advantages to using ATMOS™ LDS, the extremely successful statistical pipeline leak detection system, in conjunction with true mass flow measurement on multi-product pipelines or on pipelines with crude batches of varying densities to increase reliability and find smaller leaks without false alarms under all operating conditions. ATMOS™ LDS has detected over 400 leaks under steady-state, transient and shut-in conditions on both gas and liquid pipelines. The paper discusses how the poor repeatability from some ultrasonic flow meters reduced the sensitivity of leak detection systems. Ultrasonic flow meters typically use time-of-flight techniques to measure the time taken for the sonic wave to travel between the transmitter and receiver and then use a look-up table to calculate the fluid velocity. Unfortunately, single or dual path ultrasonic meters do not measure the actual speed of sound in the fluid and generate inconsistent flow readings when density or flow rate changes. The paper discusses how the poor repeatability from some ultrasonic flow meters can reduce the sensitivity of leak detection systems. Fore example, ultrasonic flow meters typically use time-of-flight techniques to identify the density of a product in a pipeline and then use a look-up table to correct the volumetric flow rate. Unfortunately, experience has shown that some ultrasonic meters behave inconsistently. It was found that changes in density and flow rates resulted in changes in flow difference values between the inlet and the outlet. This poor repeatability alters the flow difference between the inlet and outlet flow meters as the product density changes. If this change in flow difference is greater than the minimum detectable leak, false leak alarms may be generated. Thus, poor repeatability can force an increase in the minimum leak size (reducing sensitivity) to avoid false alarms. Only ATMOS LDS has any hope of performing leak detection with such meter by virtue of the unique on-line learning ability of the software. These flow difference changes have been found in both clamps-on and in-line ultrasonic flow meters. For this reason, it is critical that the pipeline operator evaluates each application in the design stage to determine which flow meter is suitable. The flow meter vendor should be held accountable for making the flow meters meet their specified repeatability and accuracy in the field for the complete range of products transported. ATMOS LDS has achieved excellent sensitivity in leak tests on pipelines where the ultrasonic flow meters were installed and set-up correctly. The minimum leak detected in one leak test was 0.25% of flow on a pipeline that uses 5-path in-line ultrasonic flow meters.