Relief valve leak detection

Pressure Relief Valves (PRVs) are used extensively in oil refineries, chemical plants, pipelines, power plants, and other industrial applications as part of pressure-relieving systems to protect piping, boilers, vessels, and other pressurized assets from dangerous over-pressure conditions. To help users ensure that these systems are supplied in accordance with recognized good engineering practices, a number of industry-specific standards exist. For example, American Petroleum Institute standards API 521 (ISO 23251), API 526, and RP 520 all deal specifically with PRVs and pressure-relieving systems as used in the petroleum industries.

PRVs are particularly susceptible to leaks and the consequences of such leaks can be serious:

• Dangerous operating conditions, increasing the risk to plant and personnel
  
• Damage to the environment
  
• Regulatory fines/penalties
  
• Negative public image/reputation
  
• Reduced process efficiency and profitability
  
• Loss of raw, intermediate, or finished materials

While most companies expend considerable resources to detect and mitigate leaks, usually in the form of routine inspection or observation regimens, current approaches have not proven entirely effective, and PRV leaks remain a significant challenge for industry. Additionally, PRVs are frequently subjected to unneces¬sary maintenance. This occurs when a leak cannot be confirmed, but to err on the side of caution, preventive repairs, replacements, or overhauls are carried out “just in case.” This can drive the relative maintenance costs for PRVs very high.

Cost-effective and reliable online technology is now available to monitor critical PRVs for the presence of leaks. An online approach means that leaks are caught within minutes, incident severity (and associated consequences) is minimized, and costly routine inspection regimens are eliminated.

GE’s PRV Leak Detection System uses the Bently Nevada Trendmaster™ Pro architecture to gather periodic, online temperature and acoustic measurements and automatically analyze them in System 1™ software. By using both temperature and acoustic measurements in complementary fashion, more reliable detection is available than when using either technology alone. These measurements are then processed through a special RulePak in System 1 software’s decision support engine. This RulePak contains sophisticated, embedded knowledge of PRV leak behavior, automatically detecting the presence of a leak.