Vibrating wire osmometer data processing method and system for reservoir dam

Abstract

The invention provides a vibrating wire osmometer data processing method and system for a reservoir dam. The method comprises the following steps: respectively acquiring a vibrating wire frequency signal and a temperature signal of a to-be-monitored area of a reservoir dam, determining a temperature feature vector according to the temperature signal, and performing multi-resolution spatial decomposition on the vibrating wire frequency signal according to the temperature feature vector to generate a vibrating wire dynamic feature value corresponding to each resolution space; according to the method, the vibration wire frequency signal and the temperature signal are combined to generate the vibration wire dynamic vector, and then the current state of the to-be-monitored area is determined according to the vibration wire dynamic vector, so that more accurate evaluation of the safety state of the to-be-monitored area of the reservoir dam is realized by comprehensively considering the combination of the vibration wire frequency signal and the temperature signal and utilizing multi-resolution spatial decomposition and feature extraction.

Description

Technical Field

[0001] The present application relates to data processing technology, and in particular to a vibrating-wire piezometer data processing method and system for reservoir dams. Background Art

[0002] In the field of reservoir dam safety monitoring, accurate and timely acquisition of the internal status information of the dam is crucial to ensure the safe operation of the dam. Traditionally, the monitoring of reservoir dams mainly relies on various sensors and equipment, among which the vibrating wire osmometer is a commonly used monitoring instrument. The vibrating wire osmometer can sense the changes in the osmotic pressure inside the dam and convert it into fluctuations in the frequency of the vibrating wire. By monitoring these frequency fluctuations, the osmotic pressure status inside the dam can be evaluated.

[0003] However, in the actual monitoring process, the change of the vibrating wire frequency signal is not only affected by the seepage pressure, but also by a variety of environmental factors, among which temperature is an important influencing factor. The change of temperature will cause the physical properties of the vibrating wire in the vibrating wire osmometer to change, thereby affecting the stability of the vibrating wire frequency, making it difficult for the simple vibrating wire frequency signal to accurately reflect the real seepage pressure state inside the dam. Summary of the invention

[0004] The present application provides a vibrating-wire piezometer data processing method and system for reservoir dams, which is used to comprehensively consider the combination of vibrating-wire frequency signals and temperature signals, and utilizes multi-resolution spatial decomposition and feature extraction to achieve an assessment of the safety status of the area to be monitored in the reservoir dam.

[0005] In a first aspect, the present application provides a vibrating wire piezometer data processing method for a reservoir dam, comprising: Respectively obtain a vibrating wire frequency signal and a temperature signal of the area to be monitored in the reservoir dam, wherein the vibrating wire frequency signal is data obtained by a vibrating wire osmometer; Determine a temperature characteristic vector according to the temperature signal, perform multi-resolution spatial decomposition on the vibrating string frequency signal according to the temperature characteristic vector, so as to generate corresponding vibrating string dynamic characteristic values ​​under each resolution space, so as to generate a vibrating string dynamic vector, wherein the temperature characteristic vector is associated with a decomposition depth of the multi-resolution spatial decomposition; The current state of the area to be monitored is determined according to the vibrating wire dynamic vector.

[0006] In the above scheme, the vibrating string frequency signal and temperature signal of the monitored area of ​​the reservoir dam are obtained respectively. The vibrating string frequency signal comes from the vibrating string osmometer, which can be used to sense the osmotic pressure change inside the dam and convert it into the fluctuation of the vibrating string frequency. The temperature signal reflects the temperature change inside the dam or the surrounding environment. This information is crucial for understanding the changing background of the vibrating string frequency signal. Next, the temperature characteristic vector is determined according to the temperature signal to capture the overall trend and fluctuation of the temperature change. Based on the temperature characteristic vector, the vibrating string frequency signal is further decomposed in multi-resolution space, where the decomposition depth is closely related to the temperature characteristic vector, ensuring the adaptability and pertinence of the signal decomposition. Subsequently, the corresponding vibrating string dynamic eigenvalues ​​under each resolution space are generated and spliced ​​to form a vibrating string dynamic vector to capture the dynamic change characteristics of the vibrating string frequency signal under different resolution spaces. Finally, the current state of the monitored area is determined according to the vibrating string dynamic vector, and the safety status of the dam is evaluated. Comprehensively considering the combination of the vibrating string frequency signal and the temperature signal, the multi-resolution spatial decomposition and feature extraction are used to evaluate the safety status of the monitored area of ​​the reservoir dam. This process not only improves the accuracy and efficiency of monitoring, but also provides strong technical support for the safe operation of reservoir dams.

[0007] Optionally, determining a temperature characteristic vector according to the temperature signal includes: Performing time series sampling on the temperature signal to determine a temperature sequence, and determining a temperature mean and a temperature standard deviation according to the temperature sequence; The temperature feature vector is determined according to the temperature mean, the temperature standard deviation, and the temperature sequence.

[0008] In the above scheme, by sampling the temperature signal in time series, a series of discrete temperature data points, namely the temperature series, can be obtained. The time series sampling not only captures the temperature changes inside the dam or the surrounding environment, but also retains the time sequence information of the temperature changes. Then, according to the obtained temperature series, the temperature mean and temperature standard deviation are calculated. The temperature mean reflects the average level of temperature within a certain time range, while the temperature standard deviation measures the degree of temperature fluctuation. These two statistics together constitute a comprehensive description of the temperature change characteristics. By calculating the temperature mean and standard deviation, we can more intuitively understand the overall trend and fluctuation characteristics of temperature changes. After obtaining the temperature mean, temperature standard deviation and original temperature series, the temperature feature vector is further determined based on this information. The temperature feature vector extracts the key information that can represent the temperature change characteristics by comprehensively considering the temperature mean, standard deviation and each data point in the temperature series. It not only retains the main characteristics of temperature change, but also reduces the redundancy of data and improves the efficiency and accuracy of subsequent data processing. In addition, the determination of the temperature feature vector has produced positive technical effects in many aspects. On the one hand, it provides a comprehensive and concise tool to describe temperature changes, allowing us to more accurately understand the impact of temperature changes on the vibrating wire frequency signal. On the other hand, by combining the temperature feature vector with the vibrating wire frequency signal for multi-resolution spatial decomposition and feature extraction, the safety status of the reservoir dam area to be monitored can be more accurately assessed.

