Improving LNG transfer

The development of offshore LNG projects, whether for sheltered or harsh environments, represents a vehicle for progress consistent with the LNG market growth, and opens new perspectives and new challenges for the entire LNG industry. In order to ensure that these new LNG chains will operate correctly and at an optimum rate, the development of an efficient and safe LNG transfer system is a key link, one that is still challenging the actors involved in this field. Since there are several offshore LNG transfer solutions in development, cryogenic flexible hose based LNG transfer systems appear to be able to meet the LNG Industry requirements.

However, in order to demonstrate that a new technology will operate reliably within specific limits, LNG operators generally deem it necessary to have either the evidence of such performances through a structured qualification process, or to be convinced by significant scale trials in real conditions or close to the ones that the system will be used in.

To this end, a series of full-scale LNG trials were recently performed in the context of an exiting LNG receiving terminal. These trials involved use of a new 16-in. flexible hose-based LNG transfer system designed for onshore and offshore conditions. These trials were the ultimate stage of an industrial qualification of this transfer system, and they have been carried out on a unique dynamic test bench able to reproduce open sea conditions in terms of heave amplitude and acceleration.

Offshore LNG projects

The development of offshore LNG projects is partly based on the ability of LNG industry actors to bring innovative solutions capable of meeting the new technical requirements related to the marine environment. Whether for LNG export or receiving facilities, it appears that the “transfer system” link is a key element, whose ability to fulfill its role in open sea conditions must contribute to guarantee the availability and flexibility of these new offshore LNG chains.

GDF Suez has undertaken, in partnership with Technip France, KSB and Eurodim, the last stage of industrial qualification of a new LNG transfer system called Amplitude-LNG Loading System (ALLS). This system, based on flexible hose technology and suitable for offshore and onshore conditions, was developed through industrial partnerships involving major oil and gas companies and international certification organizations.

After completion of the technical certification of ALLS main equipment, the requirement to carry out full-scale dynamic testing, with circulation of LNG, was deemed necessary by the LNG operators to demonstrate the performances, operability and safety of this innovative LNG transfer system.

A structured qualification process, managed by Det Norske Veritas (DNV), was then engaged and a full-scale testing program developed with the aim to operate the transfer system in operational conditions with a facility able to reproduce open sea conditions. For this purpose, a dedicated dynamic test bench was built inside the Montoir-de-Bretagne LNG receiving terminal and connected to the installations. After the start up phase completion, a test campaign was engaged to reach the objectives of such a project.

The first phase at ambient temperature was successfully achieved in mid-2007 with the validation of the mechanical and safety performances of the system without gas. The second one with LNG was achieved in the beginning of 2008 and concluded the qualification program. The goal here is to describe the dynamic test bench used to qualify ALLS, and present the results obtained during the recent LNG trials with a focus on the operational procedures.

Qualification trial program

The test campaign had two main objectives: on the one hand to answer the qualification requirements raised during the qualification process managed by the certification authority; and on the other hand to demonstrate the capabilities of the transfer system in term of technical performances, operability, and safety. Based on these considerations, a complete program was developed to carry this out in real operational conditions.

The first phase at ambient temperature without gas validated the handling procedures for connection and disconnection with: 

• Static vertical offsets (tide effect) 
• Lateral induced torsion in the flexible hose (mid-ship manifold misalignment) 
• Dynamic oscillations and accelerations (swell effect in heave direction).

Combinations of the above different type of parameters were applied during these trials. Regarding the safety aspects, emergency disconnection trials were also carried out in static and dynamic conditions. The second phase of the program was designed to check and validate: 

• The global behavior of the system under repeated cryogenic cycles. 
• The ALLS operational procedures established specifically with respect to the transfer system main characteristics and LNG standard practices.

Planning management was essential during the LNG trials, since they started during the winter period, which implied higher constraints on the terminal.

