Thermal fatigue assessment method and device for safety hook rigging based on artificial intelligence

Abstract

The present application relates to the field of fatigue testing, and in particular to a method and device for thermal fatigue assessment of safety hook rigging based on artificial intelligence; the method comprises: obtaining current monitoring data of the hook rigging to be assessed; obtaining a thermal fatigue prediction result of the hook rigging to be assessed through a thermal fatigue prediction model according to the current monitoring data; wherein the thermal fatigue prediction model is obtained by training with historical monitoring data and fatigue test data on sample hook rigging; the historical monitoring data is the state data of the sample hook rigging during actual use; the fatigue test data is the state data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be assessed; the present application can solve the technical problem existing in the prior art of lacking a method for determining the thermal fatigue state of the hook rigging.

Description

Technical Field

[0001] The present application relates to the field of fatigue testing, and in particular to an artificial intelligence-based thermal fatigue assessment method and device for safety hook rigging. Background Art

[0002] Materials or heavy equipment are required during the construction and maintenance of infrastructure such as power plants and mines. Hook slings are important tools in the handling process, and their safety, reliability and durability need to be strictly monitored and evaluated to ensure that they can play a safe role during the handling process.

[0003] In actual operation, equipment thermal fatigue is an important factor leading to the failure of hook rigging. In particular, when the hook rigging is subjected to high temperature and high stress for a long time, its fatigue damage gradually accumulates, which may eventually cause cracks, deformation or even breakage of the hook rigging. Among them, thermal fatigue not only affects the service life of the hook rigging, but may also cause serious safety accidents.

[0004] In practical applications, although determining the thermal fatigue state of the hook rigging is crucial to the safety of the hook rigging, there is a lack of corresponding technical solutions in the prior art. Summary of the invention

[0005] In view of this, the purpose of this application is to provide a method and device for thermal fatigue assessment of safety hook rigging based on artificial intelligence, so as to solve the technical problem existing in the prior art of lack of determination of the thermal fatigue state of the hook rigging.

[0006] In a first aspect, the present application provides a safety hook rigging thermal fatigue assessment method based on artificial intelligence, the method comprising: Obtain current monitoring data for the hook rigging to be evaluated; According to the current monitoring data, a thermal fatigue prediction result of the hook rigging to be evaluated is obtained through a thermal fatigue prediction model; Among them, the thermal fatigue prediction model is obtained by training on historical monitoring data and fatigue test data of sample hook rigging; the historical monitoring data is the status data of the sample hook rigging during actual use; the fatigue test data is the status data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be evaluated.

[0007] In a second aspect, the present application provides an artificial intelligence-based safety hook rigging thermal fatigue assessment device, the device comprising: a data module and a prediction module; The data module is used to obtain current monitoring data of the hook rigging to be evaluated; The prediction module is used to obtain the thermal fatigue prediction result of the hook rigging to be evaluated through a thermal fatigue prediction model according to the current monitoring data; Among them, the thermal fatigue prediction model is obtained by training on historical monitoring data and fatigue test data of sample hook rigging; the historical monitoring data is the status data of the sample hook rigging during actual use; the fatigue test data is the status data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be evaluated.

[0008] Beneficial effects: The present application provides a thermal fatigue assessment method for safety hook rigging based on artificial intelligence, the method comprising: obtaining current monitoring data of the hook rigging to be assessed; obtaining a thermal fatigue prediction result of the hook rigging to be assessed through a thermal fatigue prediction model based on the current monitoring data; wherein the thermal fatigue prediction model is obtained by training with historical monitoring data and fatigue test data on sample hook rigging; the historical monitoring data is status data of the sample hook rigging during actual use; the fatigue test data is status data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be assessed.

