Dam deformation prediction method and device, storage medium and product

Abstract

The invention discloses a dam deformation prediction method and device, a storage medium and a product. The prediction method comprises the following steps: preprocessing dam monitoring data; performing deep feature extraction on the preprocessed environment data to obtain environment features; clustering analysis is conducted on the preprocessed reservoir water level and environment characteristics, time domain division is conducted on the aging factors and dam measuring point displacement according to clustering analysis results, and clustering division results of different time periods are obtained; respectively constructing a sample data set according to each clustering division result; respectively training and testing the Transform-LSTM model by using each sample data set to obtain a dam deformation prediction model in each time period; and obtaining to-be-predicted data, and predicting the predicted data by using the dam deformation prediction model corresponding to the time period to which the to-be-predicted data belongs to obtain a dam deformation prediction result. The dam deformation prediction precision is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of water conservancy project dam deformation monitoring, and in particular relates to a dam deformation prediction method, equipment, storage medium and product. Background Art

[0002] As a crucial component of water conservancy projects, the structural safety of dams is directly linked to the safety and socioeconomic development of downstream areas. Therefore, dam deformation monitoring is particularly important. Deformation monitoring primarily involves real-time or periodic monitoring and analysis of deformation data of the dam and its surroundings, enabling timely identification of potential safety hazards and the implementation of appropriate measures.

[0003] Traditional deformation prediction methods mainly include statistical methods, finite element analysis, and shallow machine learning methods. However, these traditional methods have the following problems:

[0004] (1) The multivariate data processing is complex and it is difficult to analyze the data characteristics of the dam during different operation periods, resulting in poor prediction accuracy;

[0005] (2) Insufficient information utilization makes it difficult to capture the deep nonlinear relationship between long-term series deformation and monitoring factors, resulting in poor prediction accuracy;

[0006] (3) The response speed is slow, the model training time is long, and it is difficult to make quick predictions.

[0007] These problems lead to the limitation of traditional dam deformation monitoring and prediction methods in practical application, making it difficult to meet the demand for high reliability. Therefore, it is urgent and important to research and develop new deformation monitoring and prediction methods. Summary of the Invention

[0008] The purpose of the present invention is to provide a dam deformation prediction method, device, storage medium and product to solve the problem of poor prediction accuracy of traditional prediction methods.

[0009] The present invention solves the above technical problems through the following technical solutions: A method for predicting embankment deformation, comprising:

[0010] Acquire dam monitoring data; wherein the dam monitoring data includes reservoir water level, environmental data, time-effect factors, and displacement of dam measuring points; the time-effect factors are stepwise regression equations of time-effect parameters, and the time-effect parameters are the time rate of change from the monitoring time to the start time of dam operation;

[0011] Preprocessing the dam monitoring data;

[0012] Perform deep feature extraction on the preprocessed environmental data to obtain environmental features;

[0013] Performing cluster analysis on the pre-processed reservoir water level and the environmental characteristics, and dividing the time-effect factors and the displacement of the dam measuring points in the time domain according to the cluster analysis results to obtain cluster division results for different time periods;

[0014] Construct a sample data set based on each clustering result; wherein the number of sample data sets is the same as the number of clustering results;

[0015] Constructing a Transformer-LSTM model; wherein the Transformer-LSTM model is obtained by replacing the decoder of the Transformer model with LSTM;

[0016] The Transformer-LSTM model is trained and tested using each sample data set to obtain a dam deformation prediction model for each time period;

[0017] The data to be predicted is obtained, and the prediction data is predicted using a dam deformation prediction model corresponding to the time period to which the data to be predicted belongs, so as to obtain a dam deformation prediction result.

[0018] Furthermore, the environmental data includes air temperature, water temperature, dam surface temperature and dam foundation temperature.

[0019] Furthermore, the Transformer model is used to perform deep feature extraction on the preprocessed environmental data to obtain environmental features.

[0020] Furthermore, a deep fuzzy clustering algorithm is used to perform cluster analysis on the preprocessed reservoir water level and the environmental characteristics.

