April 2018

Environment and Safety

How pipeline management becomes part of a plant safety strategy

A typical refinery or chemical plant contains miles of pipelines.

Liston, C., HIMA Americas, Inc.

A typical refinery or chemical plant contains miles of pipelines. Identifying and correcting leaks in these pipelines can be problematic. The potential cost of remediation and possible financial penalties due to pipeline leaks in plants has increased, along with the threat of damage to the operator’s reputation from the possible negative environmental impact (FIG. 1).

FIG. 1. Leak incident cost escalation.

Operators have low confidence in commercially available leak detection systems, so they often delay action when an alarm is received from a leak detection system, risking both financial resources and corporate reputation. Fortunately, the reliability of detection systems has improved. These systems have been adopted into an innovative new approach, which is a hybrid leak detection solution based on a safety integrity level 3 (SIL 3) safety controller. By tightly integrating safety-related hardware and monitoring software, the solution ensures the long-term availability and safety of a plant’s pipelines. It also contributes to reducing the adverse consequences of incidents, improves safety by minimizing the size of leaks and protects the reputation of the plant and its operator.

The need for standards

Continued improvement of leak detection is necessary since hundreds of pipeline incidents are reported every year, and these numbers have not seen improvement over the last 15 yr. The leading causes of these incidents are material defects, followed by corrosion, excavation damage and incorrect operation—although excavation has figured more significantly in past years. The American Petroleum Institute (API) has released a set of standards to be used as guidelines for operators to help mitigate the risks and consequences of leaks. These standards include:

  • API 1160—Overall standard to cover all pipeline integrity management
  • API 1130—Design and implementation of leak detection systems
  • API 1149—Theoretical calculations of possible leak detection system performance
  • API 1175—Standard for the selection, operation, maintenance and continuous improvement of leak detection systems.

What to look for

Many operators do not know what to look for when searching for a reliable leak detection system. A combination of factors exist that must be evaluated to find the best solution. The four primary factors include:

  • Sensitivity—A combination of the size of a detectable leak and the time required to detect it.
  • Reliability—A measure of the system’s ability to accurately assess whether a leak exists or not.
  • Accuracy—The ability of a system to estimate leak parameters such as leak flowrate, total volume lost and leak location.
  • Robustness—The ability of a system to continue to function during unusual hydraulic conditions or when data is compromised.

The evolution of leak detection

The industry’s response to these standards, and to leak detection overall, has been evolutionary. This evolution can be represented as a set of four steps. The first step—which will be referred to as Leak Detection 1.0—uses a single method chosen from the many available. It is worth reviewing these alternatives, as they are not only used for Leak Detection 1.0, but also provide the building blocks for the more recent evolutionary stages. The detection methods must first be classified by whether they are externally or internally based. The external methods discussed are acoustic sensor, fiber optic cable and vapor sensor.

Acoustic sensors are distributed along the pipeline to detect internal noise levels. Any leak produces a low-frequency acoustic noise at its location, which the sensors can detect. This method’s advantage is its sensitivity to small leaks, while its disadvantages include a high number of false alarms, which can be caused by vehicles, valves or pumps, and the fact that the method’s efficiency and accuracy is dependent on the operator’s skill level. 

Fiber-optic leak sensing depends on installing a fiber optic cable along the entire pipeline length. This cable monitors continuously for temperature changes caused by pipeline leaks. This method’s advantages are high leak location accuracy and theft identification. However, it has a high installation cost, leak identification times can be slow, stability over time is unproven and the entire pipeline must be excavated to install the cable. Additionally, the method does not yield any leak size data.

The last external method uses a vapor sensor. A vapor-sensing tube, installed along the entire length of the pipeline, contains air moving at a constant speed toward a sensor at the end of the pipeline. During a scan, an electrolysis cell emits a test peak of hydrogen. If vapor from a leak is detected, the system will calculate where the leak is based on timing differences between the vapor peak and hydrogen peak arrivals at the sensor. Leak location and size accuracy is high, but the installation price is also steep. Scanning is performed only once or twice a day rather than continuously, so a leak could become extremely large by the time it is detected.

Internal methods

Five internal or computational pipeline monitoring (CPM) methods are available. The first is a statistical analysis method that relies on the pipeline pressure and flow profiles reacting to a leak in a typical manner. These profile reactions can be calculated by using the correlation between inlet and outlet flow, as well as inlet and outlet pressure. Unfortunately, without a steady-state condition, this correlation does not exist. This means that the method does not work in transient conditions, and leak location tends to be of low accuracy, improving only as the leak continues. However, this method does have the advantage of utilizing existing instrumentation.