[0009] Optionally, performing multi-resolution spatial decomposition on the vibrating-wire frequency signal according to the temperature eigenvector includes: Determining a target decomposition depth according to the temperature feature vector, wherein the target decomposition depth is used to characterize a maximum number of decomposition layers in a preset multi-resolution decomposition framework, wherein different wavelet basis functions are preset in the preset multi-resolution decomposition framework; The vibrating string frequency signal is subjected to wavelet decomposition according to the target decomposition depth and the preset multi-resolution decomposition framework.

[0010] In the above scheme, the target decomposition depth, as a key parameter in the preset multi-resolution decomposition framework, directly determines the maximum number of layers of wavelet decomposition, which is determined based on the temperature eigenvector. Among them, the temperature eigenvector, as a comprehensive reflection of the temperature change characteristics, contains key information such as the temperature mean and standard deviation, which is crucial for understanding the impact of temperature change on the vibrating string frequency signal. By comprehensively considering the various dimensions of the temperature eigenvector, the target decomposition depth can be determined more accurately, thereby ensuring that the wavelet decomposition can capture the main features of the signal and avoid computational redundancy caused by over-decomposition. Next, the vibrating string frequency signal is wavelet decomposed according to the target decomposition depth and the preset multi-resolution decomposition framework. The preset multi-resolution decomposition framework contains different wavelet basis functions, which have their own unique time-frequency characteristics and can adapt to different types of signal analysis requirements. After determining the target decomposition depth, the corresponding wavelet basis function can be selected according to this depth to decompose the vibrating string frequency signal layer by layer. This process not only realizes the detailed division of the signal at different frequency scales, but also retains the time-frequency characteristics of the signal at different decomposition layers, providing a rich information basis for subsequent feature extraction and state evaluation. By adjusting the target decomposition depth, the appropriate number of decomposition layers can be selected as needed to balance the computational complexity and information extraction capabilities. In addition, wavelet decomposition can capture subtle changes in signals at different frequency scales, which is crucial for identifying abnormal fluctuations in vibrating string frequency signals.

[0011] Optionally, determining the target decomposition depth according to the temperature feature vector includes: The target decomposition depth is determined according to the temperature feature vector and a maximum decomposition layer number threshold of a preset multi-resolution decomposition framework.

[0012] In the above scheme, by comprehensively considering the temperature eigenvector and the maximum decomposition layer threshold, the target decomposition depth can be determined, which not only helps to capture the main features in the vibrating string frequency signal, but also avoids computational redundancy and information loss caused by over-decomposition. The introduction of the temperature eigenvector makes it possible to adapt to the signal analysis needs under different temperature change conditions. Whether the temperature changes drastically or gently, the accuracy and efficiency of the analysis can be ensured by adjusting the target decomposition depth. By combining the maximum decomposition layer threshold, unnecessary deep decomposition can be avoided, thereby saving computing resources. This optimization not only improves the efficiency of data processing, but also reduces the dependence on hardware devices.

[0013] Optionally, the preset multi-resolution decomposition framework includes multiple decomposition layers, and a basis function set is configured in the preset multi-resolution decomposition framework. The basis function set includes multiple basis functions, and the basis functions in the basis function set are used to configure wavelet decomposition in the decomposition layer.

[0014] In the above scheme, through multi-resolution decomposition and basis function configuration, the detailed features in the vibrating string frequency signal can be captured more accurately, wherein the introduction of the basis function set can adapt to different types of vibrating string frequency signals and analysis requirements. Whether the signal is stationary or non-stationary, linear or non-linear, the accuracy of the analysis can be ensured by selecting the appropriate basis function.

[0015] Optionally, performing wavelet decomposition on the vibrating string frequency signal according to the target decomposition depth and the preset multi-resolution decomposition framework includes: A target basis function is selected from the basis function set according to the temperature characteristic vector, so as to perform wavelet decomposition on the vibrating wire frequency signal by using the target basis function.

[0016] In the above scheme, by selecting the target basis function according to the temperature eigenvector, it can ensure that the wavelet decomposition is more targeted at the characteristics of the current signal, thereby improving the accuracy and effectiveness of the decomposition. Since the temperature eigenvector changes with time and environment, the selection of the basis function will also be adjusted accordingly. This dynamic adjustment capability enables the method to adapt to the needs of signal analysis under different environments and conditions. Then, using the basis function that matches the characteristics of the target signal for wavelet decomposition can more efficiently extract the key information in the signal.

[0017] Optionally, performing wavelet decomposition on the vibrating string frequency signal according to the target decomposition depth and the preset multi-resolution decomposition framework includes: According to the temperature feature vector, corresponding target basis functions are sequentially configured for each target decomposition layer of the target decomposition depth to perform wavelet decomposition on the vibrating string frequency signal, wherein the target basis function is one of the basis functions in the basis function set, and each target decomposition layer is configured with a different basis function.

[0018] In the above scheme, by configuring basis functions for each target decomposition layer according to the temperature eigenvector, it can be ensured that each decomposition layer uses the basis function that best suits the current signal characteristics for decomposition. This customized decomposition strategy significantly improves the accuracy and effectiveness of the decomposition, making it possible to more accurately capture the key information in the signal.

[0019] In a second aspect, the present application provides a vibrating wire piezometer data processing system for a reservoir dam, comprising: An acquisition module, used to respectively acquire a vibrating string frequency signal and a temperature signal of a reservoir dam to be monitored area, wherein the vibrating string frequency signal is data acquired by a vibrating string osmometer; A processing module, configured to determine a temperature characteristic vector according to the temperature signal, to perform a multi-resolution spatial decomposition on the vibrating string frequency signal according to the temperature characteristic vector, to generate a corresponding vibrating string dynamic characteristic value under each resolution space, to generate a vibrating string dynamic vector, wherein the temperature characteristic vector is associated with a decomposition depth of the multi-resolution spatial decomposition; The processing module is further used to determine the current state of the area to be monitored according to the vibrating string dynamic vector.

