METHOD TO DETERMINE A DRY-OUT PERIOD OF A CONVERTER OF A WIND TURBINE

Abstract

A method to determine a dry-out period of a converter of a wind turbine is provided. A time dependent chronology of data is measured in the converter. The measured data-chronology reflects the actual humidity and the humidity-history in the converter cabinet. The measured data-chronology is used to determine a dry-out period of time, which is needed to reduce the humidity inside the converter below a given value by circulating heat inside the converter.

Claims

1. A method to determine a dry-out period of a converter of a wind turbine, comprising: measuring a time dependent chronology of data in the converter, wherein the measured data-chronology reflects an actual humidity and a humidity-history in a converter cabinet; and determining a dry-out period of time using the measured data-chronology is used, which is needed to reduce a humidity inside the converter below a given value by circulating heat inside the converter.

2. The method according to claim 1, wherein temperature-data are measured to form an additional part of the data-chronology.

3. The method according to claim 1, wherein the data-chronology is measured by a sensor, which is placed at or in the converter or in the converter cabinet.

4. The method according to claim 1, wherein a data logger is used to store the data-chronology.

5. The method according to claim 1, wherein the data logger is powered by a battery.

6. The method according to claim 1, wherein the stored data are processed and used for calculating the dry-out period.

Description

BRIEF DESCRIPTION

[0028] Some of the embodiments will be described in detail, with reference to the following figures, wherein like designations denote like members, wherein:

[0029] FIG. 1 shows a content of moisture in dependency of a relative air humidity

[0030] FIG. 2 shows a layer structure in reference to FIG. 1;

[0031] FIG. 3 shows a content of moisture in a gel-material in reference to FIG. 1 and FIG. 2;

[0032] FIG. 4 shows a “de-humidification process” with reference to FIG. 3;

[0033] FIG. 5 shows a moisture distribution and its dry-out in reference to the figures above;

[0034] FIG. 6 shows a time dependent variation of the situation as shown in FIG. 5; and

[0035] FIG. 7 shows the method invented based on a schematic wind turbine.

DETAILED DESCRIPTION

[0036] FIG. 1 shows a content of moisture y.sub.e[g/kg] in dependency of a relative air humidity rh[−].

[0037] The content of moisture y.sub.e[g/kg] is plotted along the vertical axis while the relative air humidity rh[−] is plotted along the horizontal axis of FIG. 1.

[0038] The curves, which are shown for three temperatures 40° C., 20° C. and 0° C., are known as so called “Equilibrium Moisture Content, EMC”-curves for a given material.

[0039] In reference to FIG. 2 this material might be a gel and is abbreviated in FIG. 2 as “gel” accordingly. In example this gel might be used to cover and protect IGBT-elements, which are central parts of any wind turbine converter.

[0040] FIG. 2 shows in a very principle sketch different layers of material above an electrical component EC.

[0041] The thickness of the material is shown in dependency of a parameter x, which could be read as parameter for the material thickness or even as “vertical”-oriented direction into the increasing depth of the respective material.

[0042] In example the electrical component EC might comprise IGBT-elements, which are central parts of any wind turbine converter (IGBT is an abbreviation of “Insulated-Gate Bipolar Transistor”).

[0043] The electrical component EC is arranged in contact with a cooling plate CP. The electrical component EC is even in contact with and covered and protected by the gel as described in FIG. 1.

[0044] The gel-layer is exposed to air on its top, which shows a relative air humidity rh(air) at a given depth x.

[0045] Along the x-direction the depth of the gel increases steadily.

[0046] Based on FIG. 1 and on FIG. 2 reference is now made to FIG. 3.

[0047] FIG. 3 shows the content of moisture y[g/kg] in the gel in dependency of a respective depth x[m], referring to FIG. 1 and to FIG. 2.

[0048] The content of moisture y[g/kg] is plotted along the vertical axis while the depth x[m] is plotted along the horizontal axis of FIG. 3.

[0049] FIG. 3 reflects a dry gel-material, which is exposed to a humid atmosphere. Initially the moisture content in the gel-material is y.sub.init.

[0050] After some time, say 1 hour, the moisture of the humid atmosphere has diffused into the gel material—this is given by curve 1.

[0051] In dependency of time τ the moisture diffuses more into the gel-material, which is given by the curve 2, curve 3 and curve 4.