Preliminary results

The results described here after are the ones obtained during the LNG trials. The technical data recorded from these trials on the transfer system performances is not addressed here, since these tests have been very only recently achieved. The numerous data are still under detailed analysis, so the information provided below is offered from an LNG operator point of view, on a qualitative basis. The results are given with respect to the operational sequences to be followed in the context of a loading / unloading operation. However, the LNG full flow and emergency disconnection trials are explained separately.

Handling and connection procedure

The main steps for the connection procedure were as follows:

Step 1. The transfer system was initially stored. The protective tap at the flexible hose end was removed. The ERS valves were closed and maintained in this position during the whole sequence.

Step 2. The flexible hose was lowered with the storage cable at a given cable length.

Step 3. The connection cable was paid in until the load was transferred from the storage cable to it (the connection cable load measurement is used to monitor this operation).

Step 4. The flexible hose was pulled up with the connection cable and guided to flange-to-flange contact smoothly with the help of the connector guiding devices. The quick connect-disconnect coupler (QCDC) clamps were then closed and the connection cable paid out in order that the entire system load rested on the QCDC clamps. Once the QCDC clamps closed, the connection phase was considered achieved. A leak test at 4 barg, with the ERS valves still closed, was then performed to check the flange-to-flange tightness (Figure 1).

During the trials, the possible wind effects on the flexible hose behavior were checked. It appeared that with wind speeds close to the operational limits for access to the structures (20 m/s), there were no added difficulties during the connection and disconnection maneuvers, even during load transfer between the cables.

Cooling down

The flexible hose cooling down cycles were performed without specific precautions. During the trials, once the LNG reached the top of the fixed tower, it overflowed into the flexible hose with a rate function of the LNG flow at the test bench inlet. It was thus possible to simulate the fast cooling down of the flexible hose with the presence or not of an LNG plug in the sag of the catenary (Figure 2). The internal corrugated pipe thickness being less than a millimeter, the cooling down of the flexible hose was fast, as can be seen in Figure 2 (right curve). Furthermore, after thermal equilibrium, there was no frost or ice on the outer sheet of the flexible hose (Figure 4).

Low LNG flow rate trials

LNG trials, based on the low flow circulation principle, were performed during three different LNG cycles. On these three trials, two of them were carried out with dynamic oscillations in order to check that the transfer system behavior was reproducible and its management (in terms of process) was repeatable. For this purpose, the trials were organized in order to perform the operations with a sequence as close as possible from the one followed today for LNG loading/unloading, that is to say: 

• Pre-cooling down of the installation 
• Purging of the installation 
• ESD tests 
• Configuration of the circuit to be in-line with the terminal one 
• Start of tank pump and opening of the transfer circuit.

In every case, no unexpected phenomenon of vibratory and/or water hammer types was observed nor measured. This included when the trial was performed with a LNG plug present, initially at the sag of the flexible hose catenary.

Purge

The principle of flexible hose purge is to drain the LNG remaining in the catenary by injection of driving gas. The aim of this operation is to evacuate the maximum of LNG in order to be able to begin the disconnection procedure. This statement is important, as this procedure principle is different from what is done today on the onshore facilities.

Several purging trials were carried out in different conditions to adjust process parameters and insure that the procedure is reproducible. Hereafter, its main principles and steps are described with regard to the test bench characteristics:

Step 1 (S1). The rigid cryogenic pipes were purged so that the LNG would remain in the flexible hose only. The flexible hose end temperature transmitters allows engineers to follow this step.

Step 2 (S2). Driving gas was injected by one flexible hose end (in our case by the oscillating arm). When the driving gas pressure was sufficient, LNG started to overflow from the opposite flexible hose end. This was observed by an abrupt change in the corresponding temperature curve, which is characteristic of liquid/solid thermal exchange. On the other side of the flexible hose, the pressure was still rising due to continuous driving gas injection that pushed the LNG plug from the catenary sag. The maximal pressure difference reached between the flexible hose ends was equivalent to the hydrostatic LNG pressure directly correlated to the height difference marked (“h” in Figure 4).

