Water turbine cavitation state analysis method
By using covariance decomposition, Hammerseley sequence and cross-fuzzy entropy calculation, combined with high-frequency acoustic emission signal detection, the problem of cavitation status assessment of large hydro turbines was solved, enabling accurate monitoring and early warning of cavitation phenomena, and improving the safety and operating efficiency of the unit.
Patent Information
- Application Number
- CN202410633323.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-21
- Publication Date
- 2024-11-08
AI Technical Summary
Existing technologies are insufficient to effectively assess the cavitation state of large hydro turbines, especially given the lack of detailed studies in field prototype tests, which affects the safe and stable operation of the units.
A method for analyzing the cavitation state of a water turbine is adopted. Through covariance decomposition, uniform sampling of Hammerseley sequence, ellipsoid projection and cross-fuzzy entropy calculation, the cavitation state of the water turbine is monitored in real time. The relationship between cavitation noise and cavitation coefficient is detected by high-frequency acoustic emission signal.
It enables effective assessment of the cavitation status of the turbine, provides early fault warning and maintenance reference, and improves the safety and operating efficiency of the unit.
Smart Images

Figure CN118916601A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of water turbine cavitation state analysis, and in particular to a water turbine cavitation state analysis method. Background Art
[0002] For turbines that use high-pressure fluid as the working medium, once cavitation occurs, it may cause a series of problems, including increased vibration and noise of the unit, changes in flow characteristics, and cavitation, which will have an adverse effect on the safe and stable operation of the unit. Therefore, in order to extend the service life of unit equipment and reduce the damage caused by cavitation, it is crucial to monitor the cavitation of large units during operation and evaluate the degree of damage.
[0003] The study of turbine cavitation is an important branch in the field of hydraulic cavitation. Since its discovery in 1891, it has been a hot topic in scientific research and engineering practice. People have conducted in-depth research on the formation mechanism, dynamic behavior, monitoring diagnosis and preventive measures of cavitation. The cavitation and erosion phenomena of turbines will not only lead to increased damage to the equipment and unstable operation, but also cause performance degradation, reduced operating efficiency and increased maintenance costs. Although many researchers at home and abroad have conducted extensive research on the generation principle of cavitation, numerical simulation, signal feature recognition and turbine cavitation monitoring technology, the current research results are mainly concentrated on model tests, and detailed prototype test reports for the field, especially large turbines, are relatively lacking.
[0004] In view of the above problems, the present invention provides a method for analyzing the cavitation state of a water turbine, which can effectively evaluate the cavitation state of the water turbine. Summary of the invention
[0005] The invention provides a method for analyzing the cavitation state of a water turbine which can effectively evaluate the cavitation state of the water turbine.
[0006] The object of the present invention is to provide a method for analyzing the cavitation state of a hydraulic turbine, the method comprising the following steps:
[0007] Step 1: Given a ternary signal x(t), perform covariance decomposition E{xx T}=VΛV T , where V is the eigenvector matrix, Λ=diag{λ1,λ2,λ3} is the eigenvalue matrix, and the eigenvalues are λ1>λ2>λ3;
[0008] Step 2: Use the Hammerseley sequence to uniformly sample the sphere and determine the azimuth angle φ of its projection H and the inclination angle θ H , thereby determining the Cartesian coordinates of the uniformly sampled sphere;
[0009] Step 3: Determine the non-uniform projection on the sphere by constructing an ellipsoid with the following parameters The specific calculation formula is:
[0010]
[0011] Among them, the rotating ellipsoid Make the direction of highest curvature be sampled
[0012] Step 4: Calculate the cross fuzzy entropy of each component to obtain the entropy CFE of the component as the characteristic parameter e i , i = 1, 2, 3, construct the eigenvalue vector: H = [e1, e2, e3];
[0013] Step 5: By analyzing the changes in the characteristic vectors under different operating conditions of the turbine, the cavitation state of the turbine can be obtained in real time.
[0014] Furthermore, the specific calculation steps of the cross fuzzy entropy in step 4 are as follows:
[0015] Step 1: For two normalized time series {i(i), 1≤i≤N} and {v(j), 1≤j≤N}, given the embedding dimension m, construct Nm vectors {u(i), v(i+1), ..., u(i+m-1), 1≤i≤NM} and {u(j), v(j+1), ..., v(j+m-1), 1≤j≤Nm} respectively, and define and The specific calculation formula for the distance between is:
[0016]
[0017]
[0018]
[0019]
[0020]
[0021] Step 2: and The synchronization or similarity between Through the fuzzy function Calculation, the specific calculation formula is:
[0022]
[0023]
[0024] Among them, the fuzzy function is an exponential function, r and n are the width and gradient of the exponential function boundary respectively.