[0009] In summary, the present application learns the relationship between the state of the hook rigging and the degree of thermal fatigue based on the trained thermal fatigue prediction model, and subsequently inputs the current monitoring data of the hook rigging to be evaluated into the model to obtain the thermal fatigue prediction result. Therefore, the present application solves the technical problem of the lack of determination of the thermal fatigue state of the hook rigging in the prior art; in addition, since the technical solution provided by the present application can be used to accurately evaluate the thermal fatigue state of the hook rigging, a basis is provided for subsequent maintenance and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. The following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0011] Figure 1 A schematic flow chart of a thermal fatigue assessment method for safety hook rigging based on artificial intelligence provided in an embodiment of the present application; Figure 2 A schematic diagram comparing the noise suppression capabilities of different sample expansion methods provided in the embodiments of the present application; Figure 3 A schematic diagram comparing the data generation capabilities of different sample expansion methods provided in the embodiments of the present application; Figure 4 A schematic diagram comparing the prediction capabilities of different sample expansion methods provided in the embodiments of the present application; Figure 5 A schematic structural diagram of an artificial intelligence-based safety hook rigging thermal fatigue assessment device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution of the present application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0013] First, the embodiment of the present application provides a method for evaluating thermal fatigue of safety hook rigging based on artificial intelligence, such as Figure 1 As shown, Figure 1 A schematic flow chart of a safety hook rigging thermal fatigue assessment method based on artificial intelligence provided in an embodiment of the present application, the method includes: S110~S120, the details of which are as follows: S110: Acquire current monitoring data of the hook rigging to be evaluated during use.

[0014] Specifically, in the embodiment of the present application, the hook rigging to be evaluated refers to the hook rigging that is being used during the construction and maintenance of power plants and mines and needs to be evaluated (ie, predicted) for whether it is thermally fatigued.

[0015] In actual operation, the monitoring data of the hook rigging includes: the rigging monitoring data of the hook rigging to be evaluated in the current working process and the environmental monitoring data of the current working environment of the hook rigging to be evaluated.

[0016] The monitoring data is usually collected through sensors such as temperature sensors. The sensor data can reflect the various status data of the hook rigging to be evaluated at the time of data collection; the sensor for collecting rigging monitoring data is installed on the hook rigging, and the sensor for collecting environmental monitoring data is set in the working environment of the hook rigging.

[0017] Among them, the types of sensors include at least: temperature sensors, force sensors, strain gauges and acceleration sensors, etc. All sensors are connected to the data acquisition terminal for communication, and are used to transmit the status data collected in real time to the background server for storage and processing; among them, the acceleration sensor is used to measure the vibration data of the hook rigging to be evaluated; the rigging monitoring data includes at least: the rigging temperature, stress and deformation data of the sample hook rigging, and the environmental monitoring data includes at least: the load of the sample hook rigging and the ambient temperature of the working environment, etc.

[0018] In actual operation, monitoring data is stored in the database in the form of time series, and the data storage format is a structured file format such as JSON, CSV or HDF5.

[0019] In actual operation, since the training samples of the embodiment of the present application include monitoring data of sample hook rigging under different environmental conditions, and different monitoring data can reflect different degrees of thermal fatigue, the labels of the training samples of the embodiment of the present application may be "no fatigue", "mild fatigue", "moderate fatigue" and "severe fatigue" to indicate the degree of thermal fatigue, etc.; the labeling method is manual labeling based on expert experience, and complies with the labeling specifications of the machine learning training data set.

[0020] In the process of monitoring the hook rigging to be evaluated that is being used during the construction and maintenance of power plants and mines, it is necessary to regularly collect monitoring data according to a preset time period; the preset time period can be determined based on actual needs, and this application does not make any specific restrictions on this.

[0021] S120: obtaining a thermal fatigue prediction result of the hook rigging to be evaluated through a thermal fatigue prediction model according to the current monitoring data; Among them, the thermal fatigue prediction model is obtained by training on the historical monitoring data and fatigue test data of the sample hook rigging; the historical monitoring data is the sample hook rigging in the actual use process, and the fatigue test data is the status data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be evaluated.

[0022] Specifically, the historical monitoring data is the status data collected by sensors in the working environment, that is, the historical monitoring data is the actual status data about the sample hook rigging; the fatigue test data is the status data collected by sensors when performing fatigue testing in a simulated environment, that is, the fatigue test data is the simulated status data about the sample hook rigging, and the fatigue test data can be used to determine the rigging monitoring data of the sample hook rigging under any environmental monitoring data.

[0023] In one implementation, the fatigue test data includes: conventional fatigue test data and accelerated fatigue test data; before S110, the method further includes: steps (1) to (5), the details of which are as follows: Step (1): obtaining a training sample set of sample hook rigging including historical monitoring data, conventional fatigue test data and accelerated fatigue test data; Among them, the conventional fatigue test data is the status data obtained by conducting a conventional fatigue test on the sample hook rigging, and the conventional fatigue test is a test that simulates the actual use process of the sample hook rigging; the accelerated fatigue test data is the status data obtained by conducting an accelerated fatigue test on the sample hook rigging.

[0024] Specifically, both conventional fatigue test data and accelerated fatigue test data are data collected by sensors in a simulated working environment. The difference is that conventional fatigue test data is collected in a simulated working environment that is close to the actual working environment, while accelerated fatigue test data is collected in a simulated working environment that is quite different from the actual working environment. They are used to shorten the test time to directly obtain the status data of the sample hook rigging during long-term working process.

[0025] Step (2): Determine the original feature vector of each original sample in the multiple original samples included in the training sample set.

[0026] Specifically, various device data included in each original sample are converted into original feature vectors in vector form through operations such as encoding.

[0027] Step (3): Construct a new feature vector based on the original feature vector and the linear interpolation formula.

[0028] Specifically, firstly, insufficient training samples may easily lead to poor generalization ability of models such as thermal fatigue prediction models, and at the same time affect the prediction accuracy of models such as thermal fatigue prediction models. Therefore, the embodiment of the present application realizes "sample expansion" by constructing a new feature vector.

[0029] In addition, since the historical monitoring data, conventional fatigue test data and accelerated fatigue test data of the sample hook rigging have different statistical distribution characteristics, such as conventional fatigue test data may be more stable, while the historical monitoring data is greatly affected by the working environment, and the environmental monitoring data included in the accelerated fatigue test data may include extreme cases, in order to avoid the prediction errors that may be caused by the above-mentioned different statistical distribution characteristics, the embodiment of the present application adopts a symmetric polynomial to ensure the nonlinear expression ability of the features, so that the features from different data sources can be integrated in the high-order feature space, thereby improving the diversity of the generated samples; for example, in the historical monitoring data, the real-time fluctuation data of the load-bearing capacity of the hook rigging may contain high-frequency noise, and the symmetric polynomial can effectively extract stable feature patterns to avoid the influence of abnormal data on sample expansion. In the embodiment of the present application, “sample expansion” is achieved by constructing a feature vector in the feature space.

[0030] The linear interpolation formula is as follows: ; ; In the formula, represents the original feature vector; represents the newly added feature vector; express Symmetric polynomial of ; Indicates interpolation coefficients; Indicates interpolation matrix; represents the number of interpolation coefficients and interpolation matrices, The value can be determined according to actual needs; represents a set of vectors selected from multiple original feature vectors; Indicates converting a vector into a diagonal matrix; represents the natural exponential function; represents the L2 norm.

[0031] In actual operation, the newly added feature vectors obtained by "sample expansion" using the linear interpolation formula can not only fill the distribution gaps of existing data, but also optimize the data from specific data sources. For example, accelerated fatigue test data is usually used to predict the wear of hook rigging in long-term use. The distribution of accelerated fatigue test data may deviate from historical monitoring data. In order to improve data consistency, the linear interpolation formula uses a set of vectors randomly selected from the data set and weights them through the L2 norm, so that the generated newly added feature vectors are more in line with actual working conditions. When performing sample expansion, data points close to the limit load can be selected for interpolation, so that the generated newly added feature vectors are more representative. is determined according to the symmetric polynomial formula; the symmetric polynomial formula is as follows: ; ; In the formula, Symmetric polynomial functions; express Middle The weight of the item, express The number of items; express The corresponding A vector of features; express the number of eigenvectors included; express Corresponding to The power of the vector of features, , Indicates the highest order; express Corresponding to The vector of features Power; , , and All are positive integers.