[0020] Optionally, the processing module is specifically used for: Performing time series sampling on the temperature signal to determine a temperature sequence, and determining a temperature mean and a temperature standard deviation according to the temperature sequence; The temperature feature vector is determined according to the temperature mean, the temperature standard deviation, and the temperature sequence.

[0021] Optionally, the processing module is specifically used for: Determining a target decomposition depth according to the temperature feature vector, wherein the target decomposition depth is used to characterize a maximum number of decomposition layers in a preset multi-resolution decomposition framework, wherein different wavelet basis functions are preset in the preset multi-resolution decomposition framework; The vibrating string frequency signal is subjected to wavelet decomposition according to the target decomposition depth and the preset multi-resolution decomposition framework.

[0022] Optionally, the processing module is specifically used for: The target decomposition depth is determined according to the temperature feature vector and a maximum decomposition layer number threshold of a preset multi-resolution decomposition framework.

[0023] Optionally, the preset multi-resolution decomposition framework includes multiple decomposition layers, and a basis function set is configured in the preset multi-resolution decomposition framework. The basis function set includes multiple basis functions, and the basis functions in the basis function set are used to configure wavelet decomposition in the decomposition layer.

[0024] Optionally, the processing module is specifically used for: A target basis function is selected from the basis function set according to the temperature characteristic vector, so as to perform wavelet decomposition on the vibrating wire frequency signal by using the target basis function.

[0025] Optionally, the processing module is specifically used for: According to the temperature feature vector, corresponding target basis functions are sequentially configured for each target decomposition layer of the target decomposition depth to perform wavelet decomposition on the vibrating string frequency signal, wherein the target basis function is one of the basis functions in the basis function set, and each target decomposition layer is configured with a different basis function.

[0026] In a third aspect, the present application provides an electronic device, including: processor; and, A memory, configured to store executable instructions of the processor; The processor is configured to perform any possible method described in the first aspect by executing the executable instructions.

[0027] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement any possible method described in the first aspect.

[0028] The present application provides a vibrating-string piezometer data processing method and system for a reservoir dam, which respectively obtains the vibrating-string frequency signal and the temperature signal of the area to be monitored of the reservoir dam, determines the temperature characteristic vector according to the temperature signal, performs multi-resolution spatial decomposition on the vibrating-string frequency signal according to the temperature characteristic vector, generates the corresponding vibrating-string dynamic characteristic values ​​in each resolution space, generates the vibrating-string dynamic vector, and then determines the current state of the area to be monitored according to the vibrating-string dynamic vector, thereby comprehensively considering the combination of the vibrating-string frequency signal and the temperature signal, and utilizing multi-resolution spatial decomposition and feature extraction to achieve a more accurate assessment of the safety state of the area to be monitored of the reservoir dam. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0030] Figure 1 It is a flow chart of a vibrating wire osmometer data processing method for a reservoir dam according to an exemplary embodiment of the present application; Figure 2 is a flow chart of a vibrating-wire piezometer data processing method for a reservoir dam according to another exemplary embodiment of the present application; Figure 3 is a structural schematic diagram of a vibrating-wire osmometer data processing system for a reservoir dam according to an exemplary embodiment of the present application; Figure 4 It is a schematic diagram of the structure of an electronic device according to an exemplary embodiment of the present application.

[0031] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0032] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0033] Figure 1 FIG. 1 is a flow chart of a vibrating wire osmometer data processing method for a reservoir dam according to an exemplary embodiment of the present application. Figure 1 As shown, the method provided in this embodiment includes: S101, respectively obtaining a vibrating string frequency signal and a temperature signal of a reservoir dam area to be monitored.

[0034] Specifically, a vibrating-wire osmometer is installed in the monitored area of ​​the reservoir dam. The osmometer can sense the change of osmotic pressure inside the dam and convert it into fluctuation of the vibrating-wire frequency. The vibrating-wire frequency signal output by the vibrating-wire osmometer is collected regularly or in real time by a data acquisition system to ensure the continuity and accuracy of the signal. Optionally, the collected vibrating-wire frequency signal can be pre-processed, such as denoising, filtering, etc., to improve the quality of the signal.

[0035] In addition, a temperature sensor is installed in or near the area to be monitored of the reservoir dam to monitor the temperature changes inside the dam or the surrounding environment. The output signal of the temperature sensor is collected synchronously with the vibrating string frequency signal to ensure the temporal consistency of the temperature signal and the vibrating string frequency signal. Similarly, the collected temperature signal is preprocessed, such as calibration, denoising, etc., to improve the accuracy of temperature measurement.

[0036] S102: Determine a temperature characteristic vector according to the temperature signal.

[0037] Specifically, the preprocessed temperature signal can be subjected to feature extraction, such as calculating the average value, fluctuation range, and change trend of the temperature to form a temperature feature vector. The temperature feature vector should be able to fully reflect the overall trend and fluctuation characteristics of temperature changes, and provide a basis for subsequent multi-resolution spatial decomposition.

[0038] In a possible implementation, the temperature signal can be sampled in time series to determine the temperature series. ,in, is the sampling frequency within the preset sampling period; Using formula 1, and according to the temperature series Determine the mean temperature , where Formula 1 is: ; Using Formula 2, and based on the temperature series Determining the temperature standard deviation , where Formula 2 is: ; Using Formula 3 and based on the mean temperature And the temperature standard deviation Determine the temperature eigenvector , where Formula 3 is: ; in, is the temperature feature vector The first Temperature characteristic element.

[0039] It is worth noting that the element values ​​of the temperature feature vector obtained by processing Formula 3 contain both the degree of deviation of the temperature value and the relative importance of this deviation in the temperature distribution. Such feature vectors are easier to be used by subsequent multi-resolution spatial decomposition algorithms, thereby more accurately capturing the correlation between the vibrating string frequency signal and the temperature signal. The introduction of the Gaussian function dynamically adjusts the weight of the feature element according to the degree of deviation of the temperature value from the mean, so that the temperature value close to the mean occupies a more important position in the feature vector, thereby more accurately reflecting the main characteristics of the temperature distribution.

