An arch dam body shape multi-objective intelligent design and optimization method based on machine learning

Abstract

The application belongs to the technical field of arch dam shape design and artificial intelligence, and discloses a kind of arch dam shape multi-objective intelligent design and optimization method based on machine learning;Including the following steps: step 1, collect and arrange arch dam shape design data of arch dam shape;Step 2, carry out data cleaning to the collected arch dam shape design data, carry out correlation analysis based on arch dam shape design data, and utilize principal component analysis to carry out feature dimension reduction, complete the feature engineering of arch dam shape design data;Step 3, call the machine learning model of design optimization, and input the input parameter to be designed, utilize machine learning model to carry out intelligent design and optimization of arch dam shape design, output arch dam crown beam curve equation and horizontal arch ring line equation.The application realizes multi-objective intelligent design and optimization of arch dam shape design, effectively improves the efficiency of arch dam shape design.

Description

Technical Field

[0001] This invention relates to the fields of arch dam shape design and artificial intelligence technology, and more specifically, to a multi-objective intelligent design and optimization method for arch dam shape based on machine learning. Background Technology

[0002] As a key structure in water conservancy and hydropower projects, the shape design of arch dams directly affects the safety, economy, and construction difficulty of the project. Traditional arch dam design mainly relies on engineers' experience and extensive manual calculations. While this method can meet engineering requirements, it has the following problems:

[0003] The design of arch dams involves complex hydraulic, geotechnical, and structural mechanics calculations, and the experience accumulated by senior engineers is difficult to pass on systematically.

[0004] Traditional arch dam design typically requires extensive parameter adjustments and optimization calculations, involving complex processes such as nonlinear optimization and trial adjustments. This results in low design efficiency, difficulty in meeting the demands of rapid construction in modern engineering projects, and a low level of intelligent design.

[0005] Traditional methods rely mainly on engineers' experience and analogy for optimization and adjustment, lacking sufficient data support, making it difficult to quickly identify the best design solution among numerous influencing factors, and thus limiting optimization capabilities.

[0006] In view of this, the present invention proposes a multi-objective intelligent design and optimization method for arch dam shape based on machine learning to solve the above problems. Summary of the Invention

[0007] To overcome the aforementioned shortcomings of the existing technology and to achieve the above objectives, the present invention provides the following technical solution: a multi-objective intelligent design and optimization method for arch dam shapes based on machine learning, comprising the following steps:

[0008] Step 1: Collect and organize the arch dam body design data, which includes: the curve equation of the upstream face of the arch crown beam, the thickness equation, the thickness of the top of the arch crown, the thickness of the bottom of the arch crown, the horizontal arch ring line equation, the central angle, the arc length of the dam crest, the dam height, and the foundation surface parameters.

[0009] Step 2: Clean the collected arch dam shape design data, conduct correlation analysis based on the arch dam shape design data, and use principal component analysis to perform feature dimensionality reduction to complete the feature engineering of the arch dam shape design data. This will yield the input and output parameters for the extracted machine learning model. The input parameters are the processed arch dam shape design data, and the output parameters are the arch dam shape design results, which include the arch crown beam curve equation and the horizontal arch ring line equation.

[0010] Step 3: Call the machine learning model for design optimization, input the input parameters to be designed, and use the machine learning model to perform intelligent design and optimization of the arch dam shape design, outputting the arch crown beam curve equation and the horizontal arch ring line equation.

[0011] Furthermore, the step of collecting and organizing the arch dam shape design data includes:

[0012] Step 101: Collect the curve equation of the upstream face of the arched beam, the thickness equation, the thickness of the top and bottom of the arch, the horizontal arch line equation, the central angle, the arc length of the dam crest, the dam height, and the foundation surface parameters as arch dam body design data. Among them, the foundation surface parameters include the valley cross-sectional shape and the physical and mechanical parameters of the dam foundation rock mass. The physical and mechanical parameters of the dam foundation rock mass include the uniaxial saturated compressive strength of the rock mass, the rock mass structure, the degree of rock mass weathering, the deformation model, and the shear resistance parameters.

[0013] Step 102: Use vectors to represent the physical and mechanical parameters of the dam foundation rock mass. Use floating-point numbers to represent the key parameters of the dam foundation rock mass physical and mechanical parameters, and use unique heat vectors to represent the key categories.

[0014] Furthermore, the steps of cleaning the collected arch dam shape design data, conducting correlation analysis based on the arch dam shape design data, and using principal component analysis for feature dimensionality reduction to complete the feature engineering of the arch dam shape design data include:

[0015] Step 201: Perform data feature processing on the arch dam shape design data processed in step 102;

[0016] Step 202: Perform correlation analysis on the data characteristics of the arch dam shape design data, use the Pearson correlation coefficient to measure the correlation between high-dimensional features, clean up features with excessive correlation, carry out principal component analysis to reduce feature dimensionality, refine complex and redundant features into key features as input parameters, i.e., the processed arch dam shape design data, and use the upstream curve equation of the arch crown beam and the horizontal arch ring line equation as output parameters, i.e., complete the feature engineering of the arch dam shape design data.

[0017] Furthermore, the steps for building, training, testing, and evaluating the machine learning model include:

[0018] Step 301: Before the machine learning model has the ability to design partitions, it is necessary to construct the corresponding training dataset and test dataset of the parallel machine learning model for model training and testing. The training dataset and test dataset are constructed using the feature representation of the input parameters and output parameters.

[0019] Step 302: Build a machine learning model, which consists of an artificial neural network model, the Bagging algorithm, and a random forest, and optimize the machine learning model using an intelligent design optimization algorithm;

[0020] Step 303: Based on the constructed training dataset, test dataset, and machine learning model, train and test the model. Once the design accuracy meets the requirements, it can be used for design.

[0021] Step 304: During the training and validation of the machine learning model, k-fold cross-validation is used, and the root mean square error (MSE) is used as the error metric for accuracy evaluation.

[0022] Further, in step 302,

[0023] An intelligent design optimization algorithm is constructed using ensemble learning technology. By combining the advantages of multiple algorithms, more accurate equations for the upstream surface curve of the arch crown beam and the horizontal arch ring shape are obtained in the arch dam body design.

[0024] Meanwhile, based on the intelligent design optimization algorithm, the hyperparameters of the constructed machine learning model are tuned to obtain the optimized design algorithm.

[0025] Furthermore, the intelligent design optimization algorithm includes a genetic algorithm.

[0026] Further, in step 302, the step of building the machine learning model includes:

[0027] Step 3021: Using artificial neural networks as the basic component of the machine learning model, construct a multi-layer structure containing an input layer, a hidden layer, and an output layer. The input layer receives the processed arch dam shape design data, the hidden layer uses the connection weights between neurons to perform nonlinear transformation and feature extraction on the data, and the output layer outputs the predicted arch dam shape design result.

[0028] Step 3022: Adjust the number of hidden layers, neurons, and connection weights. The artificial neural network model can continuously optimize its learning and prediction capabilities on arch dam shape design data, and use the Bagging algorithm to enhance the stability and generalization ability of the machine learning model.

[0029] Further, in step 3022,

[0030] The Bagging algorithm samples the original training and test datasets multiple times with replacement to generate multiple different subsets of the dataset, and trains a base model for each subset.

[0031] In the prediction phase, the prediction results of each base model are combined, usually by using a voting method or an averaging method to obtain the final prediction result.

[0032] Furthermore, each decision tree in the random forest is trained on a subset of the generated dataset based on the Bagging algorithm, and when constructing the decision tree, for each node split, a subset of features is randomly selected to find the optimal split point, rather than considering all features.

[0033] Furthermore, the steps of calling the machine learning model for design optimization, inputting the input parameters to be designed, using the machine learning model for intelligent design and optimization of the arch dam's shape, and outputting the arch crown beam curve equation and the horizontal arch ring shape equation include:

[0034] The input parameters to be designed are input into the trained machine learning model, which then outputs the predicted curve equation of the arched beam and the linear equation of the horizontal arch ring.

[0035] The technical effects and advantages of the multi-objective intelligent design and optimization method for arch dam shape based on machine learning in this invention are as follows:

[0036] 1. This invention enables intelligent design and multi-objective optimization of the arch dam shape design. The optimization includes key design parameters of the arch dam and optimization of the upstream curve equation of the arch crown beam: the upstream curve equation of the arch crown beam is used as the optimization objective. The aim is to adjust its characteristic parameters so that the shape of the upstream face of the arch dam is more in line with engineering requirements. A suitable upstream curve shape helps to improve the stress distribution of the arch dam and enhance its stability.

[0037] 2. By optimizing the characteristic parameters of the horizontal arch line equation, this invention can make the shape of the horizontal arch more reasonable. A reasonable horizontal arch line can better transmit the load on the dam body, improve the overall bearing capacity and safety of the arch dam, and achieve intelligent and precise design through a series of technical means; it greatly improves design efficiency and enhances the level of intelligence in arch dam design. Attached Figure Description

[0038] Figure 1 This is a flowchart of a multi-objective intelligent design and optimization method for arch dam shape based on machine learning, according to the present invention.

[0039] Figure 2 This is a schematic diagram of the process of the present invention;

[0040] Figure 3 For the present invention Figure 2 Enlarged schematic diagram of the second part;

[0041] Figure 4 For the present invention Figure 2 An enlarged schematic diagram of the third part. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Example 1

[0044] Please see Figure 1 , Figure 2 , Figure 3 as well as Figure 4 As shown in this embodiment, a multi-objective intelligent design and optimization method for arch dam shapes based on machine learning includes the following steps:

[0045] Step 1: Collect and organize the arch dam body design data, which includes: the curve equation of the upstream face of the arch crown beam, the thickness equation, the thickness of the top of the arch crown, the thickness of the bottom of the arch crown, the horizontal arch ring line equation, the central angle, the arc length of the dam crest, the dam height, and the foundation surface parameters.

[0046] Step 2: Clean the collected arch dam shape design data, conduct correlation analysis based on the arch dam shape design data (such as upstream surface curve equation, arch ring line equation, etc.), and use principal component analysis (PCA) to perform feature dimensionality reduction to complete the feature engineering of the arch dam shape design data, that is, to obtain the input parameters and output parameters of the extracted machine learning model. The input parameters are the processed arch dam shape design data, and the output parameters are the arch dam shape design results, including the arch crown beam curve equation and the horizontal arch ring line equation.

[0047] Step 3: Call the machine learning model for design optimization, input the input parameters to be designed, and use the machine learning model to perform intelligent design and optimization of the arch dam shape design, outputting the arch crown beam curve equation and the horizontal arch ring line equation.

[0048] Furthermore, the step of collecting and organizing the arch dam shape design data includes:

[0049] Step 101: Collect the curve equation of the upstream face of the arch crown beam, thickness equation, top thickness of the arch crown, bottom thickness, horizontal arch ring line equation, central angle, dam crest arc length, dam height, and foundation surface parameters as arch dam body design data. Among them, the foundation surface parameters include the valley cross-section shape and the physical and mechanical parameters of the dam foundation rock mass. The physical and mechanical parameters of the dam foundation rock mass include the uniaxial saturated compressive strength of the rock mass, rock mass structure, rock mass weathering degree, deformation model, and shear (fracture) resistance parameters, etc.

[0050] Step 102: Use vectors to represent the physical and mechanical parameters of the dam foundation rock mass. Use floating-point numbers to represent the key parameters of the dam foundation rock mass physical and mechanical parameters, and use unique heat vectors to represent the key categories.

[0051] Furthermore, the steps of cleaning the collected arch dam shape design data, conducting correlation analysis based on data features (such as upstream surface curve equations and arch ring line equations) in the arch dam shape design data, and using principal component analysis (PCA) for feature dimensionality reduction to complete the feature engineering of the arch dam shape design data include:

[0052] Step 201: Perform data feature processing on the arch dam shape design data processed in step 102;

[0053] Step 202: Perform correlation analysis on the data characteristics of the arch dam shape design data, use Pearson correlation coefficient and other measures to obtain the correlation between high-dimensional features, clean up features with excessive correlation, carry out principal component analysis (PCA) to reduce feature dimensionality, refine complex and redundant features into key features as input parameters, that is, the processed arch dam shape design data, and use the upstream curve equation of the arch crown beam and the horizontal arch ring line equation as output parameters, that is, complete the feature engineering of the arch dam shape design data;

[0054] Step 203: Construct training and testing datasets using the feature representations of input and output parameters;

[0055] It should be noted that normalizing the data features yields the input parameters for the arch dam shape design. Correlation analysis is then performed on the arch dam design data features, using metrics such as the Pearson correlation coefficient to determine the correlation between high-dimensional features. Features with excessively high correlation are cleaned up, and principal component analysis (PCA) is conducted for feature dimensionality reduction, refining complex and redundant features into key features as input. The Pearson correlation coefficient is then used to measure the correlation between design data features, clearly revealing the degree of association between high-dimensional features. When two or more features are highly correlated, it means they may carry similar information. For example, some features describing the geometry of an arch dam may be highly correlated due to mutual influence. By setting an appropriate correlation threshold, excessively correlated features are cleaned up, preventing the model from learning repetitive information, thereby reducing computational load, preventing overfitting, and allowing the model to focus on key information.

[0056] PCA (Principal Composition Analysis) is a powerful linear transformation technique that converts complex and redundant original features into a set of uncorrelated principal components. These principal components are linear combinations of the original features, ordered according to their variance contributions. In the data processing for arch dam design, PCA identifies the main directions of change in the data by eigenvalue decomposition of the data covariance matrix. For example, numerous original features describing the arch dam shape and geological conditions can be transformed into a few comprehensive principal components. Typically, the first few principal components retain most of the variance information of the original data, i.e., the key features. By selecting these principal components as model input, while retaining important information, the data dimensionality is significantly reduced, improving the efficiency of model training and prediction. This allows the model to learn the relationship between key factors and objectives in arch dam design more concisely and efficiently, ultimately outputting the arch crown beam curve equation and the horizontal arch ring alignment parameters.

[0057] It should be noted that the feature engineering of the arch dam shape design data requires: obtaining the parameter distribution pattern: collecting existing arch dam shape design data, including images and design information text, extracting key features from the existing arch dam shape design data of drawings and text data, statistically analyzing the cross-sectional view of the arch crown beam and the design drawing of the horizontal arch ring in the images, and statistically analyzing the curve equation, the linear equation of the horizontal arch ring, and the physical and mechanical parameters of the dam foundation rock mass in the text to obtain the distribution pattern of different parameters. Among them, the key feature extraction of drawing data: for the cross-sectional view of the arch crown beam and the design drawing of the horizontal arch ring, AutoCAD is used to directly measure and mark parameters such as the thickness of the arch crown beam and the thickness of the horizontal arch ring.

[0058] Key Feature Extraction from Text Data: Utilizing natural language processing technology, for texts containing curve equations and horizontal arch ring linear equations, mathematical expressions and related parameters are identified and extracted through syntactic and semantic analysis. For texts containing physical and mechanical parameters of the dam foundation rock mass, information extraction technology is employed to extract key parameters such as uniaxial saturated compressive strength, rock mass structure, and rock mass weathering degree from engineering reports, design documents, and other texts. Statistical analysis is performed on the extracted parameters, calculating statistical quantities such as mean, median, and standard deviation. For parameters of the upstream face curve equation of the arch crown beam, the mean and standard deviation of these parameters in different arch dams are calculated to understand their central tendency and dispersion. Frequency distribution histograms of the parameters are plotted to visually display the frequency of occurrence of parameters in different value ranges, thus clearly showing the distribution of parameters. For example, a frequency distribution histogram of the uniaxial saturated compressive strength of the dam foundation rock mass is plotted to observe the distribution characteristics of its strength values.

[0059] It should be noted that the construction of the training and test datasets involves the following: Parametric generation of multi-source heterogeneous datasets: Based on the statistical distribution of parameters, parametric adjustment techniques are employed. According to the previously obtained distribution patterns of different parameters, parameters that have a significant impact on the design of the arch dam shape are selected as adjustment targets. Within the set parameter adjustment range, different parameters are quickly generated through proportional scaling and random perturbation methods. This augments and expands the geometric and physical mechanics parameters, thus constructing multi-source heterogeneous training and test datasets.

[0060] Furthermore, the steps for building, training, testing, and evaluating the optimized machine learning model include:

[0061] Step 301: Construct training and testing datasets using the feature representations of input and output parameters;

[0062] Step 302: Build a machine learning model, which consists of an artificial neural network model, the Bagging algorithm, and a random forest, and optimize the machine learning model using an intelligent design optimization algorithm;

[0063] Step 303: Based on the constructed training dataset, test dataset, and machine learning model, train and test the model. Once the design accuracy meets the requirements, it can be used for design.

[0064] Step 304: During the training and validation of the machine learning model, k-fold cross-validation is used, and the root mean square error (MSE) is used as the error metric for accuracy evaluation.

[0065] It should be noted that k-fold cross-validation is used during model training and validation to address the issue of insufficient accuracy in intelligent design due to limited data volume. Root mean square error (MSE) is used as the error metric for accuracy evaluation. The machine learning model includes algorithms based on multivariate regression, artificial neural networks (ANN), random forests, and XGBoost. An intelligent design optimization algorithm is constructed using ensemble learning techniques, combining the advantages of multiple algorithms to obtain more accurate equations for the upstream curve of the arch crown beam and the horizontal arch ring shape in the arch dam design. Simultaneously, hyperparameter tuning of the constructed machine learning model is performed using optimization algorithms such as genetic algorithms to obtain an optimized design algorithm. The machine learning model is built, comprising an artificial neural network model, Bagging algorithm, and random forest. Based on the constructed dataset and machine learning model, model training and testing are conducted. Once the design accuracy meets the requirements, it can be used for design.

[0066] Further, in step 302,

[0067] An intelligent design optimization algorithm is constructed using ensemble learning technology. By combining the advantages of multiple algorithms, more accurate equations for the upstream surface curve of the arch crown beam and the horizontal arch ring shape are obtained in the arch dam body design.

[0068] Meanwhile, based on the intelligent design optimization algorithm, the hyperparameters of the constructed machine learning model are tuned to obtain the optimized design algorithm.

[0069] Furthermore, the intelligent design optimization algorithm includes a genetic algorithm.

[0070] Further, in step 302, the step of building the machine learning model includes:

[0071] Step 3021: Using artificial neural networks as the foundational component of a machine learning model, construct a multi-layer structure comprising an input layer, hidden layers, and an output layer. The input layer receives the processed arch dam shape design data, the hidden layer utilizes the connection weights between neurons to perform nonlinear transformations and feature extraction on the data, and the output layer outputs the predicted arch dam shape design result.

[0072] It should be noted that artificial neural network models are used as a fundamental component of machine learning models. These models possess powerful nonlinear mapping capabilities, enabling them to learn complex data patterns and relationships. In arch dam design, they can capture the complex nonlinear relationships between design parameters (such as the upstream curve equation parameters of the arch crown beam, dam height, and valley cross-sectional shape) and target results (the arch crown beam curve equation and the horizontal arch ring line shape equation). By constructing a multi-layered structure containing input, hidden, and output layers, the input layer receives processed arch dam design data, the hidden layers utilize the connection weights between neurons to perform nonlinear transformations and feature extraction on the data, and the output layer outputs the predicted arch dam design results. By adjusting the number of hidden layers, the number of neurons, and the connection weights, the artificial neural network model can continuously optimize its learning and prediction capabilities for arch dam design data.

[0073] Step 3022: Adjust the number of hidden layers, neurons, and connection weights. The artificial neural network model can continuously optimize its learning and prediction capabilities on arch dam shape design data, and use the Bagging algorithm to enhance the stability and generalization ability of the machine learning model.

[0074] It's important to note that the Bagging algorithm enhances the stability and generalization ability of a model. Based on bootstrapping, it samples multiple times with replacement from the original training and test datasets to generate several distinct subsets. For each subset, a base model is trained (this base model can be an artificial neural network or a decision tree in a random forest). During the prediction phase, the predictions from each base model are combined, typically using a voting method (for classification problems) or an averaging method (for regression problems) to obtain the final prediction. In this way, the Bagging algorithm effectively reduces the variance of a single model, minimizes the risk of overfitting, and improves the overall stability and generalization performance of the model, enabling the machine learning model to exhibit relatively stable predictive capabilities when faced with design data for different arch dam shapes.

[0075] Random Forest: Random forest is an ensemble model composed of multiple decision trees, playing a crucial role in this ensemble learning architecture. Each decision tree is trained on a subset of the dataset generated by the Bagging algorithm. In the process of building the decision tree, for each node split, a subset of features is randomly selected to find the optimal split point, rather than considering all features. This randomization operation gives different decision trees a certain degree of independence, further enhancing the model's generalization ability.

[0076] Further, in step 3022,

[0077] The Bagging algorithm samples the original training and test datasets multiple times with replacement to generate multiple different subsets of the dataset, and trains a base model for each subset.

[0078] In the prediction phase, the prediction results of each base model are combined, usually by using a voting method or an averaging method to obtain the final prediction result.

[0079] Furthermore, each decision tree in the random forest is trained on a subset of the generated dataset based on the Bagging algorithm, and when constructing the decision tree, for each node split, a subset of features is randomly selected to find the optimal split point, rather than considering all features.

[0080] It's important to note that random forests, by integrating the predictions of multiple decision trees, can more accurately capture complex patterns and relationships in data. In arch dam design, random forests can handle high-dimensional design data, uncovering potential connections between different design parameters and providing rich information for predicting dam shapes. They complement artificial neural network models and the Bagging algorithm, jointly improving the performance of the machine learning model. In this machine learning model, the artificial neural network model, the Bagging algorithm, and the random forest work together. The Bagging algorithm provides diverse training data for the artificial neural network model and the random forest, reducing model variance; the artificial neural network model utilizes its non-linear learning capabilities to handle complex design data relationships; and the random forest, through the ensemble of multiple decision trees, enhances the model's generalization ability and its ability to handle high-dimensional data. Finally, the predictions from multiple models are fused to output a more accurate and reliable prediction of the arch dam shape design. Based on the constructed dataset and machine learning model, model training and testing are performed, and once the design accuracy meets the requirements, it can be used for design.

[0081] Furthermore, the steps of inputting the input parameters to be designed, using a machine learning model for intelligent design and optimization of the arch dam's shape, and outputting the arch crown beam curve equation and the horizontal arch ring shape equation include:

[0082] The input parameters to be designed are input into the trained machine learning model, which then outputs the predicted curve equation of the crown beam and the linear equation of the horizontal arch ring.

[0083] It should be noted that the reference Figure 3 In the design of arch dams, y cu (z) is the mathematical expression for the upstream surface of the arched beam. The equation of the upstream surface is generally set as a cubic curve of the z-coordinate, i.e., y = (z - z) / (z - z). cu (z) = a0 + a1z + a2z 2 +a3z 3 In the formula, y cu (z) is a function with elevation z as the independent variable, used to quantitatively describe the geometric profile of the upstream surface of the arched beam, where a0, a1, a2, and a3 are the parameters of the upstream surface of the arched beam;

[0084] T c (z) is the mathematical expression for the thickness equation of the arched beam. Generally, the thickness of the arched beam is also a cubic curve of the z-coordinate, i.e., T. c (z) = b0 + b1z + b2z 2 +b3z 3 In the formula, T c (z) is a function of height z, and b0, b1, b2, b3 are parameters of the thickness equation of the arched beam;

[0085] Furthermore, the equation y of the downstream surface of the arched beam cd The mathematical expression for (z) is: y cd (z)=y cu (z)+T c (z), where y cu (z), T c (z) represents the upstream curved surface of the arched beam and the thickness of the arched beam, respectively;

[0086] In arch dam design, the arch axis is generally a curve, but it can also be a quadratic curve such as a parabola, ellipse, or hyperbola. These different curve forms can be comprehensively described by a quadratic equation, namely x. 2 =ay 2 +by, where a and b are parameters that determine this quadratic equation, x represents the horizontal distance from a point on the cross-section of the arch dam to the centerline of the arch axis, and y represents the vertical distance (i.e., elevation) from that point to the bottom of the arch dam. Note that when a > 0, the quadratic equation is a hyperbola; when a < 0, the quadratic equation is an ellipse. For an ellipse, there is a definition for the radius of curvature, R, which is expressed as: In the formula, the coefficients a and b in the arch axis equation are predicted by machine learning, and the radius of curvature R can be obtained from the predicted a and b. For example, for y=a+bx, it is necessary to determine this straight line, that is, to determine a and b, and this straight line can be represented in space. The same applies to the design of arch dams, where a... 0、 a 1、 a 2、 a 3、 b 0、 b1、b 2、 b3 is the equation of the upstream surface of the arched beam. cu The parameters (coefficients) of the thickness equation Tc(z) and the crown beam are determined when determining these two cubic functions. These relevant parameters are all predicted by machine learning in this invention.

[0087] In this embodiment, the present application enables intelligent design and multi-objective optimization of the arch dam's shape design. The optimization includes key design parameters of the arch dam and optimization of the upstream curve equation of the arch crown beam: using the upstream curve equation of the arch crown beam as the optimization objective, the aim is to adjust its characteristic parameters to make the shape of the upstream face of the arch dam more in line with engineering requirements. A suitable upstream curve shape helps to improve the stress distribution of the arch dam and enhance its stability. By optimizing the characteristic parameters of the horizontal arch ring's shape equation, the shape of the horizontal arch ring can be made more reasonable. A reasonable horizontal arch ring shape can better transmit the load on the dam body, improve the overall bearing capacity and safety of the arch dam, and achieve intelligent and precise design through a series of technical means; significantly improving design efficiency and enhancing the level of intelligence in arch dam design.

[0088] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0089] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0090] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0091] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A machine learning-based multi-objective intelligent design and optimization method for arch dam body shape, characterized in that, The method comprises the following steps: Step 1, collecting and sorting arch dam body type design data of arch dam body type, the arch dam body type design data comprising: upstream curve equation of arch crown beam, thickness equation, arch crown top thickness, bottom thickness, horizontal arch ring line type equation, central angle, dam top arc length, dam height and foundation plane parameters; Step 2, data cleaning is performed on the collected arch dam body type design data, correlation analysis based on the arch dam body type design data is carried out, and feature dimension reduction is performed by using principal component analysis to complete feature engineering of the arch dam body type design data, that is, input parameters and output parameters of an extracted machine learning model are obtained, wherein the input parameters are processed arch dam body type design data, and the output parameters are arch dam body type design results, the arch dam body type design results comprising arch crown beam curve equation and horizontal arch ring line type equation; Step 3, calling a design-optimized machine learning model, inputting the input parameters to be designed, and using the machine learning model to perform intelligent design and optimization of arch dam body type design, and outputting arch crown beam curve equation and horizontal arch ring line type equation; The steps of building, training, testing and evaluating the machine learning model comprise: Step 301: constructing a training data set and a test data set by using feature expression of input parameters and output parameters; Step 302: building a machine learning model, the model comprising an artificial neural network model, a Bagging algorithm and a random forest, and optimizing the machine learning model by using an intelligent design optimization algorithm; Step 303: based on the constructed training data set, test data set and machine learning model, model training and testing are performed, and the design precision meets the requirements and can be used for design; Step 304: during the training and verification of the machine learning model, k-fold cross-validation is adopted, and the root mean square error (MSE) is used as an error measurement standard for precision evaluation; In the step 302, the step of building the machine learning model comprises: Step 3021, taking an artificial neural network as a basic component of the machine learning model, constructing a multi-layer structure comprising an input layer, a hidden layer and an output layer, the input layer receiving processed arch dam body type design data, the hidden layer performing nonlinear transformation and feature extraction on the data by using connection weights between neurons, and the output layer outputting predicted arch dam body type design results; Step 3022, adjusting the number of layers, the number of neurons and the connection weights of the hidden layer, continuously optimizing the learning and prediction ability of the artificial neural network model on the arch dam body type design data, and enhancing the stability and generalization ability of the machine learning model by using the Bagging algorithm; In the step 3022, The Bagging algorithm samples multiple different sub-data sets from the original training data set and test data set with replacement, trains a base model for each sub-data set; In the prediction stage, the prediction results of each base model are integrated, and the final prediction result is obtained by using a voting method or an averaging method. And each decision tree in the random forest is trained on a generated subset of data based on the Bagging algorithm, and when building the decision tree, for each node split, a subset of features is randomly selected to find the best split point.