[0025] Step 3: Define the specific calculation formula of the function:
[0026]
[0027] Step 4: Expand the embedding dimension m to m+1. The specific calculation formula is:
[0028]
[0029] Step 5: Define the specific calculation formula of cross fuzzy entropy:
[0030]
[0031] Step 6: The number of data points N is limited, and when the data length is N, the result obtained through the above steps is the estimated value of CFE, and the specific calculation formula is:
[0032]
[0033] The present invention has the following advantages: by conducting cavitation experiments on a large-scale mixed flow turbine prototype under various working conditions, the present invention found that, for a mixed flow turbine, the tailwater inlet gate is the best position to monitor the cavitation of the runner. The study also pointed out that under the same working head conditions, as the output power increases, the cavitation degree of the turbine tends to increase. By selecting appropriate measurement points and using high-frequency acoustic emission signals to detect the relationship between the cavitation noise generated during the operation of the turbine and the cavitation coefficient, the degree of cavitation can be evaluated.
[0034] The present invention can effectively evaluate the cavitation state of the turbine and realize early warning of potential failures of the unit, and also provides an important reference basis for maintenance and overhaul. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 It is a curve diagram of cavitation characteristic changes of the water gate acoustic emission measuring point under different operating conditions of the present invention. DETAILED DESCRIPTION
[0036] The present invention provides a method for analyzing the cavitation state of a hydraulic turbine, the method comprising the following steps:
[0037] Step 1: Given a ternary signal x(t), perform covariance decomposition E{xx T}=VΛV T , where V is the eigenvector matrix, Λ=diag{λ1,λ2,λ3} is the eigenvalue matrix, and the eigenvalues are λ1>λ2>λ3;
[0038] Step 2: Use the Hammerseley sequence to uniformly sample the sphere and determine the azimuth angle φ of its projection H and the inclination angle θ H , thereby determining the Cartesian coordinates of the uniformly sampled sphere;
[0039] Step 3: Determine the non-uniform projection on the sphere by constructing an ellipsoid with the following parameters The specific calculation formula is:
[0040]
[0041] Among them, the rotating ellipsoid Make the direction of highest curvature be sampled
[0042] Step 4: Calculate the cross fuzzy entropy of each component to obtain the entropy CFE of the component as the characteristic parameter e i , i = 1, 2, 3, construct the eigenvalue vector: H = [e1, e2, e3];
[0043] Step 5: By analyzing the changes in the characteristic vectors under different operating conditions of the turbine, the cavitation state of the turbine can be obtained in real time.
[0044] In this embodiment, the specific calculation steps of the cross fuzzy entropy in step 4 are as follows:
[0045] Step 1: For two normalized time series {u(i), 1≤i≤N} and {v(j), 1≤j≤N}, given the embedding dimension m, construct Nm vectors {u(i), u(i+1), ..., u(i+m-1), 1≤j≤NM} and {v(j), v(j+1), ..., v(j+m-1), 1≤j≤Nm} respectively, and define and The specific calculation formula for the distance between is:
[0046]
[0047]
[0048]
[0049]
[0050]
[0051] Step 2: and The synchronization or similarity between Through the fuzzy function Calculation, the specific calculation formula is:
[0052]
[0053]
[0054] Among them, the fuzzy function is an exponential function, r and n are the width and gradient of the exponential function boundary respectively.
[0055] Step 3: Define the specific calculation formula of the function:
[0056]
[0057] Step 4: Expand the embedding dimension m to m+1. The specific calculation formula is:
[0058]
[0059] Step 5: Define the specific calculation formula of cross fuzzy entropy:
[0060]
[0061] Step 6: The number of data points N is limited, and when the data length is N, the result obtained through the above steps is the estimated value of CFE, and the specific calculation formula is:
[0062]
[0063] In this embodiment, the actual operation conditions show that the cavitation forms that most significantly affect the operation of the mixed flow turbine include: cavitation at the water inlet edge of the impeller, cavitation at the rim gap, and vortex cavitation in the tailwater pipe. Among them, the water inlet edge cavitation and gap cavitation may cause cavitation of the blades, while the vortex cavitation in the tailwater pipe may interfere with the smooth operation of the turbine. With the further development of cavitation inside the turbine, the efficiency and output will be significantly affected.
[0064] In this embodiment, the research work is carried out on a large-scale Francis turbine with a rated flow of 381.6m 3 / s, rated head is 175.5m, rated output is 609MW, speed is 125rpm, and there are 16 blades. Four acoustic emission sensors were used for monitoring in the test. Equipment that met the requirements was selected for field testing, and a complete test system was established. The acoustic emission sensor uses a resonant piezoelectric sensor with an operating frequency range of 100-400kHz and a center frequency of 170kHz. The SAEU3H-04 acquisition card is used to control the sampling of acoustic wave signals, with a maximum sampling frequency of 10MB / s. The acquisition frequency was set to 1000000Hz in the test, and 4 channels of data were collected at the same time.