[0040] S103, performing multi-resolution spatial decomposition on the vibrating string frequency signal according to the temperature eigenvector to generate corresponding vibrating string dynamic eigenvalues ​​in each resolution space to generate a vibrating string dynamic vector.

[0041] Specifically, appropriate wavelet basis functions and decomposition layers are selected, and the vibrating string frequency signal is decomposed in multi-resolution space according to the temperature eigenvector. The depth of decomposition is closely related to the temperature eigenvector. For example, when the temperature changes greatly, a deeper decomposition layer can be selected to capture more detailed information; when the temperature changes slightly, a shallower decomposition layer can be selected to improve efficiency. Through multi-resolution spatial decomposition, the vibrating string frequency signal is decomposed into sub-signals in different resolution spaces, each of which corresponds to a specific frequency range and level of detail.

[0042] Then, the sub-signals in each resolution space are subjected to feature extraction, such as calculating the energy, entropy, peak value, etc. of the sub-signals, to generate the corresponding vibrating string dynamic eigenvalues ​​in each resolution space. The vibrating string dynamic eigenvalues ​​can accurately reflect the dynamic change characteristics of the vibrating string frequency signal in different resolution spaces. The vibrating string dynamic eigenvalues ​​in each resolution space are then spliced ​​in a certain order to form a vibrating string dynamic vector.

[0043] In a possible implementation, a target decomposition depth can be determined according to a temperature feature vector, wherein the target decomposition depth is used to characterize a maximum number of decomposition layers in a preset multi-resolution decomposition framework, wherein different wavelet basis functions are preset in the preset multi-resolution decomposition framework. According to the target decomposition depth and the preset multi-resolution decomposition framework, wavelet decomposition is performed on the vibrating string frequency signal.

[0044] It is worth noting that a threshold or rule related to temperature fluctuation can be set based on the statistical analysis results of the temperature feature vector. For example, when the temperature fluctuation exceeds a certain range, it is considered that a deeper decomposition of the vibrating string frequency signal is required to capture the details of the frequency change caused by the temperature change. According to this threshold or rule, the target decomposition depth is determined. The target decomposition depth is an integer representing the maximum number of layers required for wavelet decomposition in the preset multi-resolution decomposition framework.

[0045] S104: Determine the current state of the area to be monitored according to the vibrating string dynamic vector.

[0046] Specifically, a state assessment model can be established, which can output the current state of the area to be monitored based on the input of the vibrating string dynamic vector. The state assessment model can use advanced artificial intelligence technologies such as machine learning algorithms and neural network algorithms to improve the accuracy and efficiency of the assessment. The generated vibrating string dynamic vector is input into the state assessment model for state assessment. According to the output of the state assessment model, the current state of the area to be monitored is determined, such as normal, abnormal, warning, etc. The assessment results can also be visualized or alarmed to timely discover and deal with potential safety hazards.

[0047] In this embodiment, the vibrating string frequency signal and the temperature signal of the monitored area of ​​the reservoir dam are respectively obtained, and the temperature characteristic vector is determined according to the temperature signal, so as to perform multi-resolution spatial decomposition on the vibrating string frequency signal according to the temperature characteristic vector, so as to generate the corresponding vibrating string dynamic characteristic value in each resolution space, so as to generate the vibrating string dynamic vector, and then determine the current state of the monitored area according to the vibrating string dynamic vector, so as to comprehensively consider the combination of the vibrating string frequency signal and the temperature signal, and utilize multi-resolution spatial decomposition and feature extraction to achieve a more accurate assessment of the safety status of the monitored area of ​​the reservoir dam.

[0048] Figure 2 FIG. 1 is a flow chart of a vibrating wire osmometer data processing method for a reservoir dam according to another exemplary embodiment of the present application. Figure 2 As shown, the vibrating wire osmometer data processing method for a reservoir dam provided in this embodiment includes: S201, respectively obtaining a vibrating string frequency signal and a temperature signal of a reservoir dam area to be monitored.

[0049] Specifically, a vibrating-wire osmometer is installed in the monitored area of ​​the reservoir dam. The osmometer can sense the change of osmotic pressure inside the dam and convert it into fluctuation of the vibrating-wire frequency. The vibrating-wire frequency signal output by the vibrating-wire osmometer is collected regularly or in real time by a data acquisition system to ensure the continuity and accuracy of the signal. Optionally, the collected vibrating-wire frequency signal can be pre-processed, such as denoising, filtering, etc., to improve the quality of the signal.

[0050] In addition, a temperature sensor is installed in or near the area to be monitored of the reservoir dam to monitor the temperature changes inside the dam or the surrounding environment. The output signal of the temperature sensor is collected synchronously with the vibrating string frequency signal to ensure the temporal consistency of the temperature signal and the vibrating string frequency signal. Similarly, the collected temperature signal is preprocessed, such as calibration, denoising, etc., to improve the accuracy of temperature measurement.

[0051] S202: Determine a temperature characteristic vector according to the temperature signal.

[0052] Specifically, the preprocessed temperature signal can be subjected to feature extraction, such as calculating the average value, fluctuation range, and change trend of the temperature to form a temperature feature vector. The temperature feature vector should be able to fully reflect the overall trend and fluctuation characteristics of temperature changes, and provide a basis for subsequent multi-resolution spatial decomposition.

[0053] S203: Determine the target decomposition depth according to the temperature feature vector.

[0054] Alternatively, one can use Equation 4 and calculate the temperature characteristic vector And the maximum decomposition layer threshold of the preset multi-resolution decomposition framework Determine target resolution depth , where Formula 4 is: ; in, The maximum decomposition layer threshold for the preset multi-resolution decomposition framework.

[0055] In the above formula 4, first, all elements in the temperature feature vector are accumulated, and this accumulated value reflects the overall strength or trend of the temperature feature. When the element value in the temperature feature vector is large (that is, the temperature fluctuation is more obvious), the accumulated value will also increase accordingly. Then, the accumulated value is input into the sigmoid function. The characteristic of the sigmoid function is that when the input value is small, the output value is close to 0; when the input value is large, the output value is close to 1. Therefore, through the mapping of the sigmoid function, the accumulated value is converted to a value between 0 and 1, which represents the ratio of the temperature feature strength or trend relative to the maximum value. Next, the output value of the sigmoid function is multiplied by the maximum decomposition layer threshold of the preset multi-resolution decomposition framework. The purpose of this step is to determine the target decomposition depth according to the ratio of the temperature feature strength or trend. When the temperature feature is strong, the output value of the sigmoid function is close to 1, and the target decomposition depth will also increase accordingly; when the temperature feature is weak, the output value of the sigmoid function is close to 0, and the target decomposition depth will also decrease accordingly. Finally, the product result is rounded down to ensure that the target decomposition depth is an integer.

[0056] S204, performing wavelet decomposition on the vibrating string frequency signal according to the target decomposition depth and a preset multi-resolution decomposition framework.

[0057] Optionally, the above-mentioned preset multi-resolution decomposition framework includes at least a first decomposition layer, a second decomposition layer and a third decomposition layer, and the preset multi-resolution decomposition framework is configured with a basis function set, the basis function set includes multiple basis functions, and the basis function set includes a first basis function, a second basis function and a third basis function.

[0058] Among them, a set of basis functions in the multi-resolution decomposition framework is preset. The set includes multiple basis functions with different frequency resolutions, time domain characteristics and shape characteristics, such as Haar wavelet, Daubechies wavelet, Symlet wavelet, Coiflet wavelet, etc. Each basis function has its unique application scenario and advantages. Choosing a suitable basis function is crucial to accurately capture the characteristics of the vibrating string frequency signal.

[0059] According to the analysis results of the temperature feature vector, combined with the characteristics of each basis function in the basis function set, the target basis function that best suits the current ambient temperature conditions is selected. For example, if the temperature feature vector shows that the ambient temperature changes dramatically and contains more high-frequency components, a basis function with high frequency resolution, such as Haar wavelet or Daubechies wavelet, can be selected to better capture these high-frequency features.

[0060] If the ambient temperature changes are relatively stable and the low-frequency trends are of primary interest, a basis function with low frequency resolution, such as the Morlet wavelet or the Mexican Hat wavelet, can be selected.

[0061] In another possible implementation, corresponding target basis functions are sequentially configured for each target decomposition layer of the target decomposition depth according to the temperature eigenvector to perform wavelet decomposition on the vibrating string frequency signal, wherein the target basis function is one of the basis functions in the basis function set, and each target decomposition layer is configured with a different basis function.

[0062] First decomposition layer: As the most basic decomposition layer, it is mainly responsible for capturing high-frequency components in the signal, such as noise and short-term fluctuations. This layer uses basis functions with higher frequency resolution to accurately identify and separate these components.

[0063] Second decomposition layer: Based on the first decomposition layer, the intermediate frequency components in the signal are further decomposed. These components may include changes in osmotic pressure caused by changes in reservoir water level, temperature, etc. The basis function used in the second decomposition layer has a moderate frequency resolution, which can capture the intermediate frequency changes without over-decomposing the high-frequency or low-frequency components.

[0064] The third decomposition layer: As the topmost decomposition layer, it is mainly responsible for capturing low-frequency components in the signal, such as long-term trends and seasonal changes. This layer uses basis functions with lower frequency resolution in order to accurately reflect these long-term changes.

[0065] Each decomposition layer forms a hierarchical relationship, and each layer further decomposes the signal on the basis of the previous layer, thereby realizing multi-scale and multi-level analysis of the signal. Through this hierarchical progressive approach, different components in the signal can be separated more effectively, providing a more accurate basis for subsequent data analysis and processing.

[0066] The preset multi-resolution decomposition framework is configured with a basis function set, which includes multiple basis functions for decomposing the signal at different decomposition layers. The basis function set includes at least a first basis function, a second basis function and a third basis function, which correspond to the first decomposition layer, the second decomposition layer and the third decomposition layer, respectively.

[0067] First basis function: Select a wavelet function with high frequency resolution, such as Haar wavelet or Daubechies wavelet, etc. These wavelet functions have compact support and orthogonality, and are suitable for capturing high-frequency components in the signal.

[0068] Second basis function: Select a wavelet function with medium frequency resolution, such as Symlet wavelet or Coiflet wavelet, etc. These wavelet functions have good localization performance in both frequency domain and time domain, and are suitable for capturing the medium frequency components in the signal.

[0069] Third basis function: Select a wavelet function with low frequency resolution, such as Morlet wavelet or Mexican Hat wavelet, etc. These wavelet functions have a wide bandwidth in the frequency domain and are suitable for capturing low-frequency components in the signal.

[0070] In a possible implementation, a target basis function may be selected from a basis function set according to a temperature eigenvector, so as to perform wavelet decomposition on the vibrating wire frequency signal using the target basis function.

[0071] S205 , performing wavelet decomposition on the vibrating string frequency signal according to each decomposition layer in the multi-resolution space to obtain a target characteristic wave set corresponding to each decomposition layer.

[0072] Specifically, the vibrating string frequency signal is subjected to wavelet decomposition according to each decomposition layer in the multi-resolution space to obtain a target characteristic wave set corresponding to each decomposition layer, wherein the target characteristic wave set includes an approximate wave set and a detail wave set.

[0073] It is worth noting that the vibrating string frequency signal can be subjected to wavelet decomposition using the basis functions in each decomposition layer to determine the characteristic wavelet, wherein the approximate wave set in the target characteristic wave set is the low-frequency component in the characteristic wavelet whose frequency is lower than a preset frequency threshold, and the detail wave set in the target characteristic wave set is the high-frequency component in the characteristic wavelet whose frequency is higher than a preset frequency threshold.

[0074] In the above scheme, multi-resolution analysis allows the signal to be decomposed and reconstructed at different scales (or resolutions). Through wavelet decomposition in multi-resolution space, the vibrating string frequency signal can be decomposed into multiple components of different frequency bands, which represent the characteristics of the signal at different scales.

[0075] In the process of wavelet decomposition, the basis functions in each decomposition layer play a key role. Basis functions are the basic units in wavelet analysis. They can be sine functions, cosine functions, Gaussian functions, etc., but more common are wavelet functions with localized characteristics. These basis functions can form a set of basis function systems that can cover the entire frequency range through scaling and translation operations. By using these basis functions to decompose the vibrating string frequency signal, the characteristic wavelets of the signal in different frequency bands can be extracted.

[0076] Through wavelet decomposition, we can get a series of characteristic wavelets, which represent the components of the vibrating string frequency signal in different frequency bands. These characteristic wavelets contain not only the frequency information of the signal, but also the time localization information of the signal, which enables us to more accurately analyze the changing characteristics of the signal at different times and frequencies.

[0077] After obtaining the characteristic wavelet, it needs to be divided into a target characteristic wave set. According to a preset frequency threshold, the components in the characteristic wavelet with a frequency lower than the threshold are classified as an approximate wave set, and the components with a frequency higher than the threshold are classified as a detail wave set.

[0078] The approximate wave set represents the low-frequency components in the vibrating string frequency signal. In water conservancy project monitoring, the low-frequency components are usually related to the overall structural response and long-term change trend of the reservoir dam. By analyzing the approximate wave set, important information such as the overall stability and long-term deformation trend of the dam structure can be obtained, providing strong support for engineering safety assessment.

[0079] The detail wave set represents the high-frequency components in the vibrating string frequency signal. These high-frequency components are usually related to the short-term dynamic response of the reservoir dam, such as local deformation and crack expansion. By analyzing the detail wave set, potential problems in the dam structure can be discovered in time, providing an important reference for engineering maintenance and repair.

[0080] Through the above steps, the wavelet decomposition technology is used to decompose the vibrating string frequency signal into an approximate wave set and a detail wave set, which not only improves the accuracy of data processing, but also significantly improves the processing efficiency, and can effectively extract the key feature information in the signal.

[0081] In addition, the process of wavelet decomposition can also involve selecting appropriate basis functions (such as Haar wavelet, Daubechies wavelet, etc.) and setting the number of decomposition layers according to the signal characteristics. Through discrete wavelet transform (DWT) or continuous wavelet transform (CWT), the vibrating string frequency signal is decomposed into signal components in different frequency bands to form approximate wave sets and detail wave sets corresponding to each decomposition layer. In each decomposition layer, the approximate wave set and the detail wave set are extracted respectively. The approximate wave set contains the low-frequency components of the signal and reflects the overall trend of the signal; while the detail wave set contains the high-frequency components of the signal and reflects the local details and fluctuations of the signal.

[0082] S206. Determine the corresponding approximate wave characteristic value according to the approximate wave set corresponding to each decomposition layer.

[0083] Specifically, the corresponding approximate fluctuation characteristic value is determined according to the approximate wave set corresponding to each decomposition layer, and the approximate fluctuation characteristic value is the standard deviation of each wave peak in the approximate wave set.

[0084] The approximate wave set in each decomposition layer can be subjected to wave crest detection. The wave crest detection algorithm can use a local maximum value method, a zero-crossing method, etc. to accurately identify the position and amplitude of the wave crest.

[0085] Calculate the standard deviation of each peak in the approximate wave set. The standard deviation is an important indicator to measure the degree of data dispersion. Here, it is used to quantify the degree of fluctuation of the peaks in the approximate wave set. By calculating the standard deviation, the approximate fluctuation eigenvalue corresponding to each decomposition layer can be obtained. This eigenvalue reflects the overall fluctuation characteristics of the signal at this decomposition layer.

[0086] S207. Determine the corresponding vibrating string energy eigenvalue according to the detail wave set corresponding to each decomposition layer.

[0087] Specifically, the corresponding vibrating string energy eigenvalue is determined according to the detail wave set corresponding to each decomposition layer, and the vibrating string energy eigenvalue is the integral of each detail wave in the detail wave set along the time direction.

[0088] The detail wave set in each decomposition layer is integrated along the time direction. The integration operation can capture the energy distribution and change trend of each detail wave in the detail wave set. Through the integration operation, the vibrating string energy eigenvalue corresponding to each decomposition layer is obtained. This eigenvalue reflects the local energy distribution and detail characteristics of the signal at the decomposition layer.

[0089] S208, generating a vibrating string dynamic eigenvalue according to the approximate wave eigenvalues ​​of each decomposition layer and the vibrating string energy eigenvalues.

[0090] In this step, the vibrating string dynamic eigenvalues ​​are generated according to the approximate wave eigenvalues ​​of each decomposition layer and the vibrating string energy eigenvalues, so as to splice the vibrating string dynamic eigenvalues ​​of each decomposition layer to form a vibrating string dynamic vector.

[0091] Specifically, the approximate wave eigenvalues ​​and vibrating string energy eigenvalues ​​corresponding to each decomposition layer are concatenated to form the vibrating string dynamic eigenvalues ​​of the decomposition layer. The vibrating string dynamic eigenvalues ​​of each decomposition layer are concatenated in the order of the decomposition levels to form the final vibrating string dynamic vector. The vibrating string dynamic vector is a high-dimensional eigenvector that contains the dynamic characteristic information of the vibrating string frequency signal in different resolution spaces.

[0092] S209: Determine the current state of the area to be monitored according to the vibrating string dynamic vector.

[0093] Specifically, a state assessment model can be established, which can output the current state of the area to be monitored based on the input of the vibrating string dynamic vector. The state assessment model can use advanced artificial intelligence technologies such as machine learning algorithms and neural network algorithms to improve the accuracy and efficiency of the assessment. The generated vibrating string dynamic vector is input into the state assessment model for state assessment. According to the output of the state assessment model, the current state of the area to be monitored is determined, such as normal, abnormal, warning, etc. The assessment results can also be visualized or alarmed to timely discover and deal with potential safety hazards.

[0094] In a possible implementation, the vibrating string dynamic vector may be normalized to generate a normalized vibrating string dynamic vector. The normalized vibrating string dynamic vector input value support vector machine is classified to determine the current state of the monitored area, wherein the current state includes a safe state, a warning state, and a dangerous state.

[0095] Specifically, normalizing the vibrating-wire dynamic vector is a key step to ensure the accuracy of subsequent classification. The vibrating-wire dynamic vector contains a variety of data features collected by the vibrating-wire osmometer, such as frequency, amplitude, etc., and the dimensions and numerical ranges of these features may be different. Normalization can convert these feature values ​​to the same numerical range, usually [0, 1] or [-1, 1], thereby eliminating the impact of dimension and numerical range differences on the classification results. The normalized vibrating-wire dynamic vector, that is, the normalized vibrating-wire dynamic vector, provides a more balanced and comparable data basis for subsequent classification tasks.

[0096] When the normalized vibrating wire dynamic vector is input into the support vector machine for classification, a support vector machine model needs to be trained first. During the training process, the vibrating wire dynamic vector of a known state (such as safe state, warning state, dangerous state) needs to be used as a training sample, and the classification performance is optimized by adjusting the model parameters. After the training is completed, the new normalized vibrating wire dynamic vector can be input into the model for classification, thereby determining the current state of the area to be monitored.

[0097] Through the classification of support vector machines, the status of the monitored area can be accurately divided into safe status, warning status and dangerous status. This classification result not only provides an intuitive basis for the safety monitoring of reservoir dams, but also provides strong support for subsequent engineering decision-making and maintenance. In the safe state, the existing monitoring frequency and engineering measures can be maintained; in the warning state, it is necessary to strengthen monitoring and inspection, and prepare to take necessary engineering measures; in the dangerous state, emergency measures need to be taken immediately to ensure the safety of the reservoir dam.

[0098] It is worth noting that the embodiments provided in the embodiments of the present application are intended to process the vibrating-wire osmometer data used for reservoir dams, so as to improve the adaptability of the vibrating-wire dynamic vector input into the support vector machine to the actual temperature, wherein the adaptability is specifically manifested in the ability to adapt to the signal analysis requirements under different temperature change conditions. Whether the temperature changes drastically or gently, the accuracy and efficiency of the analysis can be ensured by adjusting the target decomposition depth. By combining the maximum decomposition layer threshold, unnecessary deep decomposition can be avoided, thereby saving computing resources. This optimization not only improves the efficiency of data processing, but also reduces the dependence on hardware equipment. As for the training process of the support vector machine, it is not specifically limited in this embodiment.

[0099] In one possible implementation, historical data from vibrating wire piezometers are collected, which reflect the vibrating wire response of the dam under different conditions. For each vibrating wire dynamic vector data, there needs to be a clear label indicating the state of the dam at that time (safe, warning, dangerous). These labels are usually based on expert judgment or records of historical events.

[0100] Then, the dynamic vector of the vibrating string is normalized to eliminate the dimensional differences between data of different dimensions and improve the convergence speed and accuracy of the model. Then, outliers and missing values ​​are removed or corrected to ensure data quality.

[0101] Extract features useful for classification tasks from the normalized dynamic vector of the vibrating string. This may include changes in frequency, amplitude, phase, and other parameters or combinations thereof. If the feature dimension is too high, it may lead to overfitting of the model or inefficiency of the computation. Consider using methods such as principal component analysis to reduce the feature dimension.

[0102] Then, a kernel function is selected, such as a linear kernel, a polynomial kernel, a radial basis function (RBF) kernel, etc. A suitable kernel function is selected according to the characteristics of the data and the requirements of the classification task. For example, in this implementation, an RBF kernel may be selected.

[0103] The preprocessed data is trained using the selected kernel function and tuned parameters to obtain the support vector machine classification model.

[0104] The trained support vector machine classification model is then deployed to the actual reservoir dam monitoring system for real-time or periodic data processing and status classification. The system performance can also be monitored later, and the data can be updated regularly and the model can be retrained to adapt to changes in the dam status.

[0105] Figure 3 FIG. 1 is a schematic diagram of a vibrating wire osmometer data processing system for a reservoir dam according to an exemplary embodiment of the present application. Figure 3 As shown, the vibrating wire osmometer data processing system 300 provided in this embodiment includes: An acquisition module 310 is used to respectively acquire a vibrating-wire frequency signal and a temperature signal of a reservoir dam area to be monitored, wherein the vibrating-wire frequency signal is data acquired by a vibrating-wire osmometer; A processing module 320 is used to determine a temperature characteristic vector according to the temperature signal, to perform multi-resolution spatial decomposition on the vibrating string frequency signal according to the temperature characteristic vector, to generate corresponding vibrating string dynamic characteristic values ​​under each resolution space, to generate a vibrating string dynamic vector, wherein the temperature characteristic vector is associated with a decomposition depth of the multi-resolution spatial decomposition; The processing module 320 is further configured to determine a current state of the area to be monitored according to the vibrating wire dynamic vector.

[0106] Optionally, the processing module 320 is specifically used for: Performing time series sampling on the temperature signal to determine a temperature sequence, and determining a temperature mean and a temperature standard deviation according to the temperature sequence; The temperature feature vector is determined according to the temperature mean, the temperature standard deviation, and the temperature sequence.

[0107] Optionally, the processing module 320 is specifically used for: Determining a target decomposition depth according to the temperature feature vector, wherein the target decomposition depth is used to characterize a maximum number of decomposition layers in a preset multi-resolution decomposition framework, wherein different wavelet basis functions are preset in the preset multi-resolution decomposition framework; The vibrating string frequency signal is subjected to wavelet decomposition according to the target decomposition depth and the preset multi-resolution decomposition framework.

[0108] Optionally, the processing module 320 is specifically used for: The target decomposition depth is determined according to the temperature feature vector and a maximum decomposition layer number threshold of a preset multi-resolution decomposition framework.

[0109] Optionally, the preset multi-resolution decomposition framework includes multiple decomposition layers, and a basis function set is configured in the preset multi-resolution decomposition framework. The basis function set includes multiple basis functions, and the basis functions in the basis function set are used to configure wavelet decomposition in the decomposition layer.

[0110] Optionally, the processing module 320 is specifically used for: A target basis function is selected from the basis function set according to the temperature characteristic vector, so as to perform wavelet decomposition on the vibrating wire frequency signal by using the target basis function.

[0111] Optionally, the processing module 320 is specifically used for: According to the temperature feature vector, corresponding target basis functions are sequentially configured for each target decomposition layer of the target decomposition depth to perform wavelet decomposition on the vibrating string frequency signal, wherein the target basis function is one of the basis functions in the basis function set, and each target decomposition layer is configured with a different basis function.

[0112] Figure 4 is a schematic diagram of the structure of an electronic device according to an exemplary embodiment of the present application. Figure 4 As shown, an electronic device 400 provided in this embodiment includes: a processor 401 and a memory 402; wherein: The memory 402 is used to store computer programs, and the memory may also be a flash memory.

[0113] The processor 401 is used to execute the execution instructions stored in the memory to implement each step in the above method. For details, please refer to the relevant description in the above method embodiment.

[0114] Optionally, the memory 402 may be independent or integrated with the processor 401 .

[0115] When the memory 402 is a device independent of the processor 401, the electronic device 400 may further include: The bus 403 is used to connect the memory 402 and the processor 401 .

[0116] This embodiment further provides a readable storage medium, in which a computer program is stored. When at least one processor of an electronic device executes the computer program, the electronic device executes the methods provided in the above-mentioned various implementation modes.

[0117] This embodiment also provides a program product, which includes a computer program stored in a readable storage medium. At least one processor of the electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device implements the methods provided in the above various embodiments.

[0118] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the claims.

[0119] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A vibrating wire piezometer data processing method for a reservoir dam, characterized in that: include: Respectively obtain a vibrating wire frequency signal and a temperature signal of the area to be monitored in the reservoir dam, wherein the vibrating wire frequency signal is data obtained by a vibrating wire osmometer; Determine a temperature characteristic vector according to the temperature signal, perform multi-resolution spatial decomposition on the vibrating string frequency signal according to the temperature characteristic vector, so as to generate corresponding vibrating string dynamic characteristic values ​​under each resolution space, so as to generate a vibrating string dynamic vector, wherein the temperature characteristic vector is associated with a decomposition depth of the multi-resolution spatial decomposition; The current state of the area to be monitored is determined according to the vibrating wire dynamic vector.

2. The vibrating wire osmometer data processing method for a reservoir dam according to claim 1, characterized in that: The step of determining a temperature characteristic vector according to the temperature signal comprises: Performing time series sampling on the temperature signal to determine a temperature sequence, and determining a temperature mean and a temperature standard deviation according to the temperature sequence; The temperature feature vector is determined according to the temperature mean, the temperature standard deviation, and the temperature sequence.

3. The vibrating wire osmometer data processing method for a reservoir dam according to claim 1, characterized in that: The performing multi-resolution spatial decomposition on the vibrating-wire frequency signal according to the temperature characteristic vector comprises: Determining a target decomposition depth according to the temperature feature vector, wherein the target decomposition depth is used to characterize a maximum number of decomposition layers in a preset multi-resolution decomposition framework, wherein different wavelet basis functions are preset in the preset multi-resolution decomposition framework; The vibrating string frequency signal is subjected to wavelet decomposition according to the target decomposition depth and the preset multi-resolution decomposition framework.

4. The vibrating wire osmometer data processing method for a reservoir dam according to claim 3, characterized in that: The step of determining the target decomposition depth according to the temperature characteristic vector comprises: The target decomposition depth is determined according to the temperature feature vector and a maximum decomposition layer number threshold of a preset multi-resolution decomposition framework.

5. The vibrating wire osmometer data processing method for a reservoir dam according to claim 3 or 4, characterized in that: The preset multi-resolution decomposition framework includes multiple decomposition layers. A basis function set is configured in the preset multi-resolution decomposition framework. The basis function set includes multiple basis functions. The basis functions in the basis function set are used to configure wavelet decomposition in the decomposition layer.

6. The vibrating wire osmometer data processing method for a reservoir dam according to claim 5, characterized in that: The step of performing wavelet decomposition on the vibrating string frequency signal according to the target decomposition depth and the preset multi-resolution decomposition framework includes: A target basis function is selected from the basis function set according to the temperature characteristic vector, so as to perform wavelet decomposition on the vibrating wire frequency signal by using the target basis function.

7. The vibrating wire osmometer data processing method for a reservoir dam according to claim 5, characterized in that: The step of performing wavelet decomposition on the vibrating string frequency signal according to the target decomposition depth and the preset multi-resolution decomposition framework includes: According to the temperature feature vector, corresponding target basis functions are sequentially configured for each target decomposition layer of the target decomposition depth to perform wavelet decomposition on the vibrating string frequency signal, wherein the target basis function is one of the basis functions in the basis function set, and each target decomposition layer is configured with a different basis function.

8. A vibrating wire piezometer data processing system for a reservoir dam, characterized in that: include: An acquisition module, used to respectively acquire a vibrating string frequency signal and a temperature signal of a reservoir dam to be monitored area, wherein the vibrating string frequency signal is data acquired by a vibrating string osmometer; A processing module, configured to determine a temperature characteristic vector according to the temperature signal, to perform a multi-resolution spatial decomposition on the vibrating string frequency signal according to the temperature characteristic vector, to generate a corresponding vibrating string dynamic characteristic value under each resolution space, to generate a vibrating string dynamic vector, wherein the temperature characteristic vector is associated with a decomposition depth of the multi-resolution spatial decomposition; The processing module is further used to determine the current state of the area to be monitored according to the vibrating string dynamic vector.

9. An electronic device, characterized in that: include: processor; as well as, A memory, configured to store executable instructions of the processor; The processor is configured to perform the method of any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images