Method for cleaning a component

Abstract

One aspect of the invention relates to a method for cleaning a component of a circuit for circulating a fluid in a gas turbine, the component being at risk of coking, the method comprising, in this order: - injecting (110) a cleaning product into the circulation circuit in order to cover a surface at risk of coking of the component, the cleaning product being in a semi-solid form; - injecting (120) a rinsing product into the circulation circuit.

Description

DESCRIPTION TITLE: Component Cleaning Procedure TECHNICAL FIELD OF THE INVENTION

[0001] The technical field of the invention is that of the maintenance of turbomachinery.

[0002] In particular, the invention relates to a method for cleaning components of fuel injection systems in turbomachinery, especially in gas turbines. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0003] The components of a gas turbine's fuel injection system, particularly the injectors, are highly susceptible to fuel degradation caused by temperature. This phenomenon, known as coking, occurs when three conditions are met: fuel is present in retention areas, there is significant residual heat from the engine after operation above a certain temperature, and the residual fuel remains exposed to that temperature for an extended period. This results in the formation of carbon deposits within the fuel injection system components, for example, in the form of plaques or agglomerates. The fuel injection system components are particularly vulnerable to this phenomenon due to the resulting disruptions in fuel flow, or even blockages.

[0004] Flow disturbances caused by coking have mechanical repercussions on components located downstream—relative to the direction of fuel flow—of the fuel injection system, as well as on combustion quality. Coking leads to flow rate differences between injectors and consequently to differences in flame temperatures within the combustion chamber. This is detrimental to the combustion chamber and the turbine, as it creates hot spots and therefore damage. This damage results in a reduced lifespan for the combustion chamber and turbine, and also shortens the average time between injector removals because they then need to be cleaned more frequently. Furthermore, injector clogging impacts the quality of fuel atomization, which becomes uneven.

[0005] In the most severe cases, coking can lead to an inability to start an engine on the ground, an inability to restart an engine in flight, thermal perforation of the combustion chamber, or even destruction of the turbine blades. These factors also result in reduced engine performance at very high RPMs or a decreased shutdown margin during very strong decelerations.

[0006] Due to coking, it is necessary to regularly clean the components of the fuel injection system. Current cleaning methods can be separated into two categories, depending on the type of component to be cleaned: "with removal" of the component, meaning that the component to be cleaned is removed from the engine before being cleaned, and "without removal" of the component, meaning that the component to be cleaned is left in place on the engine and cleaned while it is mounted on the engine.

[0007] Cleaning processes involving removal include pyrolytic cleaning and wet cleaning using a degreasing agent (alkaline or hydrocarbon solvent) as the cleaning solution. Removing the engine component results in significant downtime and substantial maintenance (requiring personnel to remove the component, clean it, allow it to cool in the case of pyrolytic cleaning, and then reinstall it). Furthermore, pyrolytic processes require specialized equipment (e.g., a pyrolytic oven).

[0008] Cleaning processes without removal can be used, for example, for cleaning nozzles and are carried out using a wet method. According to these processes, a cleaning solution is injected into the fuel circulation circuit (hereafter simply called the "fuel circuit") in several stages and over a relatively long period (between 20 and 30 minutes in total) to ensure that the circulating cleaning solution properly cleans the component to be cleaned. A rinsing solution is then injected into the fuel circuit to remove the cleaning solution, and finally the fuel circuit is dried by blowing air through it. These processes are preferentially used for new or recently cleaned components, as they are not suitable for cleaning components that are already clogged. Furthermore, implementing these processes is complex, as it involves a series of steps. according to a very precise sequence with specific means, and it still requires a fairly high maintenance load (need to circulate the cleaning solution inside the circuit for a sufficient time, then to circulate the cleaning solution inside the circuit until the cleaning solution is completely evacuated).