[0065] In this embodiment, during the operation of the turbine, it is necessary to consider and determine in advance whether cavitation occurs and how to identify interference noise during the test. The research results show that cavitation may occur in a Francis turbine under all working conditions. Usually, the hydrodynamic and mechanical noises generated in the test belong to the low-frequency category, which has little effect on the high-frequency cavitation data. In order to identify whether there is noise caused by electromagnetic force, data under no-load conditions are also collected in the test for comparative analysis with data under other working conditions. During the test, the gross head of the turbine was 174m, and the load gradually increased from low to high, first no-load, then 60MW, 120MW, 180MW, ... 540MW and 600MW (the load has increased by 10% of the rated load). Considering that the rotation frequency of the turbine is 2.08 Hz, the data acquisition system takes 3.3 seconds as a sampling period, and continuously collects data. The test time for each working condition is nearly 2 minutes. During the analysis, 20 sets of data are taken for each working condition, and their characteristic parameter values are obtained according to the method proposed in the present invention. The characteristic parameter changes of the data collected by the tailwater gate acoustic emission measurement point under different operating conditions of the turbine are shown in Table 1 and Figure 1 shown.
[0066] In this embodiment, by analyzing Figure 1 It can be seen from Table 1 that cavitation occurs when the unit is running at 80% to 100% load. From 80% load, the energy value recorded at the measuring point shows an increasing trend, which shows that under the same working head conditions, with the increase of flow rate, vibration and cavitation are gradually intensified. It can be seen that if a suitable measurement position is selected, it is effective to use high-frequency acoustic emission signals to detect the relationship between cavitation noise and cavitation coefficient during turbine operation and judge the degree of cavitation accordingly. Compared with traditional analysis technology, acoustic emission detection method has higher mutation and sensitivity, can detect the occurrence of cavitation earlier, and is a more accurate monitoring method, especially suitable for determining the initial occurrence point of cavitation and evaluating the relative intensity of cavitation.
[0067] Table 1 shows the characteristic parameters of cavitation signals.
[0068] Although the specific implementation of the present invention is described in detail in conjunction with the drawings, it should not be understood as limiting the scope of protection of this patent. Within the scope described in the claims, various modifications and variations that can be made by those skilled in the art without creative work still fall within the scope of protection of this patent.
[0069] Table 1
[0070]
Claims
1. A method for analyzing cavitation state of a hydraulic turbine, characterized in that: The method comprises the following steps: Step 1: Given a ternary signal x(t), perform covariance decomposition E{xx T }=VΛV T , where V is the eigenvector matrix, Λ=diag{λ1,λ2,λ3} is the eigenvalue matrix, and the eigenvalues are λ1>λ2>λ3; Step 2: Use the Hammerseley sequence to uniformly sample the sphere and determine the azimuth angle φ of its projection H and the inclination angle θ H , thereby determining the Cartesian coordinates of the uniformly sampled sphere; Step 3: Determine the non-uniform projection on the sphere by constructing an ellipsoid with the following parameters The specific calculation formula is: Among them, the rotating ellipsoid Make the direction of highest curvature be sampled Step 4: Calculate the cross fuzzy entropy of each component to obtain the entropy CFE of the component as the characteristic parameter e i , i = 1, 2, 3, construct the eigenvalue vector: H = [e1, e2, e3]; Step 5: By analyzing the changes in the characteristic vectors under different operating conditions of the turbine, the cavitation state of the turbine can be obtained in real time.
2. A method for analyzing cavitation state of a hydraulic turbine according to claim 1, characterized in that: The specific calculation steps of the cross fuzzy entropy in step 4 are as follows: Step 1: For two normalized time series {u(i), 1≤i≤N} and {v(j), 1≤j≤N}, given the embedding dimension m, construct Nm vectors {u(i), u(i+1), ..., u(i+m-1), 1≤i≤NM} and {v(j), v(j+1), ..., v(j+m-1), 1≤j≤Nm} respectively, and define and The specific calculation formula for the distance between is: Step 2: and The synchronization or similarity between Through the fuzzy function Calculation, the specific calculation formula is: Among them, the fuzzy function is an exponential function, r and n are the width and gradient of the exponential function boundary respectively. Step 3: Define the specific calculation formula of the function: Step 4: Expand the embedding dimension m to m+1. The specific calculation formula is: Step 5: Define the specific calculation formula of cross fuzzy entropy: Step 6: The number of data points N is limited, and when the data length is N, the result obtained through the above steps is the estimated value of CFE, and the specific calculation formula is:
