Method for determining the degree of damage caused by cavitation in a hydropower plant
The method uses a structure-borne sound sensor to analyze frequency spectra and determine cavitation damage, addressing the inefficiency of existing methods by identifying safe operating states in hydroelectric power plants.
Patent Information
- Application Number
- DE102019121978
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2019-08-15
- Publication Date
- 2024-12-05
- Estimated Expiration
- 2039-08-15
AI Technical Summary
Existing methods for detecting cavitation in hydroelectric power plants unnecessarily restrict the operating range due to the inability to differentiate between damaging and non-damaging cavitation states, leading to inefficient operation.
A method using a structure-borne sound sensor to analyze frequency spectra and determine the degree of cavitation-related damage by correlating amplitude shifts in specific frequency bands, allowing for the assessment of safe operating states without complex inspections.
Enables accurate determination of cavitation damage without extensive plant inspections, optimizing operational efficiency by identifying safe cavitation states.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The present invention relates to the operation of a hydropower plant with a structure-borne sound sensor which is suitable for detecting cavitation.
[0002] Cavitation can occur during the operation of a hydropower plant. Cavitation usually occurs in operating ranges that lie outside the optimal operating range for which the hydraulic machine was primarily designed. Methods for detecting cavitation in hydropower plants are known from the prior art. DE 10 2015 201 203 A1, for example, describes a method for detecting cavitation using at least one structure-borne sound sensor. The measured signal is processed using a specific algorithm to derive a characteristic value, which is compared with a global limit value. Studies have shown that this method can detect incipient cavitation with approximately 90% probability. The aim of the known methods is to prevent damage to the affected hydropower plant by avoiding operating ranges in which cavitation is detected.
[0003] Since it is possible that no or only very slight damage to the hydropower plant occurs in certain operating ranges, even though cavitation occurs in these operating ranges, the methods known from the state of the art result in such operating ranges being avoided even though this might not actually be necessary, thus unnecessarily restricting the usable operating range. The disadvantage described could obviously be avoided if it were known whether a certain cavitation-prone operating condition leads to damage and what the respective damage rate is. In principle, the necessary information could be obtained by inspecting the hydropower plant for the damage caused after operation in a cavitation-prone operating condition for a certain period. However, the effort required for this is so great that such an approach appears impractical and uneconomical.
[0004] The document JAMIL, I. [et al.]: Technical communication of Condition Monitoring System of hydroelectric generating unit of HPP. In: IEEE 4th International Conference on Electronics Information and Emergency Communication, Beijing, China, 2013, discloses a method for determining the degree of cavitation-related damage in a hydropower plant. CN 203 629 869 U discloses a similar method.
[0005] The object of the present invention is to provide an alternative method that allows the determination of cavitation-related damage to a hydropower plant from the measurement signal of a structure-borne sound sensor. The method according to the invention thus provides the desired information without the need for complex inspections of the hydropower plant.
[0006] The stated problem is solved by the features of the independent claim. Advantageous embodiments emerge from the dependent subclaims.
[0007] The inventive solution is explained below with reference to figures. They show the following in detail: Fig. 1 hydropower plant on which the method according to the invention for determining the degree of cavitation-related damage can be carried out; Fig. 2 Diagram of mass loss due to cavitation; Fig. 3 Frequency spectrum; Fig. 4 Course of the signal amplitude in a first frequency band; Fig. 5 Course of the signal amplitude in a second frequency band; Fig. 6 Flowchart of the method according to the invention.
[0008] Fig. Figure 1 shows a highly schematic representation of a hydropower plant on which the method according to the invention for determining the degree of cavitation-induced damage can be carried out. The hydropower plant is designated by 1. The hydropower plant 1 comprises at least one impeller, designated by 2. Furthermore, the hydraulic machine 1 comprises at least one structure-borne sound sensor, designated by 3, and means suitable for carrying out the steps of the method. These means are designated by 4. These can be, for example, one or more computer-based evaluation or control units. Fig. Figure 1 further shows a computer-readable medium on which a computer program for carrying out the method according to the invention is stored. The computer-readable medium is designated 5.
[0009] The method according to the invention is based on the results of a study conducted by the inventors to investigate cavitation-induced damage to hydropower plants. Starting from an undamaged state, a hydropower plant's wing was exposed to damaging cavitation over a period of several hundred hours. The progressive degree of damage was quantitatively determined by determining the increasing mass loss of the wing.
[0010] Fig. Figure 2 shows the mass of the wing measured during the study as a function of the time the wing was exposed to cavitation. The y-axis values are given in grams, and the x-axis values are given in hours. The wing loses mass with increasing damage time, with the mass loss occurring approximately linearly. The increasing mass loss is a measure of the degree of damage to the wing.
[0011] During cavitation damage, measurement data were recorded using a structure-borne sound sensor. The measurement data from such a sensor consists of an analog electrical signal as a function of time. The data provided by the sensor were recorded and stored at a suitable sampling rate. Frequency spectra were calculated from this data using a Fourier transform. The inventors analyzed the frequency spectra in the ultrasonic range (> 20 kHz). Fig. Figure 3 shows such a frequency spectrum.
[0012] From the frequency spectra recorded successively over the damage period of almost 300 hours, the inventors were able to determine that the amplitude in certain frequency bands increased over time, while the amplitude in other frequency bands decreased over time. This revealed that the amplitude strength increasingly shifted from higher frequencies to lower frequencies.
[0013] Fig. Figure 4 shows the amplitude curve in a first frequency band. The frequency band extends from 50 to 51 kHz. The x-axis indicates the damage time in hours. It can be seen that the amplitude increases approximately linearly. The linear trend is overlaid by a relatively strong noise. Other suitable frequency bands in which the amplitudes vary similarly to those in Fig. 4 can be found in the frequency range between 20 and 90 kHz.
[0014] Fig. Figure 5 shows the amplitude curve in a second frequency band. The frequency band extends from 320 to 330 kHz. The x-axis indicates the damage time in hours. It can be seen that the amplitude decreases approximately linearly. The linear trend is overlaid by a relatively strong noise. Other suitable frequency bands in which the amplitudes are similar to those in Fig. 4 can be found in the frequency range between 100 and 1000 kHz.
[0015] This finding allowed the inventors to establish a correlation between the shift in amplitude strength in the frequency spectrum and the degree of damage. Depending on the size of the frequency spectra taken into account, different correlation metrics result. For example, the Fig. 4 and Fig. The values shown in Figure 5 are used for the correlation. Fig. 4, the correlation factor is positive, since the degree of damage increases with increasing amplitude in the frequency band. Fig. For the quantity shown in Figure 5, the correlation factor is negative, since the degree of damage increases with decreasing amplitude in the frequency band.
[0016] The described finding and the correlation derived from it allow the inventors to provide a method for the first time that makes it possible to determine the degree of cavitation-related damage in a hydropower plant on the basis of data recorded with a structure-borne sound sensor.
[0017] The flow chart of the method according to the invention is shown in Fig. 6. In step V1, measurement data is acquired using a structure-borne sound sensor in a cavitation-prone operating state of the hydropower plant. In step V2, at least one frequency spectrum is calculated from the measurement data acquired in step V1. In step V3, the degree of damage is determined from the frequency spectrum calculated in step V2 using a predetermined correlation metric.
[0018] The correlation metric is predetermined using a prototype or a model test. The correlation metric includes a reference value for the correlation quantity under consideration and a correlation factor. The reference value can refer to either the undamaged state or any other state with a known degree of damage. The correlation factor indicates by how much the degree of damage changes when the correlation quantity changes by one unit. When measuring according to the method according to the invention, the degree of damage can then be determined as follows: The reference value is subtracted from the value of the correlation quantity determined by the measurement, and this difference is multiplied by the correlation factor.
[0019] Out of Fig. 4 and Fig.Figure 5 shows that correlation values derived from frequency spectra can contain significant noise. The harmful influence of noise can be reduced by all known methods, for example, simply by determining a large number of frequency spectra during a measurement and then averaging (either the spectra themselves or the correlation values derived from them).
[0020] The method according to the invention can also be used very well for assessing cavitation-prone operating conditions of hydropower plants. Measurements are carried out according to the method according to the invention over a sufficiently long period of time. If the degree of damage does not increase or increases only very slightly during this period, then the corresponding operating condition can be maintained for longer periods without fear of excessive damage from cavitation. Otherwise, the operating condition should be avoided. It is clear that it is sufficient to consider only the correlation variable: If there is no or only a negligible change in the correlation variable, then the corresponding operating condition is harmless. Determining a correlation variable alone is equivalent to determining the degree of damage, even if this is not explicitly calculated.Therefore, the correlation factor is not even needed to assess an operating condition, since it is sufficient to observe the correlation value with regard to a change.
Claims
[1] Method for determining the degree of damage caused by cavitation in a hydropower plant, comprising the following steps: - V1: Recording measurement data in a structure-borne sound sensor (3) in a cavitation-affected operating state of the hydropower plant; - V2: Calculation of a frequency spectrum from the measurement data acquired in step V1; - V3: Calculation of the degree of damage from the frequency spectrum calculated in V2 using a predetermined correlation metric; characterized by that the degree of damage is calculated as follows: - Determination of the signal amplitude on a frequency band; - Formation of a difference between the signal amplitude and a predetermined reference signal amplitude; - Multiplying the difference by a predetermined correlation factor. [2] The method according to claim 1, wherein the frequency spectrum is calculated on a frequency band of 20 to 1000 kHz. [3] A method according to claim 1 or 2, wherein the frequency band in which the signal amplitude is determined is in the range between 20 and 90 kHz, and wherein the correlation factor is positive. [4] A method according to claim 1 or 2, wherein the frequency band in which the signal amplitude is determined is in the range between 100 and 1000 kHz, and wherein the correlation factor is negative. [5] A method according to claim 3, wherein the width of the frequency band in which the signal amplitude is determined is 1 kHz. [6] A method according to claim 4, wherein the width of the frequency band in which the signal amplitude is determined is 1 to 20 kHz. [7] Method for assessing a cavitation-affected operating state of a hydropower plant, wherein the method according to one of the preceding claims is carried out several times over a period in which the operating state is assumed, wherein in step V3 the correlation variable of the associated metric is determined, and the assessment of the operating state is carried out on the basis of a change in the correlation variable in the said period. [8] Hydroelectric power plant with a structure-borne sound sensor and means designed to carry out the steps of the method according to one of the preceding claims. [9] A computer program comprising instructions causing the apparatus of the preceding claim to carry out the method steps of any one of claims 1 to 7. [10] A computer-readable medium on which the computer program according to the preceding claim is stored.
Citation Information
Patent Citations
Water turbine cavitation monitoring and analyzing system
CN203629869U
method for detecting cavitation
DE102015201203A1
CN000203629869U
