Water turbine cavitation state monitoring method, product and equipment based on cavitation characteristic parameter perception
By setting up an acoustic emission sensor on the turbine, using Fourier decomposition and cross-fuzzy entropy calculation, the cavitation feature vector is constructed, which solves the problem of accuracy in cavitation state monitoring of the turbine, and early warning and maintenance support for potential faults is achieved.
Patent Information
- Application Number
- CN202410989327.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-12
AI Technical Summary
The prior art is difficult to accurately monitor and predict the cavitation status of the turbine and its degree of damage, resulting in equipment damage, unstable operation, reduced performance and increased maintenance costs.
Using a method based on cavitation characteristic parameter perception, acoustic emission sensors are set on the turbine to obtain the acoustic emission signals under different operating conditions, empirical Fourier decomposition and cross-fuzzy entropy calculation are carried out, and cavitation characteristic vectors are constructed and their analysis is analyzed. Change to determine the cavitation state.
It improves the accuracy of the pilot cavitation status monitoring, realizes early warning of potential faults, and provides important decision-making support for maintenance and maintenance.
Smart Images

Figure CN118934399A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of hydraulic cavitation, and in particular to a method, product and equipment for monitoring the cavitation state of a water turbine based on the perception of cavitation characteristic parameters. Background Art
[0002] For turbines that use high-pressure, high-flow fluids as working media, once cavitation occurs, it may cause vibration of the unit, leading to increased noise, thereby changing the flow characteristics of the unit and even causing cavitation, ultimately threatening the safety and stable operation of the unit. Cavitation and cavitation inside the turbine will not only accelerate the damage of the equipment, but also lead to unstable operation, reduced performance, increased operation and maintenance costs, and increased repair costs. Therefore, in order to increase the service life of the equipment and reduce the damage caused by cavitation, it is crucial to monitor the cavitation status of large units and predict the extent of their damage.
[0003] As an important branch of the field of hydraulic cavitation, the study of turbine cavitation has been a hot topic in scientific research and engineering practice since the cavitation phenomenon was discovered in 1891. Historically, the formation, characteristics, monitoring, diagnosis and prevention of cavitation have been extensively studied. Although many researchers at home and abroad have conducted in-depth research on the generation principle, numerical simulation, signal characteristics and identification methods of cavitation, as well as turbine cavitation monitoring technology, the current research results are mainly concentrated on model experiments. There is still a lack of detailed prototype test reports on the field, especially for large turbines, which makes it impossible to accurately judge the cavitation state of the turbine and its degree of damage. Summary of the invention
[0004] The purpose of the present invention is to provide a method, product and equipment for monitoring the cavitation state of a water turbine based on the perception of cavitation characteristic parameters, so as to improve the accuracy of monitoring the cavitation state of the water turbine.
[0005] To achieve the above object, the present invention provides the following solutions:
[0006] A method for monitoring the cavitation state of a hydraulic turbine based on the perception of cavitation characteristic parameters, comprising:
[0007] A plurality of acoustic emission sensors are arranged on the turbine to obtain a plurality of groups of acoustic emission signals of each acoustic emission sensor under different operating conditions;
[0008] Performing empirical Fourier decomposition on multiple groups of acoustic emission signals to obtain components of the acoustic emission signals in the time domain;
[0009] Based on multiple components, the cross fuzzy entropy is calculated, and the cross fuzzy entropy of each component is obtained as the cavitation characteristic parameter to construct the cavitation characteristic vector;
[0010] The changes of cavitation characteristic vectors under different operating conditions are analyzed to determine the cavitation state of the turbine.
[0011] Optionally, the step of arranging a plurality of acoustic emission sensors on the turbine to obtain a plurality of groups of acoustic emission signals from each acoustic emission sensor under different operating conditions specifically includes:
[0012] An acoustic emission sensor is arranged at the tailwater gate of the turbine;
[0013] An acoustic emission sensor is arranged every 120° along the edge of the top cover of the turbine;
[0014] Each acoustic emission sensor is used to obtain multiple groups of acoustic emission signals under different operating conditions; the different operating conditions include no-load operating conditions and operating conditions under different loads.
[0015] Optionally, performing empirical Fourier decomposition on the multiple groups of acoustic emission signals to obtain components of the acoustic emission signals in the time domain specifically includes:
[0016] Performing Fourier transform on the acoustic emission signal to obtain the Fourier spectrum of the acoustic emission signal;
[0017] The Fourier spectrum is segmented by using an improved segmentation technique to obtain multiple continuous frequency segments after segmentation;
[0018] Construct an ideal filter bank based on the boundaries of each frequency band after segmentation;
[0019] The Fourier spectrum is filtered using an ideal filter bank to obtain a filtered signal;
[0020] Perform inverse Fourier transform on the filtered signal to obtain the components of the acoustic emission signal in the time domain.
[0021] Optionally, performing Fourier transform on the acoustic emission signal to obtain the Fourier spectrum of the acoustic emission signal specifically includes:
[0022] The formula for the acoustic emission signal f(t) is Perform Fourier transform to obtain the Fourier spectrum of the acoustic emission signal Among them, ω is frequency and t is time.
[0023] Optionally, the adopting of an improved segmentation technique to segment the Fourier spectrum to obtain a plurality of continuous frequency segments after segmentation specifically includes:
[0024] Using improved segmentation technology, the formula Calculate the boundary ω of the nth frequency bin in the Fourier spectrum n ; Wherein, N is the number of frequency segments; 1≤n≤N; Indicates frequency Ω nand frequency Ω n+1 The Fourier spectrum amplitude between ;
[0025] The Fourier spectrum is segmented based on the boundary of each frequency segment in the Fourier spectrum to obtain a plurality of continuous frequency segments after segmentation.
[0026] Optionally, constructing an ideal filter bank based on the boundary of each divided frequency segment specifically includes:
[0027] Based on the boundary ω of the nth frequency segment after segmentation n Constructing an ideal filter bank
[0028] Optionally, the cross fuzzy entropy calculation is performed based on multiple components to obtain the cross fuzzy entropy of each component as a cavitation feature parameter and construct a cavitation feature vector, specifically including:
[0029] Select multiple components containing main information to calculate cross fuzzy entropy and obtain the cross fuzzy entropy of each component;
[0030] The cross fuzzy entropy of the i-th component is used as the cavitation characteristic parameter e i , i = 1, 2, ..., K, construct the cavitation feature vector H = [e1, e2, ..., e K ]; where K is the number of components.
[0031] Optionally, the selecting of multiple components containing main information to perform cross fuzzy entropy calculation to obtain the cross fuzzy entropy of each component specifically includes:
[0032] Select multiple components containing the main information as the time series {u(i), 1≤i≤M} and {v(j), 1≤j≤M}, where i represents the i-th point in the time series; j represents the j-th point in the time series; M is the total number of points in the time series;
[0033] Given an embedding dimension m, construct Mm vectors {u(i),u(i+1),…,u(i+m-1),1≤i≤Mm} and Mm vectors {v(j),v(j+1),…,v(j+m-1),1≤j≤Mm} respectively;
[0034] Given an embedding dimension m, construct an m-dimensional vector corresponding to the time series {u(i), 1≤i≤M} The m-dimensional vector corresponding to the time series {v(j), 1≤j≤M}
[0035] calculate and The distance between in, It means finding the maximum value of m |u(i+k)-v(j+k)|; u(i+k) is the kth vector of the constructed Mm vectors {u(i),u(i+1),…,u(i+m-1),1≤i≤Mm}; v(j+k) is the kth vector of the constructed Mm vectors {v(j),v(j+1),…,v(j+m-1),1≤j≤Mm};
[0036] based on and Through the fuzzy function Determine the similarity between the m-dimensional vectors corresponding to each time series Among them, the fuzzy function is an exponential function, r and n are the width and gradient of the exponential function boundary respectively;
[0037] according to Define cross fuzzy entropy CFE(m,n,r,M), in,
[0038] is the similarity between the m+1 dimensional vectors corresponding to each time series, thereby obtaining the cross fuzzy entropy of each component.
[0039] A computer program product comprises a computer program, which, when executed by a processor, implements the method for monitoring the cavitation state of a water turbine based on the perception of cavitation characteristic parameters.
[0040] A computer device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for monitoring the cavitation state of a water turbine based on the perception of cavitation characteristic parameters.