[0052] Finally the moisture content into the gel-material will be y.sub.e everywhere, while y.sub.e is even the moisture content at the boundary line between air and gel.

[0053] Thus the parameter y.sub.e marks an maximum amount of moisture-content in the gel or even an equilibrium of moisture-content in the gel.

[0054] It can be seen that the relative humidity y(g/kg) at the boundary line between air and gel is high—this is shown by the joint point of intersection of all three curves.

[0055] Summoned up the figures show this principle over time: “the higher the relative humidity in the air, the higher the moisture content in the gel”.

[0056] This is based on the fact that—for many materials including gel material—the resistance for moisture, flowing into the surface of the material, is small compared to the resistance for the moisture, diffusing inside the material.

[0057] The “humidification process” as described above will happen if a respective IGBT-element is not heated or temperature controlled, i.e. if there is no auxiliary power available for this purposes due to an OFF-grid-situation.

[0058] FIG. 4 shows a respective “de-humidification process” in view to the process described in FIG. 3—thus the moisture is dried out now.

[0059] FIG. 4 shows the content of moisture y[g/kg] in the gel (vertical axis) in dependency of a respective depth x[m] (horizontal axis), referring to the figures above.

[0060] Here the initial moisture content in the gel is denoted by y.sub.init. The moisture content of the dry air is denoted by y.sub.e.

[0061] After a respective period of time τ the gel-material will be dried out, ending up with the moisture content y.sub.e inside the whole gel-material.

[0062] This “drying out process” needs to be done whenever an IGBT-element was face with an OFF-grid-time.

[0063] FIG. 5 shows a specific case: the dry gel-material is exposed to a humid atmosphere, for example for a time period of one hour, in a first step.

[0064] This results in a distribution of moisture content y.sub.e as shown and in dependency of the depth x.

[0065] In a second step a dry out of this humidity is initiated. In example after a period of time of approximately four hours the gel material is dried out again.

[0066] FIG. 6 shows a similar case but with these differences: the gel has been exposed to a humid atmosphere for a longer time before the drying-out process is started.

[0067] As a consequence the drying-out will take longer time.

[0068] In reference to the figures above the curves as shown can be simulated by a model.

[0069] Based on this model and based on the gathered data-chronology of the humidity inside the converter a required and optimized dry-out time can be calculated.

[0070] The dry put time is needed to reduce the humidity inside the converter below a given value—for example a fluid (like air or water or the like) might be circulated inside the converter for heating purposes.

[0071] As a consequence the dry-out time can be reduced in most cases to an optimized time period.

[0072] Alternatively (instead of simulation based on a model) the needed dry-out time could be estimated and based on a measured humidity inside the GEL as well.

[0073] FIG. 7 shows the method invented based on a schematic wind turbine.

[0074] A number of blades BL of a wind turbine WT is driven by wind and transfers rotational energy to a generator GEN.

[0075] The generator GEN transforms the rotational energy into electrical power, which is passed on to a converter CONV.

[0076] The converter CONV is arranged into a converter cabinet CAB and converts electrical power with a varying frequency into electrical power with constant frequency.

[0077] According to the method invented a time dependent chronology of data TDCD is measured in the converter CONV of the wind turbine WT.

[0078] The measured data-chronology TDCD reflects the actual humidity and the humidity-history in the cabinet CAB of the converter CONV.

[0079] In a second step S2 the measured data-chronology TDCD is used to determine a dry-out period of time, which is needed to reduce the humidity inside the converter CONV and its respective converter cabinet CAB below a given value.

[0080] This could be done by circulating of a heated fluid (air or water) inside the converter. It is even possible to use any other heat (i.e. any direct or indirect heating, electrical power-based heating, or the like).

[0081] The data-chronology of the humidity is measured by a sensor SEN, which is placed in the converter CONV or in the respective converter cabinet CAB.

[0082] The data-chronology is stored in a data-logger DL, which might be power by a battery.

[0083] The stored data TDCD are processed and used for calculating the dry-out period.

[0084] Although the present invention has been disclosed in the form of preferred embodiments and variations thereon, it will be understood that numerous additional modifications and variations could be made thereto without departing from the scope of the invention.

[0085] For the sake of clarity, it is to be understood that the use of “a” or “an” throughout this application does not exclude a plurality, and “comprising” does not exclude other steps or elements.