Step 3 (S3). Once the peak of pressure difference was reached, the driving gas started to bubble over/in the remaining LNG. At this time, the driving gas flow rate was raised continuously until a quasi-equilibrium between the flexible hose ends pressures was observed and a change in the outlet temperature curve was measured (this change of curvature is, this time, characteristic of a gas-solid thermal exchange).

Step 4 (S4). The driving gas injection could then be stopped at this step in order to start the disconnection procedure.

Disconnection

Even after the purging operations had been completed, some LNG remained in the sag of the flexible hose. To disconnect the transfer system, the following maneuvers were performed: 

• Nitrogen was injected inside the flexible hose from the flexible hose side where the connecting system is in order to inert the area to be opened (QCDC). This way of injection is compulsory since there is LNG remaining in the sag of the flexible hose. 
• After a continuous sweeping of nitrogen for a duration that depends on the piping characteristics, methane concentration measurements were carried out upstream and downstream the ERS valves. 
• Once the measurements concluded that the QCDC area was inerted, the ERS valves were closed and remained in this position until the end of the procedure. The LNG remaining in the flexible hose is thus isolated from the opening area. 
• The clamps were then opened and the flexible hose transferred to storage position. Protective taps were put on both QCDC and flexible hose flanges. 
• Once the flexible hose was stored and secured, gas was then injected to help the LNG remaining vaporization until the next connection operation.

This procedure appeared to be reproducible and safe for the operators whatever the amount of LNG remaining in the sag of the flexible hose.

Full LNG flow rate trial

As noted, the full LNG flow rate trial needed to use ship pumps during an unloading operation. In order to measure the transfer system performances and characteristics at different working flow rate, the trial specifications were: 

• Stable LNG flow at different rates for temperature, pressure and vibrational measurements. 
• Dynamic cycles for a significant duration at the maximal flow rate reached.

Since this trial took place during a commercial LNG unloading operation, it was necessary to work in cooperation with the ship owner, charter company, the ship crew and the technical manager of the ship, the operations teams of the terminal, and the insurance representatives of all the actors involved. These steps were taken in order to identify and mitigate the technical and financial risks related to the use of the ship for prototype trial. This preparatory work, in turn, required: 

• A navigability study for the selected ship with regard to the volume and distribution of LNG in its tanks at different time of the unloading operation. The conclusions allowed to precisely determined at what time of the unloading operation the trial would have to begin in order to guarantee, in case of ship pumps failure or malfunction, the navigability of the ship; 
• A preliminary trial during a ship unloading operation with the aim to raise the terminal pipe pressure while unloading the LNG.

The effective trial was performed in January 2008. The numerous data recorded are still under analysis to establish final results. However, from a qualitative point of view, the different flow rate steps can be seen in Figure 5.

Once the maximum LNG flow rate reached around 3000 m3/h, dynamic oscillations were applied to the system for one hour with parameters corresponding to minimum amplitude of 2.5 m with a 13-second period. During this trial, as previously said for low LNG flow rate, no unexpected phenomenon of vibratory and/or water hammer types was measured or visually observed even during transitory phases when a ship pump was started.

In addition, the maximum LNG flow rate reached during this trial was not correlated to the transfer system performances. Instead, it was related to a combination of added pressure drops to the existing 32-in. unloading circuit and ship pumping capabilities.

Emergency disconnection

An emergency disconnection trial was carried out with the flexible hose full of LNG but without flow during the ESD sequences. The objectives were to check: 

• Opening of the emergency release coupler (ERC) 
• Lowering of the flexible hose end, slowed down by the speed limiter 
• Transfer of the flexible hose end from the speed limiter cable onto the storage cable 
• Handling of the flexible hose full of LNG to storage position 
• Purging and inerting for ERC re-assembly.

As a reminder, two emergency disconnection trials were performed during the ambient temperature trials in static and dynamic conditions. These trials allowed for the validation of the speed limiter function, which is used to slow down the flexible hose lowering before the load transfer to the storage cable. The various steps of this trial are discussed below.

Opening of the ERC. As can be seen in the Figure 6 photos, the effective time needed for the system to be physically disconnected is around 1 second. During this phase, no contact between the different parts of the transfer system was observed.

Flexible hose lowering. The flexible hose lowering just after ERC opening is shown in Figure 7. The different steps of this phase of the emergency disconnection process can be detailed as follows:

Steps 1 and 2. ERC opening. The flexible hose end lowering is immediately slowed down by the speed limiter cable which is unrolled at a constant speed.

Step 3. The flexible hose continues to be lowered down slowly.

Step 4. The load of the transfer system is transferred from the speed limiter cable to the storage cable.

During the trial, the lateral movements of the flexible hose after the load transfer on the storage cable were monitored. At the end of this step, the flexible hose is then stored and secured before starting the purging process.

Purge. The purge of the flexible hose follows the same principles than the ones explained in case of purging before disconnection. It means that, thanks to dedicated connections just below the half ERS valve remaining on the flexible hose, gas is injected in order to push the LNG. This operation is again controlled by temperature and pressure measurements at both extremities of the flexible hose (Figure 8).

During this trial, before starting any operations, gas detectors were used to insure that the emergency release system valves on both parts of the transfer system were gas tight. After that, the purging process was carried out and globally the same behavior was observed and recorded. The time needed to realize the operation was a little bit longer due to the injection-piping diameter. Depending on the situation and operational constraints, the gas injection could be maintained as long as needed to vaporize a maximum of LNG.

Flexible hose inerting. In order to re-assemble the ERC, all the LNG present in the flexible hose had to be removed. For this purpose, the flexible hose was thus put on the ground (Figure 9). Thanks to other connections at the lower part of the connecting system, the flexible hose was linked to the terminal purge network. The LNG remaining in the flexible hose was purged by gas injection from the other flexible hose end. The connecting system was then been re-assembled. After a check list procedure to verify the operability of the system on the ground, the whole system was connected again and a 4 barg pressurization test was performed with no leak detected.

Summary of the LNG trials

Table 1 gives an overview of the LNG trials successfully performed during the qualification campaign of the ALLS transfer system.

Conclusion

A GDF Suez team, specifically trained to operate this unique dynamic test bench and LNG transfer system, successfully conducted a test campaign with LNG that enabled the team to confirm the following points: 

• Operability of the ALLS transfer system in LNG conditions. 
• Safety procedures of the transfer system, in particular achieving an LNG emergency disconnection. 
• Technical performances in real LNG transfer conditions with a maximal flow rate reached around 3000 m3/h. This stable operating point allowed the team to record vibrations, temperatures and pressure drops.

Since the ALLS transfer system was designed for onshore or offshore conditions, all the trials were also carried out with dynamic oscillations applied to the system. The entire results obtained during the successful ambient and LNG trials should therefore lead to the qualification of the ALLS transfer system by an independent certification organization of which a representative witnessed some of the trials.

Over and above this certification process, the LNG full-scale test campaign enabled the approval of the operational procedures for all the phases related to the use of ALLS, which is based on a cryogenic flexible hose technology and a multipurpose connecting system. This added value should facilitate the integration of this innovative transfer system for either onshore coastal weather exposed terminals or offshore projects with side-by-side or tandem configurations.

This world-first development has thus opened a way that should allow the LNG Industry to be confident with the use of cryogenic flexible hose-based transfer systems for commercial LNG transfer for the next generation of LNG facilities, whether they are fixed or floating, liquefaction plants or regasification terminals.

Acknowledgments

Based on a paper presented at the Offshore Technology Conference held in Houston, Texas, U.S.A., May 5-8, 2008.