[0041] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0042] The present invention provides a method, product and equipment for monitoring the cavitation state of a turbine based on cavitation characteristic parameter perception. The method extracts the cavitation signal characteristics (cavitation characteristic parameters) of the turbine by combining empirical Fourier decomposition and cross fuzzy entropy. Based on the accurately extracted cavitation characteristic parameters of the turbine, the accuracy of the cavitation state monitoring of the turbine is improved, and early warning of potential faults can be achieved, providing important decision support for maintenance and overhaul. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0044] Figure 1 A schematic flow chart of a method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception provided by the present invention;
[0045] Figure 2 A schematic diagram of the variation trend of the cavitation characteristic parameter e1 of the acoustic emission measuring point at the tailwater gate under different operating conditions provided by an embodiment of the present invention;
[0046] Figure 3 A schematic diagram of the change trend of the cavitation characteristic parameter e2 of the acoustic emission measurement point at the tailwater gate under different operating conditions provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0048] Cavitation is a crucial technical challenge in the design and operation of turbines, which requires special attention. Identifying and analyzing cavitation signals and deriving the cavitation law of turbine operation are urgent problems that need to be solved for the safety and efficiency of hydropower units. Therefore, the purpose of the present invention is to provide a method, product and equipment for monitoring the cavitation state of turbines based on the perception of cavitation characteristic parameters, so as to improve the accuracy of monitoring the cavitation state of turbines.
[0049] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0050] The turbine cavitation state monitoring method based on cavitation characteristic parameter perception of the present invention adopts a stacked ensemble learning method. The stacked ensemble learning method mainly trains multiple base learner models and uses their prediction results as the input of the meta-learner model to find the optimal combination of base learners, so as to fully mine the original data set and comprehensively utilize the learning advantages of each base learner, thereby improving the regression accuracy.
[0051] Specifically, the present invention adopts support vector machine and random tree method as base learners, and adopts adaptive neural network fuzzy inference system method as meta learner, and the specific process of the constructed stacked ensemble learning framework is as follows:
[0052] (1) Divide the obtained sample data set (X, Y) into a training set (X tr , Y tr ) and the test set (X te , Y te ); X = {x i , i = 1, 2, 3, ..., n}, where n is the total number of samples, x i = {active power, working head}, represents the input parameters of the ith sample; Y = {y i ,i=1,2,…,n}, where y i represents the response value corresponding to the i-th sample, i.e., the vibration parameter value of the unit; (X tr , Y tr )={(x i ,y i ), i = 1, 2, 3, ..., nk}, where 1 ≤ k < n, indicating that nk samples are selected as the training set; (X te , Y te )={(x i ,y i ), i=1, 2, 3, ..., k}, indicating that k samples are selected as the test set.
[0053] (2) During the training process, each base learner is trained to obtain the regression result of each base learner for the entire original training set. The prediction result P of each base learner is obtained. tr The new training set (P tr , Y tr ), and used as the training sample of the meta-learner, where P tr = {P tr-支持向量机 , P tr-随机树}, P tr-支持向量机 is the prediction result of the support vector machine, P tr-随机树 is the prediction result of the random tree.
[0054] (3) During the test, each trained base learner is used to obtain the prediction value P corresponding to the original test set. te , and construct a new test set (P te , Y te ) is used to test the trained meta-learner, where P te = {P te-支持向量机 , P te-随机树}. The final regression result F is obtainedte ={f i , i = 1, 2, 3, ..., k}, where f i represents the i-th regression result, which is obtained through the meta-learner.
[0055] Compared with traditional analysis techniques, acoustic emission detection technology has higher mutation and sensitivity and can detect the occurrence of cavitation earlier. Therefore, it is an accurate monitoring method for determining the starting point of cavitation and evaluating the relative intensity of cavitation.
[0056] like Figure 1 As shown, the method for monitoring the cavitation state of a turbine based on cavitation characteristic parameter perception disclosed in the present invention includes:
[0057] Step 1: multiple acoustic emission sensors are set on the turbine to obtain multiple groups of acoustic emission signals from each acoustic emission sensor under different operating conditions.
[0058] When using multiple sets of high-frequency acoustic emission signals to monitor the relationship between the cavitation acoustic signal characteristics (cavitation characteristic parameters) and the cavitation coefficient during the operation of the turbine, selecting the appropriate location of the acoustic emission measurement point can effectively improve the reliability of the turbine cavitation state monitoring. As a preferred embodiment, the present invention arranges an acoustic emission sensor at the tailwater gate of the turbine and arranges an acoustic emission sensor every 120° along the edge of the top cover of the turbine. Each acoustic emission sensor is used to obtain multiple sets of acoustic emission signals under different operating conditions. Among them, different operating conditions include no-load operating conditions and operating conditions under different loads.
[0059] Step 2: Perform empirical Fourier decomposition on multiple groups of acoustic emission signals to obtain the components of the acoustic emission signals in the time domain.
[0060] The present invention provides a method for accurately sensing characteristic parameters of cavitation of a hydraulic turbine based on empirical Fourier decomposition and cross fuzzy entropy coupling. It is necessary to perform empirical Fourier decomposition on the acoustic emission signal of the hydraulic turbine under different operating conditions, and use the obtained components to construct the cavitation state quantity of the hydraulic turbine. The empirical Fourier decomposition process of the acoustic emission signal is specifically as follows.
[0061] The step 2 specifically includes:
[0062] Step 2.1: Perform Fourier transform on the acoustic emission signal to obtain the Fourier spectrum of the acoustic emission signal.
[0063] Specifically, the decomposed signal, that is, the acoustic emission signal f(t), is subjected to Fourier transform to obtain the Fourier spectrum of the acoustic emission signal:
[0064]
[0065] Where f(t) is the representation of the acoustic emission signal in the time domain; ω is the frequency; t is the time; and j is the imaginary unit.
[0066] Step 2.2: Use the improved segmentation technology to segment the Fourier spectrum to obtain multiple continuous frequency segments after segmentation.
[0067] Based on the Fourier spectrum obtained in step 2.1 Determine the boundary ω of the nth frequency segment in the Fourier spectrum using improved segmentation technology n Specifically, in the improved segmentation technique, [0, π] is segmented into N consecutive frequency segments. During the sorting process, the amplitude of the Fourier spectrum at ω = 0 and ω = π and its local maximum are identified and extracted 200 to a series. All the magnitudes in the sequence segmented into N consecutive frequency segments are sorted in descending order. The frequencies corresponding to the first N maximum values in the sorted sequence are represented by [Ω1, Ω2, ...Ω N ] indicates. In addition, Ω0=0 and Ω N+1 =π. Among them, the boundary of the nth frequency segment ω n The specific formula is:
[0068]
[0069] Where, 1≤n≤N; Indicates frequency Ω n and frequency Ω n+1 After determining the boundary of each frequency segment in the Fourier spectrum, the Fourier spectrum is segmented based on the boundary of each frequency segment in the Fourier spectrum to obtain multiple continuous frequency segments after segmentation.
[0070] Step 2.3: Construct an ideal filter bank based on the boundaries of each frequency segment after segmentation.
[0071] Based on the boundary ω of the nth frequency segment after segmentation obtained in step 2.2 n , an ideal filter bank is constructed by the following formula
[0072]
[0073] Step 2.4: Use an ideal filter bank to filter the Fourier spectrum to obtain a filtered signal.
[0074] Use the ideal filter bank obtained in step 2.3 Fourier spectrum Filter and get the filtered signal in the frequency domain
[0075]
[0076] Step 2.5: Perform inverse Fourier transform on the filtered signal to obtain the components of the acoustic emission signal in the time domain.
[0077] The filtered signal obtained in step 2.4 Perform inverse Fourier transform to obtain the component f of the original acoustic emission signal in the time domain n (t).
[0078]
[0079] Step 3: Calculate the cross fuzzy entropy based on multiple components, obtain the cross fuzzy entropy of each component as the cavitation feature parameter, and construct the cavitation feature vector.
[0080] After obtaining the components of each group of acoustic emission signals in the time domain through step 2, multiple components containing main information from the components corresponding to the multiple groups of acoustic emission signals are selected to perform cross fuzzy entropy calculation to obtain the cross fuzzy entropy of each component.
[0081] For the cross fuzzy entropy, its principle and calculation formula are specifically shown as follows.
[0082] Cross fuzzy entropy is generalized from cross sample entropy and can measure the synchronization between time series. The more synchronized the two time series are, the smaller the value of their cross fuzzy entropy is. The specific algorithm is as follows:
[0083] S1: For two normalized time series {u(i), 1≤i≤M} and {v(j), 1≤j≤M}, where u(i) = f n (t); i represents the i-th point in the time series; j represents the j-th point in the time series; M is the total number of points in the time series. Given an embedding dimension m, construct Mm vectors {u(i),u(i+1),…,u(i+m-1),1≤i≤Mm} and {v(j),v(j+1),…,v(j+m-1),1≤j≤Mm} respectively.
[0084] For the time series {u(i), 1≤i≤M}, we can sequentially construct an m-dimensional vector For the time series {v(j), 1≤j≤M}, we can sequentially construct an m-dimensional vector definition and The distance between for:
[0085]
[0086] In the formula, It means to find the maximum value of m |u(i+k)-v(j+k)|. l=0,1,...,m-1.
[0087] S2: and The synchronization or similarity between Through the fuzzy function calculate:
[0088]
[0089] In the formula, the fuzzy function is an exponential function:
[0090]
[0091] Where r and n are the width and gradient of the exponential function boundary, respectively.
[0092] S3: Define functions Specifically:
[0093]
[0094] Then expand the embedding dimension m to m+1, and we get:
[0095]
[0096] S4: Define the cross fuzzy entropy CFE(m,n,r) as:
[0097]
[0098] In fact, the number of data points M is limited, and when the data length is M, the result obtained by the above steps is the estimated value of CFE, which can be written as:
[0099]
[0100] Based on the above steps, the cross fuzzy entropy of multiple components containing the main information is calculated. After the cross fuzzy entropy of each component is obtained, the cross fuzzy entropy of the i-th component is used as the cavitation characteristic parameter e i , i = 1, 2, ..., K, construct the cavitation feature vector H = [e1, e2, ..., e K ]; where K is the number of components.
[0101] Step 4: Analyze the changes of cavitation characteristic vectors under different operating conditions and determine the cavitation state of the turbine.
[0102] After constructing the cavitation characteristic vector H through the above steps, the change of the cavitation characteristic vector H at each acoustic emission measuring point under different operating conditions is analyzed, and a chart is drawn according to the change of the cavitation characteristic vector to determine the cavitation state of the turbine.
[0103] As a specific application example, the research work was carried out on a large Francis turbine. The rated flow of the turbine is 381.6m 3 / s, rated head is 175.5m, rated output is 609MW, speed is 125rpm, and there are 16 blades. It can be seen from the actual operation on site that the main cavitation types that affect the operation of the mixed flow turbine are: impeller water inlet cavitation, rim gap cavitation and tailwater tube vortex cavitation. Water inlet edge cavitation and gap cavitation will cause cavitation of the blades, and tailwater tube vortex cavitation will affect the stable operation of the turbine. With the continuous development of unit cavitation, the efficiency and output of the turbine will also be greatly affected. Therefore, the test process of the present invention mainly monitors the positions where the above cavitation may occur, such as the tailwater gate as an acoustic emission measuring point, and the top cover is provided with three acoustic emission measuring points, specifically, an acoustic emission measuring point is arranged every 120° on the edge of the top cover.
[0104] Four acoustic emission sensors were used for monitoring in the test, and equipment that met the requirements was selected for field testing to establish a complete test system. Among them, the acoustic emission sensor uses a resonant piezoelectric sensor, whose operating frequency range is 100-400kHz and the center frequency is 170kHz. The SAEU3H-04 acquisition card is used to control the sampling of acoustic wave signals, and its maximum sampling frequency is 10MB / s. The acquisition frequency in the test is set to 1000000Hz, and 4 channels of data are collected at the same time.
[0105] Whether cavitation occurs during turbine operation and the identification of interference noise during the test are issues that must be considered before data collection. Studies have shown that cavitation occurs in Francis turbines under various operating conditions. Generally speaking, the hydrodynamic noise and mechanical noise in the test are low-frequency noises, which have little effect on the high-frequency part of the cavitation data. In order to identify whether electromagnetic force noise exists, data under no-load and unprocessed conditions were collected in the test for comparison with other operating conditions.
[0106] 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 increased by 10% of the rated load. Due to the fluctuation of the load, the specific load value may vary). Considering that the rotation frequency of the turbine is 2.08Hz, the data acquisition system continuously collects data with a sampling period of 3.3 seconds. The test time for each working condition is nearly 2 minutes.
[0107] During the analysis, 20 sets of data were taken for each working condition. After empirical Fourier decomposition of the 20 sets of acoustic emission signals, n components (f1(t), f2(t), ..., f n(t)), in this example, the first and fifth components containing the main information are taken. After calculating the cross fuzzy entropy of each component, the cross fuzzy entropy of the component is obtained as the cavitation characteristic parameter e i , i = 1, 2. Construct the cavitation feature vector H = [e1, e2] to prepare for the subsequent state recognition.
[0108] The values of cavitation characteristic parameters e1 and e2 at the tailwater gate acoustic emission measurement point under different operating conditions of the turbine are shown in Table 1. Figure 2 and Figure 3 shown.
[0109] Table 1 Cavitation characteristic parameters
[0110]
[0111] Depend on Figure 2 The data in show that cavitation occurs when the unit operates in the load range of 60% to 100%, and from 70% load, the monitored cavitation characteristic parameters show an increasing trend, which indicates that under fixed head conditions, with the increase of flow rate, the intensity of vibration and cavitation also gradually increases. It can be seen that for large-scale Francis turbines, the present invention has carried out cavitation experiments covering a variety of working conditions, and the results show that the outlet area of the Francis turbine, that is, the tailwater inlet gate, is an ideal position for monitoring runner cavitation. Under the same working head conditions, with the increase of output power, the cavitation intensity of the turbine generally shows an increasing trend. The analysis results of this specific application example show that the turbine cavitation state monitoring method based on cavitation characteristic parameter perception of the present invention can effectively evaluate the cavitation state of the turbine by analyzing the changes in the characteristic vector under different operating conditions of the turbine, and can achieve early warning of potential faults, providing important decision support for maintenance and repair.
[0112] In summary, the turbine cavitation state monitoring method based on cavitation characteristic parameter perception of the present invention selects appropriate acoustic emission measurement point positions, uses high-frequency acoustic emission signals to monitor the relationship between cavitation acoustic signal characteristics and cavitation coefficients during turbine operation, and evaluates the severity of the cavitation state accordingly. It is an effective method.
[0113] In addition, the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of a method for monitoring the cavitation state of a turbine based on the perception of cavitation characteristic parameters.
[0114] Furthermore, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for monitoring the cavitation state of a turbine based on the perception of cavitation characteristic parameters.
[0115] Furthermore, the present invention also provides a computer device, including a processor, a memory, an input / output interface (Input / Output, referred to as I / O), a communication interface, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the turbine cavitation state monitoring method based on cavitation characteristic parameter perception.
[0116] The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store pending transactions. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the turbine cavitation state monitoring method based on cavitation characteristic parameter perception is implemented.
[0117] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.
[0118] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ReadOnlyMemory, ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (Magnetoresistive RandomAccess Memory, MRAM), ferroelectric random access memory (Ferroelectric RandomAccess Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (RandomAccess Memory, RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited thereto. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited thereto.
[0119] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0120] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of the present invention. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A method for monitoring the cavitation state of a hydraulic turbine based on the perception of cavitation characteristic parameters, characterized in that: include: A plurality of acoustic emission sensors are arranged on the turbine to obtain a plurality of groups of acoustic emission signals of each acoustic emission sensor under different operating conditions; Performing empirical Fourier decomposition on multiple groups of acoustic emission signals to obtain components of the acoustic emission signals in the time domain; Based on multiple components, the cross fuzzy entropy is calculated, and the cross fuzzy entropy of each component is obtained as the cavitation characteristic parameter to construct the cavitation characteristic vector; The changes of cavitation characteristic vectors under different operating conditions are analyzed to determine the cavitation state of the turbine.
2. The method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception according to claim 1, characterized in that: The method of arranging a plurality of acoustic emission sensors on the turbine to obtain a plurality of groups of acoustic emission signals of each acoustic emission sensor under different operating conditions specifically includes: An acoustic emission sensor is arranged at the tailwater gate of the turbine; An acoustic emission sensor is arranged every 120° along the edge of the top cover of the turbine; Each acoustic emission sensor is used to obtain multiple groups of acoustic emission signals under different operating conditions; the different operating conditions include no-load operating conditions and operating conditions under different loads.