[0021] Furthermore, when the deep fuzzy clustering algorithm is used to perform cluster analysis on the pre-processed reservoir water level and the environmental characteristics, the particle swarm optimization algorithm is used to optimize the randomly assigned initial cluster centers to avoid the deep fuzzy clustering algorithm falling into a local optimal solution.

[0022] Furthermore, when the Transformer-LSTM model is trained, a grid search algorithm is used to globally solve the hyperparameters of the Transformer-LSTM model based on the sample data set to find the optimal hyperparameters of the Transformer-LSTM model.

[0023] Based on the same concept, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program / instruction stored in the memory, wherein the processor executes the computer program / instruction to implement the dam deformation prediction method as described above.

[0024] Based on the same concept, the present invention also provides a computer-readable storage medium having a computer program / instruction stored thereon, which implements the above-mentioned dam deformation prediction method when executed by a processor.

[0025] Based on the same concept, the present invention also provides a computer program product, including a computer program / instruction, which implements the above-mentioned dam deformation prediction method when executed by a processor.

[0026] Beneficial effects

[0027] Compared with the prior art, the advantages of the present invention are:

[0028] The present invention conducts deep feature mining on environmental data, studies the nonlinear relationship between air temperature and dam-related temperature in different time periods, and then performs cluster analysis on the mined environmental features and reservoir water levels, thereby improving clustering accuracy and, in turn, the accuracy of constructing sample data sets for each time period. An improved deep learning algorithm (i.e., the Transformer-LSTM model) is used to mine the deep correlation between each clustering result, thereby improving the prediction accuracy of dam deformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only one embodiment of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 is a flow chart of a method for predicting embankment deformation in an embodiment of the present invention;

[0031] Figure 2 is a flow chart of cluster analysis and time domain partitioning in an embodiment of the present invention;

[0032] Figure 3 1 is a schematic diagram of time domain division of embankment measurement point deformation in an embodiment of the present invention;

[0033] Figure 4 This is a structural diagram of the Transformer-LSTM model in an embodiment of the present invention;

[0034] Figure 5 1 is a schematic diagram comparing evaluation results of different models in an embodiment of the present invention; wherein the vertical axis represents the evaluation index value;

[0035] Figure 6 2 is a schematic diagram comparing the prediction effect of the test set and the measured data in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0037] The following specific embodiments are used to describe the technical solution of the present application in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0038] Example 1

[0039] Traditional methods for processing extensive monitoring data typically rely on linear dimensionality reduction, which fails to effectively capture the nonlinear relationships between different monitoring data, resulting in poor prediction accuracy. To address these technical issues, the present invention provides a dam deformation prediction method based on clustering and the Transformer-LSTM model. Figure 1 The flowchart of the dam deformation prediction method provided by the present invention is shown in FIG. Figure 1 As shown, the dam deformation prediction method includes the following steps:

[0040] Step 1: Obtain embankment monitoring data.

[0041] Dam monitoring data includes reservoir water level, environmental data, time-sensitive factors, and dam measurement point displacement; environmental data includes air temperature, water temperature, dam surface temperature, and dam foundation temperature; time-sensitive factors refer to the stepwise regression equation of time-sensitive parameters, and time-sensitive parameters refer to the time rate of change from the monitoring time to the start of dam operation. The specific expression is:

[0042] θ=(t-t0) / 100(1)

[0043] δ θ =c1θ+c2lnθ(2)

[0044] Among them, θ represents the time-effect parameter, t represents the monitoring time, t0 represents the starting time of dam operation, δ θ represents the timeliness factor, and c1 and c2 represent statistical coefficients. When constructing the sample dataset, only the timeliness parameter θ is required. The statistical coefficients c1 and c2 are automatically determined during Transformer-LSTM model training.

[0045] Factors influencing dam deformation typically include hydraulic and temperature factors. However, for concrete dams, the longer the dam is built, the more likely it is to fail. Therefore, aging is also a significant factor affecting dam deformation. By adding aging factors to the monitoring data of the present invention, the accuracy of dam deformation prediction can be improved.

[0046] A set of dam monitoring data can be obtained at each monitoring moment, which includes the reservoir water level, air temperature, dam surface temperature, dam foundation temperature, time factor calculated based on the current monitoring moment, and dam measuring point displacement (i.e., dam deformation) at the current monitoring moment.

[0047] For example, a dam is located in the upper reaches of the eastern source of the Huaihe River, a tributary of the Huaihe River. Its main function is flood control, combined with irrigation, power generation, and shipping. The dam was built in 1954, with a maximum dam height of 75.9m and a dam top axis length of 510m. It is equipped with an automatic monitoring system that can monitor deformation, settlement, stress and strain, and temperature (including air, water, and internal temperature) over a long period of time. The dam has a total of 20 cycloids (PL) and 3 inverted pendulums (IP) arranged on the dam body and foundation for deformation monitoring, and numerous thermometers are also installed inside the dam body for temperature monitoring.

[0048] Step 2: Preprocess the embankment monitoring data.

[0049] In order to improve data quality, the present invention further preprocesses the dam monitoring data. The preprocessing in this embodiment includes cleaning, interpolation, denoising and normalization.

[0050] Step 3: Perform deep feature extraction on the preprocessed environmental data to obtain environmental features.

[0051] The environmental data in this embodiment includes air temperature and dam-related temperatures (i.e., dam surface temperature and dam foundation temperature). The magnitude of change in air temperature and dam-related temperatures varies across time periods (e.g., seasons), indicating a nonlinear relationship between these two variables. To further explore the nonlinear relationship between air temperature and dam-related temperatures and improve the accuracy of constructing sample datasets for each time period, the present invention performs deep feature extraction on the preprocessed environmental data to obtain environmental features.

[0052] In a specific embodiment of the present invention, a Transformer model is used to perform deep feature extraction on the preprocessed environmental data to obtain environmental features.

[0053] The Transformer model is a neural network model based on the self-attention mechanism. It consists of multiple layers of encoders and decoders, each containing multiple attention mechanism modules and feedforward neural network modules. The encoder encodes the air temperature, dam surface temperature, and dam foundation temperature at each monitoring moment into a high-dimensional feature vector representation, while the decoder decodes the high-dimensional feature vector representation output by the encoder into environmental features. The Transformer model employs a parallel mechanism, making its structure more lightweight. It also uses techniques such as residual connections and layer normalization to accelerate model convergence and improve performance. The Transformer model deeply explores the nonlinear relationship between the amplitude of air temperature changes and the amplitude of changes in dam-related temperatures, facilitating more accurate clustering and achieving the division of different time periods.

[0054] Step 4: Perform cluster analysis on the pre-processed reservoir water level and environmental characteristics, and divide the pre-processed time-effect factors and dam measurement point displacements into time domains based on the cluster analysis results to obtain cluster division results for different time periods.

[0055] To accurately segment dam monitoring data across different time periods, cluster analysis is performed on the preprocessed reservoir water level and environmental characteristics. Time-sensitive factors and dam measurement point displacements are then segmented in the time domain based on the cluster analysis results. In a specific embodiment of the present invention, a Deep Fuzzy C-Means (DF) algorithm is used to perform cluster analysis on the preprocessed reservoir water level and the environmental characteristics obtained in step 3.

[0056] Compared with K-means clustering, the deep fuzzy clustering algorithm has better clustering performance and improves the accuracy of data partitioning. The clustering results of the deep fuzzy clustering algorithm are significantly affected by the initial cluster centers. Usually, the initial cluster centers of the deep fuzzy clustering algorithm are randomly assigned according to the data. To avoid the random assignment of initial cluster centers, which may cause the deep fuzzy clustering algorithm to fall into a local optimal solution, the present invention uses the particle swarm optimization algorithm (PSO) to optimize the randomly assigned initial cluster centers.

[0057] In this embodiment, the prediction effect is best when the number of clusters is 4. That is, cluster analysis is performed on the pre-processed reservoir water level and environmental characteristics to obtain 4 cluster analysis results; then, based on the 4 cluster analysis results, the displacement of the dam measurement point and the time effect factor are divided into time domains to obtain 4 cluster division results, such as Figure 3As shown in Figure 2, the displacements of measuring points in similar time domains belong to similar clusters. Each clustering result corresponds to a time period, and each clustering result is the reservoir water level, environmental data, timeliness factors, and displacements of measuring points in the corresponding time period.

[0058] Step 5: Construct a sample data set based on each clustering result.

[0059] To improve the accuracy of dam deformation prediction, each time period corresponds to a dam deformation prediction model. Therefore, a sample dataset is constructed based on each clustering result. The number of sample datasets is equal to the number of clustering results. In this example, there are four clustering results, so four sample datasets are constructed. Each sample dataset is divided into a training set and a test set to facilitate the subsequent training and testing of the Transformer-LSTM model.

[0060] Step 6: Build the Transformer-LSTM model.

[0061] The Transformer model uses a parallel mechanism to make the model structure more lightweight. However, when the Transformer model is applied to time series data prediction, the input sequence may lack time information, which affects the prediction results. In order to solve this technical problem, the present invention replaces the attention mechanism module in the decoder of the Transformer model with LSTM (Long Short-Term Memory) to obtain the Transformer-LSTM model, such as Figure 4 As shown in the figure, LSTM, as a recurrent neural network, is more suitable for modeling time series data prediction tasks. It can better capture the dynamic pattern of embankment deformation changes and improve the accuracy of embankment deformation prediction.

[0062] Step 7: Use each sample dataset to train and test the Transformer-LSTM model to obtain the embankment deformation prediction model for each time period.

[0063] This example uses four sample datasets. Each sample dataset is used to train and test the Transformer-LSTM model, resulting in a dam deformation prediction model corresponding to the time period of the sample dataset. This yields four dam deformation prediction models. In a specific embodiment of the present invention, during training, a grid search algorithm is used to globally solve the hyperparameters of the Transformer-LSTM model based on the sample datasets to find the optimal hyperparameters for the Transformer-LSTM model.

[0064] After the training is completed, each sample in the test set is input into the trained Transformer-LSTM model to obtain the prediction results. Then all the prediction results are summarized in chronological order, and the trained Transformer-LSTM model is evaluated using the evaluation index to obtain the embankment deformation prediction model. In this embodiment, the evaluation index is the determination coefficient R 2 , root mean square error RMSE, mean absolute percentage error MAPE.

[0065] Figure 5 The comparison of the evaluation results of the model of the present invention (i.e., DF-Transformer-LSTM model) and other models is shown. The DF-Transformer model represents a model based on cluster analysis and Transformer, the DF-LSTM model represents a model based on cluster analysis and LSTM, and the Transformer-LSTM model represents a model based on Transformer-LSTM. Figure 5 As can be seen, the DF-Transformer-LSTM model of the present invention achieved the best results in all evaluation indicators. In the comparison of various models on the test set, the DF-Transformer-LSTM model outperformed the Transformer-LSTM model, indicating that the clustering method effectively analyzed the characteristics of the environmental data and effectively improved the accuracy of the prediction. The DF-Transformer-LSTM model outperformed the DF-Transformer model, indicating that the Transformer-LSTM model built on the Transformer effectively improved the prediction accuracy.

[0066] Figure 6 The figure shows the comparison between the prediction effect of the test set and the measured data of the measurement points. Figure 6 It can be seen that the prediction results of the present invention are close to the measured data, indicating that it has a high accuracy in predicting dam deformation.

[0067] Step 8: Obtain the data to be predicted, and use the embankment deformation prediction model corresponding to the time period to which the data to be predicted belongs to predict the data to obtain the embankment deformation prediction result.

[0068] The data to be predicted include reservoir water level, environmental data and time factors. By inputting the data to be predicted into the embankment deformation prediction model corresponding to the time period to which the data to be predicted belongs, the embankment displacement can be output to realize the embankment deformation prediction.

[0069] Example 2

[0070] An embodiment of the present invention further provides an electronic device comprising: a memory, a processor, and a computer program / instructions stored in the memory, wherein the processor executes the computer program / instructions to implement the dam deformation prediction method in the embodiment of the present application.

[0071] Although not shown, the electronic device includes a processor that can perform various appropriate operations and processes based on programs and / or data stored in a read-only memory (ROM) or programs and / or data loaded from a storage portion into a random access memory (RAM). The processor can be a multi-core processor or can include multiple processors. In some embodiments, the processor can include a general-purpose main processor and one or more special coprocessors, such as a central processing unit, a graphics processing unit (GPU), a neural network processor (NPU), a digital signal processor (DSP), etc. Various programs and data required for device operation are also stored in RAM. The processor, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0072] The processor and memory are used together to execute the program / instructions stored in the memory. When the program / instructions are executed by the computer, the methods, steps or functions described in the above embodiments can be implemented.

[0073] Although not shown, an embodiment of the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the dam deformation prediction method in the embodiment of the present application.

[0074] Computer-readable storage media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0075] Although not shown, an embodiment of the present invention further provides a computer program product, including: a computer program / instruction, which, when executed by a processor, implements the dam deformation prediction method in the embodiment of the present application.

[0076] The above disclosure is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or modifications within the technical scope disclosed in the present invention, and they should all be covered by the scope of protection of the present invention.

Claims

1. A method for predicting dam deformation, characterized in that: The prediction method comprises: Acquire dam monitoring data; wherein the dam monitoring data includes reservoir water level, environmental data, time-effect factors, and displacement of dam measuring points; the time-effect factors are stepwise regression equations of time-effect parameters, and the time-effect parameters are the time rate of change from the monitoring time to the start time of dam operation; Preprocessing the dam monitoring data; Perform deep feature extraction on the preprocessed environmental data to obtain environmental features; Performing cluster analysis on the pre-processed reservoir water level and the environmental characteristics, and dividing the time-effect factors and the displacement of the dam measuring points in the time domain according to the cluster analysis results to obtain cluster division results for different time periods; Construct a sample data set based on each clustering result; wherein the number of sample data sets is the same as the number of clustering results; Constructing a Transformer-LSTM model; wherein the Transformer-LSTM model is obtained by replacing the decoder of the Transformer model with LSTM; The Transformer-LSTM model is trained and tested using each sample data set to obtain a dam deformation prediction model for each time period; The data to be predicted is obtained, and the prediction data is predicted using a dam deformation prediction model corresponding to the time period to which the data to be predicted belongs, so as to obtain a dam deformation prediction result.

2. The dam deformation prediction method according to claim 1, characterized in that: The environmental data include air temperature, water temperature, dam surface temperature and dam foundation temperature.

3. The dam deformation prediction method according to claim 1, characterized in that: The Transformer model is used to perform deep feature extraction on the preprocessed environmental data to obtain environmental features.

4. The dam deformation prediction method according to claim 1, characterized in that: The deep fuzzy clustering algorithm is used to perform cluster analysis on the preprocessed reservoir water level and the environmental characteristics.

5. The dam deformation prediction method according to claim 4, characterized in that: When the deep fuzzy clustering algorithm is used to perform cluster analysis on the pre-processed reservoir water level and the environmental characteristics, the particle swarm optimization algorithm is used to optimize the randomly assigned initial cluster centers.

6. The dam deformation prediction method according to any one of claims 1 to 5, characterized in that: When training the Transformer-LSTM model, a grid search algorithm is used to globally solve the hyperparameters of the Transformer-LSTM model based on the sample data set.

7. An electronic device comprising a memory, a processor, and a computer program / instruction stored in the memory, characterized in that: The processor executes the computer program / instructions to implement the dam deformation prediction method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the dam deformation prediction method according to any one of claims 1 to 6 is implemented.

9. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the dam deformation prediction method according to any one of claims 1 to 6 is implemented.

Images