[0009] There is therefore a need for a fuel injection system component cleaning process that is simple and quick to implement, and that provides effective results even on partially clogged components. SUMMARY OF THE INVENTION

[0010] The invention offers a solution to the problems mentioned above by injecting a semi-solid cleaning product—that is, a product in the form of a paste, cream, or highly viscous fluid—into the fuel circulation circuit. The cleaning product may, for example, be in the form of a gel, foam, or cream. The cleaning product thus covers, at least partially, any coked areas, and its semi-solid state allows it to remain in place within the fuel circulation circuit, in contact with the component to be cleaned. A rinsing solution is then injected to remove the cleaning product along with any coked compounds.Thus, according to the proposed solution, the cleaning of the component is done by contact of the cleaning product with the component (i.e. that the cleaning is due to the action, on any coked compounds, of the cleaning product which is in "static" contact with the component), whereas in the wet processes of the prior art, the cleaning is done by circulation of the cleaning liquid in the component (therefore by a "dynamic" action).

[0011] One aspect of the invention relates to a method for cleaning a component of a fluid circulation circuit in a gas turbine, the component being subject to a risk of coking, the method comprising, in this order:

[0012] - inject, via a first inlet port of the fluid circulation circuit located upstream of the component relative to the direction of fluid flow in the circulation circuit, a cleaning product into the circulation circuit so as to coat a surface at risk of coking of the component, the cleaning product being in a semi-solid form; and

[0013] - inject, via a second inlet port of the fluid circulation circuit located at the same level or upstream of the first inlet port with respect to the direction of fluid circulation in the circulation circuit, a rinsing product into the circulation circuit.

[0014] A "fluid circulation circuit" is defined as any component or set of components forming a circuit through which a fluid can circulate. For example, the fluid could be, but is not limited to, oil or fuel. Such a fluid circulation circuit includes at least one inlet point for the fluid to enter the circuit, and one outlet point for the fluid or fluid residue to exit the circuit.

[0015] The first and second inlet ports may be a single port or two separate ports and constitute one or more entry points into the circulation circuit. These inlet ports are located upstream of the component to be cleaned, so that the cleaning agent and the rinsing agent can reach the component. Furthermore, the second inlet port is located at the same level (particularly in cases where the first and second inlet ports are the same) or upstream of the first inlet port (so that all portions of the circulation circuit can be rinsed with the fluid into which the cleaning agent has been introduced).

[0016] According to the invention, the cleaning product is in a semi-solid form. Such a form is defined as a form between the solid and liquid states, in the sense that a semi-solid material possesses certain properties characteristic of solids (in particular, a consistency characteristic of solids, which allows the material to not, or only slightly, "flow") and certain properties characteristic of liquids (in particular, to conform to the shape of the mold or component into which they are injected).

[0017] A semi-solid material is thus a material with a viscosity within a predefined range, which allows it to possess both good strength and conformational properties. Typically, the viscosity can be between 10 Pa·s and 100 Pa·s.

[0018] Injecting a cleaning product in semi-solid form advantageously allows for coating the surface of the component where coked deposits are likely to be present, and for remaining in contact with this The cleaning agent is applied to the surface without flowing directly to an outlet in the fluid circulation circuit. This allows the cleaning agent to remain in place to perform its cleaning action (in this case, by adhering to the coked deposit and / or by dissolving or disintegrating the coked deposit). During the injection of the rinsing agent, a mixture comprising the cleaning solution with any coked deposits (if present on the component surface) and the rinsing solution is discharged from the fluid circulation circuit to an outlet.

[0019] Thus, according to the invention, cleaning the component is linked to "prolonged" contact of the cleaning product with any coked deposits. Here, the term "prolonged" means that the cleaning product remains in contact with any coked deposits until the rinsing product is injected. Conversely, in existing wet processes, cleaning the component is linked to the circulation of a quantity of fluid along the surface on which any coked deposits are present.

[0020] The cleaning process according to the invention is very simple to implement and requires little maintenance (personnel are only needed for the injection of the two products but remain available for other tasks between the two steps). Furthermore, the quantity of cleaning product is advantageously reduced compared to existing processes. Finally, contact cleaning proves more effective than circulation cleaning.

[0021] In one or more embodiments, between the injection of the cleaning product and the injection of the rinsing product, the cleaning product may remain in contact with the surface at risk of coking of the component.

[0022] In one or more embodiments, the process can be implemented when the component is installed on the gas turbine.

[0023] In these embodiments, the component is therefore left in place (i.e., installed or mounted) on the gas turbine, and it is not necessary to remove it. This is thus a "no-removal" process for the component. Such embodiments reduce the maintenance burden compared to processes involving component removal (i.e., dismantling the component for separate cleaning), as well as the downtime (in processes involving removal, the equipment remains unavailable during dismantling, cleaning, and reassembly).

[0024] In other embodiments, the process can be used on a component that has been disassembled or uninstalled from the gas turbine (process with component removal).

[0025] In one or more embodiments, the injection of the rinsing product is carried out after a predefined time following the injection of the cleaning product, the predefined time being between 2 hours and 12 hours.

[0026] Thus, the cleaning product remains in contact with the surface of the component on which any coked deposits are present for the entire predefined time, allowing optimal action of the cleaning product on any coked deposits.

[0027] In one or more embodiments, the component may be a hollow component comprising an internal surface delimiting an interior space, and the cleaning product may be injected into the interior space so as to cover the entire internal surface of the component.

[0028] In particular, in which the cleaning product can be injected into the interior space so as to occupy the entire interior space.

[0029] In one or more embodiments, the component may be a component of a gas turbine fuel injection circuit.

[0030] For example, the component could be a component of an injection rail, such as a half-rail or a preferred injector.

[0031] In one or more embodiments, the cleaning product may be a thixotropic fluid.

[0032] A thixotropic fluid is a fluid whose flow properties vary depending on the applied stress. Using a thixotropic fluid as a cleaning agent allows it to be in a liquid state upon injection, to solidify when at rest (typically between the injection of the cleaning agent and the injection of the rinsing agent), and to return to a liquid state upon the injection of the rinsing agent. The injection of the cleaning agent, its action by contact on any coked deposits, and the removal of the cleaning agent are thus advantageously simplified and implemented more efficiently.

[0033] In one or more embodiments, the cleaning product is in the form of a paste, cream or gel, and wherein the cleaning product has a viscosity between 10 Pa.s and 100 Pa.s.

[0034] In one or more alternative embodiments, the cleaning product may be in a foam material having an expansion ratio of less than 15.

[0035] In one or more embodiments, the cleaning product may include a solvent or a detergent.

[0036] In one or more embodiments, the cleaning product may have an action of adhesion and / or dissolution on coked compounds.

[0037] In one or more embodiments, the rinsing product can be a demineralized water-based solution or a solution based on the fluid intended to circulate in the fluid circulation circuit when the gas turbine is in operation.

[0038] In one or more embodiments, the fluid circulation circuit may be a fuel circulation circuit, and the flushing product may include fuel.

[0039] In one or more embodiments, the injection of the rinsing product may include an injection of a first rinsing product, then an injection of a second rinsing product, the second rinsing product comprising the fluid intended to circulate in the fluid circulation circuit when the gas turbine is in operation.

[0040] In one or more embodiments, the first inlet port and the second inlet port may be common.

[0041] The implementation of the process is thus advantageously simplified. In addition, using the same orifice to inject both the cleaning product and the rinsing product ensures that any residues of the cleaning product at the inlet of the cleaning product injection orifice are eliminated.

[0042] In one or more embodiments, the process may further include, after the injection of the rinsing product:

[0043] - to recover, at an outlet of the fluid circulation circuit, the cleaning product and the rinsing product.

[0044] For example, a mixture including the cleaning product, the rinsing product and any coked compounds can be recovered in a container placed at an outlet of the fluid circulation circuit.

[0045] In one or more embodiments, the component is a component of a fuel injection circuit of the gas turbine and the outlet of the fluid circulation circuit is located in a combustion chamber of the gas turbine.

[0046] The invention and its various applications will be better understood by reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0047] Other features and advantages of the invention will become apparent from the description, which can be read in conjunction with the figures. These figures are provided for illustrative purposes only and are not intended to limit the scope of the invention.

[0048] Figure 1 represents an example of a flowchart of a cleaning process according to an embodiment of the invention.

[0049] Figure 2 represents an example of a gas turbine in which the cleaning process according to the invention can be applied.

[0050] Figure 3 represents an example of a component of a fluid circulation circuit on which the cleaning process according to the invention can be used. DETAILED DESCRIPTION

[0051] Figure 1 represents an example of a flowchart of a cleaning process according to an embodiment of the invention.

[0052] It is noted that the cleaning process can be used to clean any component susceptible to coking. In particular, the component to be cleaned can be any component of a fluid circulation circuit in which coking may occur. Such a circulation circuit could be, for example, a fuel injection circuit of a gas turbine. In particular, The component to be cleaned may be a component of a gas turbine injector, for example a half fuel injection rail or a preferred injector.

[0053] It is noted that the cleaning method according to the invention is not limited to a fuel injection circuit. It can be used for any fluid circulation circuit susceptible to coking, particularly the oil circuit in a turboshaft engine such as a turbojet. Indeed, oils are used in turboshaft engines to cool and lubricate the bearings and gear systems within the various housings, and these oils can also cokify with heat, thus causing obstruction of the passages and / or filters constituting the oil circuit, which can have serious consequences for the operation of the turboshaft engine.

[0054] In step 110, a cleaning product is injected into the fluid circulation circuit to which the component to be cleaned belongs. To ensure the cleaning product reaches the areas of the component susceptible to coking, it is injected through a first inlet (or first injection point) of the fluid circulation circuit, located upstream of the component relative to the direction of fluid flow within the circuit.

[0055] The cleaning product injected in step 110 is injected so as to cover a surface of the component on which deposits of coked compound(s) are likely to be present.

[0056] For example, when the component is a hollow element, such as a half-injection rail, inside which coking may occur, the cleaning agent is injected so as to penetrate the component, coating an internal surface and thus any coked compounds deposited on that internal surface. Specifically, the cleaning agent can be injected to completely fill the interior of the component to be cleaned.

[0057] The cleaning product, according to the invention, is in a "semi-solid" form. A semi-solid product (also called a "quasi-solid" or "false-solid") is a product in a state between a solid and a liquid. In particular, a semi-solid product exhibits many of the properties of a solid, such as the ability to support its own weight and retain its shape, but also certain properties of a liquid, such as conforming to the shape of an object that applies pressure and the ability to flow under pressure. Semi-solid forms include, in particular, gels, creams, pastes, and foams. Semi-solid products have sufficient viscosity to maintain their shape without "running." In the context of the invention, the cleaning product can thus have a viscosity between 10 Pa·s (Pascal seconds) and 100 Pa·s, that is, between 100 P (Poise) and 1000 P.

[0058] The advantage of this form for the cleaning product is that it prolongs the contact time with the potentially coked component. Indeed, unlike a liquid that would flow directly to an outlet of the fluid circulation circuit, a semi-solid cleaning product flows less quickly (and may even remain in place at the end of the injection 110) and therefore remains in contact with the component for longer, allowing it to act on any coked compounds on the component's surface.

[0059] In embodiments of the invention, the cleaning product may be in the form of a gel or a foam. For example, the cleaning product may be a gel with a viscosity between 10 Pa·s and 100 Pa·s. In another example, the cleaning product may be a foam with an expansion ratio of less than 15. It should be noted that the expansion ratio is the ratio between the volume of foam obtained and the volume of the foaming solution used to produce that foam.

[0060] In one or more embodiments, the cleaning product may be thixotropic, meaning that its flow properties vary over time depending on an applied shear stress (or velocity gradient). This phenomenon is due to a change in the material's structure when subjected to shear stress (the product "breaks down"). A thixotropic material left at rest can thus (re)structure itself and its viscosity increases. Under the application of shear stress, the thixotropic material breaks down and its viscosity decreases, allowing it to flow.

[0061] This thixotropic property allows for easy injection of the cleaning product (due to a lower viscosity) and better product adhesion. the interior of the fluid circulation circuit (due to a higher viscosity once the injection is complete). Furthermore, it also allows for easier removal of the cleaning product during rinsing step 120 described below. Indeed, the injection of a rinsing product in step 120 exerts a shear stress on the cleaning product, which causes a decrease in its viscosity, allowing it to be removed more easily and efficiently. When the cleaning product is thixotropic, it can have a viscosity under stress between 10 Pa·s and 100 Pa·s. "Viscosity under stress" refers to the viscosity of the cleaning product when a stress is applied to it (Le, in its "unstructured" state).

[0062] In some embodiments, the cleaning product may have a composition comprising an active substance that has a cleaning action on coked deposits. For example, the cleaning product may have a dissolving effect on coked deposits. The active substance may be a solvent (e.g., water, alcohol, ester, or hydrocarbon) and / or a detergent (e.g., ethanolamine, sodium metasilicate). In some embodiments, the composition of the cleaning product may include acidic or basic, organic or inorganic compounds, potentially in combination with a solvent and / or a detergent. In embodiments consistent with the preceding ones, the cleaning product may include particles having abrasive properties on coked deposits (e.g., silica or alumina particles) to maximize the cleaning action.

[0063] In alternative or combined embodiments with the preceding ones, the cleaning product can have an adhesive action on the coked deposits: the product attaches to the coked deposits, and carries them away with it when it is expelled from the fluid circulation circuit during the rinsing step 120 described below.

[0064] In addition, the composition of the cleaning product is established according to the component to be cleaned and any other components of the fluid circulation circuit with which the cleaning product is intended to come into contact, to ensure compatibility with this or these components (and to ensure in particular that the cleaning product does not damage the component(s), or does not cause a harmful chemical reaction that would release dangerous compounds).

[0065] The amount of cleaning product injected depends on the surface area of ​​the component to be cleaned, or, in the case of a hollow component, on the internal volume defined by its inner wall. This amount can therefore be predetermined based on the dimensions of the component to be cleaned, as well as the injection point in the fluid circulation circuit (to ensure that a sufficient quantity of cleaning product reaches the component). This quantity can advantageously be determined in such a way as to limit the excessive amount of cleaning product injected, that is, the amount of cleaning product exceeding the optimal amount required to cover the surface of the component.

[0066] In addition, to reduce the amount of cleaning product to be injected, the first inlet port can be located as close as possible to the component to be cleaned, upstream of that component.

[0067] The injection of the cleaning product in step 110 can be carried out via any product injection device, for example via a syringe, or via a pressurizing device such as a compressed air gun (for example a foam gun or a siphon, when the cleaning product is in foam form).

[0068] After the cleaning agent is injected (step 110), a rinsing agent is injected (step 120) into the fluid circulation circuit to rinse the parts of the fluid circulation circuit that are or have been in contact with the cleaning agent and to remove the cleaning agent from the fluid circulation circuit. To ensure that the rinsing agent reaches all parts that may be coated with cleaning agent (or traces of cleaning agent), the rinsing agent is injected through a second inlet (or second injection point) of the fluid circulation circuit, located at the same level as, or upstream of, the first inlet with respect to the direction of fluid flow in the fluid circulation circuit. Thus, in one or more embodiments, the first and second inlet ports are the same port.According to alternative embodiments, the second inlet port is located upstream of the first inlet port.

[0069] The injection of the rinsing agent allows for mechanical removal (the flow of the rinsing agent "pushes away" the cleaning agent) and / or chemical removal (the rinsing agent dissolves the cleaning agent). The rinsing agent can be, for example, a liquid. In particular, the rinsing agent can be a water-based rinsing solution, such as demineralized water or a mixture of water and methanol, or a hydrocarbon-based rinsing solution (fuel, solvent, etc.), or even the fluid intended to circulate within the fluid circulation system.

[0070] The advantage of using the fluid intended to circulate in the fluid circulation circuit (for example, fuel in a fuel circuit or oil in an oil circuit) as a flushing product is that it is not necessary to dry the circuit after the cleaning process (if there is any flushing product left in the circuit, this does not pose any problem in terms of engine operation, and this remaining flushing product will be mixed with the fluid in the circuit during engine operation), and that the fluid is, by nature, compatible with the elements of the circulation circuit.

[0071] The disadvantage of using the fluid intended to circulate in the fluid circulation circuit as a rinsing product is that such a fluid is more expensive and more dangerous to handle than demineralized water, for example.

[0072] Thus, in one or more embodiments, the flushing agent injection in step 120 may include a plurality of flushing agent injections. For example, step 120 may include an injection of a first flushing agent (mineralized water, for example, but not limited to) followed by an injection of a second flushing agent (for example, fluid intended to circulate in the fluid circulation circuit when the engine is running).

[0073] The amount of rinsing product injected can be predetermined during tests according to the dimensions of the component to be cleaned, and possibly the point of injection in the fluid circulation circuit, so as to expel all the cleaning product present in the fluid circulation circuit, without however using an excessive amount of rinsing product.

[0074] The injection of rinsing product in step 120 can be carried out via any product injection device, for example via a syringe, or via a compressed air pressurization device.

[0075] In one or more embodiments, step 120, the rinsing agent injection step, is carried out after a predefined time following step 110, the cleaning agent injection step. During this predefined time, the cleaning agent injected in step 110 remains in contact with the component to be cleaned.

[0076] This predetermined time depends on the cleaning power of the cleaning product. For example, if the cleaning product acts by adhesion, and the removal of coked compounds is achieved by flushing away the cleaning product to which the coked compounds are attached, through the injection of rinsing agent, the predetermined time is related to the time required for the cleaning product to bind to the coked compounds. If the cleaning product acts by dissolving the coked compounds, the predetermined time is related to the time required for the cleaning product to satisfactorily dissolve the coked compounds. As mentioned previously, these two modes of action of the cleaning product can be combined.

[0077] This predefined time also depends on the amount of coked compounds on the surface of the component, which can be estimated by simulations or determined by endoscopy, for example.

[0078] Thus, depending on the degree of coking, the type of action of the cleaning product, and its active component, this predefined time can range from a few minutes to several hours. In one or more embodiments, this predefined time is between 1 and 24 hours, for example, between 2 and 12 hours, between 2 and 8 hours, or between 2 and 6 hours.

[0079] It is understood that the cleaning process according to the invention requires significantly less maintenance and a much smaller quantity of product than existing wet processes. In existing wet processes, a large quantity of cleaning fluid must be circulated to remove coked compounds, since the cleaning action is primarily achieved through the circulation of the cleaning fluid (it is the flow that acts primarily on the (coked deposits) – which is why these processes are not recommended when components are moderately or heavily soiled. Furthermore, they require the mobilization of maintenance personnel or a very specific installation to be able to inject the cleaning fluid over a prolonged period and possibly in several stages, followed by the injection of the rinsing fluid over a prolonged period and possibly in several stages. Conversely, according to the cleaning process of the invention, the quantity of products injected, particularly the cleaning product, is greatly reduced, no specific installation is required, and the maintenance cost is advantageously reduced (personnel are only mobilized to inject the cleaning and rinsing products, but remain available between the two stages – which can potentially be carried out over two consecutive days).

[0080] The flushing agent, mixed with the cleaning agent and any coked compounds loosened from the component, is discharged through an outlet in the fluid circulation circuit. In the case of a fuel circuit, this outlet may be located in the combustion chamber. Specifically, this outlet may be an orifice located in a low-lying part (i.e., closest to the ground) of the combustion chamber, so that the flushing agent flows out of the combustion chamber by gravity.

[0081] To recover the mixture comprising the rinsing agent, the cleaning agent, and any loosened coking compounds, the cleaning process may optionally include a step 130 for recovering this mixture. For example, a container may be placed at the outlet of the fluid circulation circuit through which the mixture escapes, in order to collect the mixture.

[0082] Finally, in an optional step 140, the fluid circulation circuit can be dried, for example by blowing air into the circuit, via a circuit inlet which can be the same as the first inlet port or the second inlet port, or be a third inlet port in the fluid circulation circuit located upstream of the second inlet port.

[0083] This step is not mandatory, especially when step 120 of rinsing ends with an injection of fluid intended to circulate in the fluid circulation circuit (e.g., fuel in the case of a fuel circuit).

[0084] The cleaning procedure in Figure 1 can be used with or without removing the component to be cleaned. When the cleaning procedure is used without removing the component (i.e., the component is still in place on the engine and has not been disassembled), the maintenance load is significantly reduced.

[0085] Figure 2 represents an example of a gas turbine in which the cleaning process according to the invention can be applied.

[0086] In this example, the gas turbine is a gas turbine 200 from a turboshaft engine and comprises a gas generator with a centrifugal compressor 210 and a high-pressure turbine 220. The gas turbine 200 also includes a free turbine 230 and a reduction gearbox (not shown in Figure 2) with an output shaft. Figure 2 also shows the air inlet 240, the combustion chamber 250, and the fuel inlet 260. This fuel inlet includes an opening to the combustion chamber, which can be used as a common orifice for injecting the cleaning and rinsing agents.

[0087] Furthermore, the combustion chamber 250 includes, in its lower part, an opening 270 closed by a valve. This opening 270 (once the valve is open) can serve as an outlet to evacuate the mixture of cleaning product and rinsing product (with any coking compounds).

[0088] Figure 3 represents an example of a component of a fluid circulation circuit on which the cleaning process according to the invention can be used.

[0089] In the example in Figure 3, the fluid circulation circuit is a fuel circulation circuit (or simply "fuel circuit"). The components to be cleaned are injector components: the two half-rails 310, 320, and the preferred injector 330.

[0090] In this example, the cleaning product and the rinsing product are injected through the same inlet port using a 340 syringe. It is understood that the Syringe 340 can be replaced by any other injection method, as mentioned above. Furthermore, in this example, the inlet port is a spring preload adjuster cover, repurposed as an injection point, thus allowing the cleaning and rinsing products to be injected upstream and as close as possible to the components to be cleaned. It is understood that any other entry point in the fuel system located upstream of the components to be cleaned can be used as an injection point. In particular, the injection point can be any existing entry point in a "standard" fuel system (whose primary function is not to receive the cleaning or rinsing product), notably one created by removing a component (e.g., a sensor or valve) belonging to the fuel system and providing access to the component(s) to be cleaned.The injection point can also be a dedicated injection point, considered from the design stage of the fuel circuit for this purpose.

[0091] Of course, the present invention is not limited to the embodiments described above by way of example, it extends to other variants.

Claims

DEMANDS

1. A method for cleaning a component (310, 320, 330) of a fluid circulation circuit in a gas turbine, the component (310, 320, 330) being subject to a risk of coking, the method comprising, in this order: - inject (110), via a first inlet port of the fluid circulation circuit located upstream of the component (310, 320, 330) relative to a direction of fluid flow in the circulation circuit, a cleaning product into the circulation circuit so as to coat a surface at risk of coking of the component (310, 320, 330), the cleaning product being in a semi-solid form; and - inject (120), via a second inlet port of the fluid circulation circuit located at the same level or upstream of the first inlet port with respect to the direction of fluid circulation in the circulation circuit, a rinsing product into the circulation circuit.

2. Method according to claim 1, the method being implemented when the component (310, 320, 330) is installed on the gas turbine.

3. A method according to claim 1 or 2, wherein the injection (120) of the rinsing product is carried out after a predetermined time following the injection (110) of the cleaning product, the predetermined time being between 2 hours and 12 hours.

4. A method according to any one of the preceding claims, wherein the component (310, 320, 330) is a component of a gas turbine fuel injection circuit.

5. A method according to any one of the preceding claims, wherein the cleaning product is a thixotropic fluid.

6. A method according to any one of the preceding claims, wherein the cleaning product is in the form of a paste, cream or gel, and wherein the cleaning product has a viscosity between 10 Pa.s and 100 Pa.s.

7. A method according to any one of claims 1 to 5, wherein the cleaning product is in a foam material having an expansion ratio of less than 15.

8. A method according to any one of the preceding claims, wherein the first inlet orifice and the second inlet orifice are common.

9. A method according to any one of the preceding claims, further comprising, after the injection (120) of the rinsing product: - recovering (130), at an outlet (270) of the fluid circulation circuit, the cleaning product and the rinsing product.

10. A method according to the preceding claim in combination with claim 4, wherein the outlet (270) of the fluid circulation circuit is located in a combustion chamber (250) of the gas turbine.

Images