Real-time transient modeling (RTTM) uses basic physical laws, such as conservation of mass, conservation of momentum and conservation of energy, to create mathematical models of the flow within the pipeline. The pressure and flow profiles are calculated in time steps. When the measured flow deviates from the model, a leak is identified. To design a reliable system with minimal false alarms, the noise level should be continuously inspected to modify the models. RTTM works very well in transient conditions, and can potentially use existing flow, pressure, temperature and density instrumentation. However, it is very expensive to program, and continuous tuning is required. Training costs for operators to tune the system must be allocated, and it is not always possible to obtain all of the parameters necessary for programming.

The volume balance method is based on the principle of the conservation of mass; what goes in must come out, unless there is a leak. This method is also used in some supervisory control and data acquisition (SCADA) systems. The compensated volume balance variant is the optimum variant to use in a leak detection system, as this optimizes its functionality. This version of the method accounts for changes in both pressure and temperature. Rising temperatures result in expansion, and building pressures cause compression. This method uses proven technology and algorithms, utilizes existing instrumentation with minimal programming, and remains effective in transient conditions. However, it can estimate only the leak location. 

The pressure drop method is a simple approach that uses existing instrumentation; during shutdown conditions, a pressure drop indicates a leak. This method can detect the smallest of leaks—known as seepages. However, it too can estimate only the leak location. 

The negative pressure wave method works on the principle that as a leak occurs, it generates a negative pressure wave of a known velocity both upstream and downstream of the leak. The leak location can be calculated by comparing the arrival times of the negative wave at each transmitter. The method utilizes existing instrumentation to provide extreme leak sensitivity and excellent leak location accuracy, with a reduction in false alarms.

Leak Detection 2.0 and 3.0

With the ‘building block’ detection methods in mind, Leak Detection 2.0 simultaneously uses multiple leak detection methods to provide comprehensive coverage. The approach is a combination of the best three internal leak detection methods: Enhanced pressure wave, compensated volume balance and the pressure drop method (FIG. 2). By applying these methodologies simultaneously, system availability can be ensured for all phases of the pipeline, while also significantly reducing false alarms. This approach reduces programming costs for the plant operator. Additionally, the system requires little, if any, tuning to compensate for changes in the physical properties of the pipeline, such as corrosion or debris buildup.

FIG. 2. Optimal leak detection methods.

The next evolutionary phase, Leak Detection 3.0, introduces the concept of emergency shutdown (ESD) action, as well as monitoring. It concerns detection of ruptures, which are more serious than leaks and must be handled accordingly. Typically, a leak is classified as a rupture if it reaches or exceeds approximately 30% of the pipeline flowrate, although the precise value is defined by each plant operator’s individual risk analysis. Rupture detection systems were created as standalone systems operating independently of the leak detection implementation, and are designed to shut down a pipeline in the event of a rupture. The leak percentage threshold at which the system reacts can be raised or lowered as necessary. These systems are used particularly in environmentally sensitive areas where a delayed response can be damaging to the environment.

A hybrid solution

Leak Detection 4.0 is an innovative approach that integrates the leak detection methodologies of Leak Detection 2.0 with an SIL 3 safety system with ESD capability. This method creates a hybrid solution to optimize a plant’s leak detection system (FIG. 3).

FIG. 3. A hybrid solution for advanced leak detection.

This solution can be installed as a complete pipeline management automation solution to help improve safety in refineries and chemical plants. The solution controls and regulates safety-related processes for uninterrupted operation over the plant’s entire lifecycle. It can continuously monitor pipelines, shut them down automatically in hazard situations and prevent or significantly reduce damage. This hybrid approach helps prevent false alarms, and allows unlimited alterations, modifications, extensions, improvements and even mandatory proof tests while the plant is in operation.

Takeaways

The use of a hybrid approach solution for pipeline management offers plant operators considerable benefits. Uninterrupted operation and maximum availability are ensured, while the system complies with present and upcoming global safety standards according to SIL 3. This type of system ensures maximum functional safety and extremely high reliability by automatically shutting down affected areas during critical situations. As a result, it cuts pipeline operating costs, significantly reduces false alarms and increases the profitability of installations. HP

NOTES

     a Refers to HIMA’s FLOWorX leak detection software

     Refers to HIMA’s HIMax system

The Author

Related Articles

From the Archive

Comments