3. The method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception according to claim 1, characterized in that: The empirical Fourier decomposition of the multiple groups of acoustic emission signals is performed to obtain components of the acoustic emission signals in the time domain, specifically including: Performing Fourier transform on the acoustic emission signal to obtain the Fourier spectrum of the acoustic emission signal; The Fourier spectrum is segmented by using an improved segmentation technique to obtain multiple continuous frequency segments after segmentation; Construct an ideal filter bank based on the boundaries of each frequency band after segmentation; The Fourier spectrum is filtered using an ideal filter bank to obtain a filtered signal; Perform inverse Fourier transform on the filtered signal to obtain the components of the acoustic emission signal in the time domain.
4. The method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception according to claim 3 is characterized in that: The Fourier transform of the acoustic emission signal to obtain the Fourier spectrum of the acoustic emission signal specifically includes: The formula for the acoustic emission signal f(t) is Perform Fourier transform to obtain the Fourier spectrum of the acoustic emission signal Among them, ω is frequency and t is time.
5. The method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception according to claim 4 is characterized in that: The improved segmentation technology is used to segment the Fourier spectrum to obtain a plurality of continuous frequency segments after segmentation, specifically including: Using improved segmentation technology, the formula Calculate the boundary ω of the nth frequency bin in the Fourier spectrum n ; Wherein, N is the number of frequency segments; 1≤n≤N; Indicates frequency Ω n and frequency Ω n+1 The Fourier spectrum amplitude between ; The Fourier spectrum is segmented based on the boundary of each frequency segment in the Fourier spectrum to obtain a plurality of continuous frequency segments after segmentation.
6. The method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception according to claim 5, characterized in that: The step of constructing an ideal filter bank based on the boundary of each divided frequency segment specifically includes: Based on the boundary ω of the nth frequency segment after segmentation n Constructing an ideal filter bank 7. The method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception according to claim 1, characterized in that: The cross fuzzy entropy calculation is performed based on multiple components to obtain the cross fuzzy entropy of each component as a cavitation feature parameter and construct a cavitation feature vector, specifically including: Select multiple components containing main information to calculate cross fuzzy entropy, and obtain the cross fuzzy entropy of each component; The cross fuzzy entropy of the i-th component is used as the cavitation feature parameter e i , i = 1, 2, ..., K, construct the cavitation feature vector H = [e1, e2, ..., e K ]; where K is the number of components.
8. The method for monitoring the cavitation state of a hydraulic turbine based on cavitation characteristic parameter perception according to claim 7, characterized in that: The selecting of multiple components containing main information to perform cross fuzzy entropy calculation to obtain the cross fuzzy entropy of each component specifically includes: Select multiple components containing the main information as the time series {u(i), 1≤i≤M} and {v(j), 1≤j≤M}, where i represents the i-th point in the time series; j represents the j-th point in the time series; M is the total number of points in the time series; Given an embedding dimension m, construct Mm vectors {u(i),u(i+1),…,u(i+m-1),1≤i≤Mm} and Mm vectors {v(j),v(j+1),…,v(j+m-1),1≤j≤Mm} respectively; Given an embedding dimension m, construct an m-dimensional vector corresponding to the time series {u(i), 1≤i≤M} The m-dimensional vector corresponding to the time series {v(j), 1≤j≤M} calculate and The distance between in, It means finding the maximum value of m |u(i+k)-v(j+k)|; u(i+k) is the kth vector of the constructed Mm vectors {u(i),u(i+1),…,u(i+m-1),1≤i≤Mm}; v(j+k) is the kth vector of the constructed Mm vectors {v(j),v(j+1),…,v(j+m-1),1≤j≤Mm}; based on and Through the fuzzy function Determine the similarity between the m-dimensional vectors corresponding to each time series Among them, the fuzzy function is an exponential function, r and n are the width and gradient of the exponential function boundary respectively; according to Define cross fuzzy entropy CFE(m,n,r,M), in, is the similarity between the m+1 dimensional vectors corresponding to each time series, thereby obtaining the cross fuzzy entropy of each component.
9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for monitoring the cavitation state of a water turbine based on the perception of cavitation characteristic parameters as described in any one of claims 1 to 8 is implemented.
10. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the turbine cavitation state monitoring method based on cavitation characteristic parameter perception as described in any one of claims 1 to 8.
Citation Information
Cited By
Method for evaluating stability of mixed-flow water turbine by fusing impact characteristic values of sound and vibration signals
CN119878421A
Stability evaluation method of mixed-flow hydraulic turbine by fusing shock characteristic values of acoustic and vibration signals
CN119878421B