2. The machine learning-based multi-objective intelligent design and optimization method for arch dam shape according to claim 1, characterized in that, The step of collecting the arch dam body type design data of the arch dam body type includes: Step 101: Collect the arch crown beam upstream surface curve equation, thickness equation, arch crown top thickness, bottom thickness, horizontal arch ring line type equation, central angle, dam top arc length, dam height and foundation surface parameters as the arch dam body type design data, wherein the foundation surface parameters include the valley section shape and the dam foundation rock mass physical and mechanical parameters, and the dam foundation rock mass physical and mechanical parameters include the rock mass uniaxial saturated compressive strength, rock mass structure, rock mass weathering degree, deformation model and shear resistance parameter; Step 102: The vector is used to represent the text of the dam foundation rock mass physical and mechanical parameters, and the key parameters of the dam foundation rock mass physical and mechanical parameters are represented by floating point numbers, and the key categories are represented by one-hot vectors.

3. The machine learning-based multi-objective intelligent design and optimization method for arch dam shape according to claim 2, characterized in that, The step of carrying out data cleaning on the collected arch dam body type design data, carrying out correlation analysis based on the arch dam body type design data, and utilizing principal component analysis for feature dimension reduction to complete the feature engineering of the arch dam body type design data includes: Step 201: Data feature processing is performed on the arch dam body type design data processed in step 102; Step 202: Correlation analysis is performed on the data features of the arch dam body type design data, the Pearson correlation coefficient is used to measure the correlation between high-dimensional features, the features with too high correlation are cleaned, principal component analysis is performed for feature dimension reduction, and complex redundant features are refined into key features as input parameters, that is, the processed arch dam body type design data, and the arch crown beam curve equation and the horizontal arch ring line type equation are used as output parameters, that is, the feature engineering of the arch dam body type design data is completed.

4. The machine learning-based multi-objective intelligent design and optimization method for arch dam shape according to claim 1, characterized in that, In the step 302, An integrated learning technology is used to construct an intelligent design optimization algorithm, and the advantages of multiple algorithms are combined to obtain more accurate arch crown beam curve equation and horizontal arch ring line type equation in arch dam body type design; At the same time, based on the intelligent design optimization algorithm, the hyperparameters of the constructed machine learning model are optimized to obtain an optimized design algorithm.

5. The machine learning based multi-objective intelligent design and optimization method for arch dam shape according to claim 4, characterized in that, The intelligent design optimization algorithm includes a genetic algorithm.

6. The machine learning based multi-objective intelligent design and optimization method for arch dam shape according to claim 1, characterized in that, The step of inputting the input parameters to be designed, utilizing the machine learning model to intelligently design and optimize the arch dam body type, and outputting the arch crown beam curve equation and the horizontal arch ring line type equation includes: The input parameters to be designed are input into the trained machine learning model, and the machine learning model outputs the predicted arch crown beam curve equation and horizontal arch ring line type equation.

Images