[0032] In the embodiment of the present application, in order to verify the noise suppression capability of the technical solution provided in the embodiment of the present application on the data, the noise suppression capabilities of different sample expansion methods are compared, such as Figure 2 As shown, Figure 2 A schematic diagram comparing the noise suppression capabilities of different sample expansion methods provided in the embodiments of the present application, Figure 2 The horizontal axis is the frequency, Figure 2 The vertical axis is the spectrum amplitude. The high-frequency noise achieved by the traditional interpolation method still exists, and there is an obvious noise signal in the 40Hz-100Hz region. However, the high-frequency noise achieved by the embodiment of the present application is significantly reduced, the curve is smoother and the key features are retained.

[0033] It should be emphasized that Figure 2 The bump near 50 Hz is actually the main frequency component of the signal, that is, the dominant signal feature of the monitoring data. This signal distribution is in line with normal conditions, because the signal generated by the embodiment of the present application is a sine wave (50 Hz) and the embodiment of the present application does not remove the real signal components, but mainly eliminates high-frequency noise.

[0034] is determined according to the first coefficient update formula; the first coefficient update formula is as follows: ; ; ; ; In the formula, Indicates the updated interpolation coefficients; Indicates the interpolation coefficient adjustment amount; Indicates The weights of the features are determined based on the Euclidean distance; express The corresponding The verification index corresponding to the vector of features; Indicates all and The average of the Euclidean distances between represents the regularization parameter; express The corresponding The importance coefficient of the feature vector.

[0035] In actual operation, the optimization difference step of the embodiment of the present application can ensure the rationality of sample expansion. Since there are differences in data density between different sources of sample data, such as historical monitoring data are usually distributed more evenly, while conventional fatigue test data may be densely distributed under specific working conditions, the regularization strategy can adapt to this non-uniform distribution, especially in historical monitoring data. Due to environmental factors, there may be fewer data points under specific conditions. For example, the rigging force data in high humidity or high temperature environments may be less. The regularization strategy can perform additional interpolation on such data to fill in data gaps and improve the adaptability of the model.

[0036] The calculation formula is as follows: ; ; ; In the formula, represents the regularization parameter; express Corresponding to A vector of features; express Corresponding to A vector of features; In the feature space and The Euclidean distance between and denote adjustment factor and threshold, respectively; No. The weights of the features are adaptively assigned based on the importance of the data features; Represents the feature vector The Features Represents the feature vector The Features Indicates The variance of a feature in the entire dataset.

[0037] In actual operation, the quality of generated data is ensured by calculating the L2 norm between data points. For example, in historical monitoring data, there are fewer deformation data of hook rigging under certain extreme loads. The L2 norm constraint can prevent the generated data from deviating from the real physical laws. For example, when interpolating between 20% of the rated load and 80% of the rated load, if the generated data deviates from the actual measured deformation curve, it can be filtered out by setting a threshold to eliminate data points that do not conform to the actual working conditions.

[0038] In one implementation, after constructing the newly added feature vector, it is also necessary to verify the newly added feature vector using a verification formula, and the verification formula is as follows: ; In the formula, Represents the newly added feature vector In actual implementation, if Exceeding the preset threshold , it is considered that the new feature vector If it is unqualified, it needs to be discarded.

[0039] In the embodiment of the present application, in order to verify the superiority of the newly added feature vector generated by the technical solution provided in the embodiment of the present application, the similarity between the newly added feature vector generated by different sample expansion methods and the original feature vector is compared, such as Figure 3 As shown, Figure 3 A schematic diagram comparing the data generation capabilities of different sample expansion methods provided in the embodiments of the present application, Figure 3 The horizontal axis indicates different data augmentation methods (i.e., sample augmentation methods). Figure 3 The ordinate is the L2 norm, Figure 3 The L2 norm is used to measure the error. A lower L2 norm indicates that the newly added feature vector is closer to the original feature vector, thereby improving the credibility of the model. Experimental results show that the technical solution provided in the embodiment of the present application can more accurately generate high-quality newly added feature vectors.

[0040] Step (4): Combine the original feature vector and the newly added feature vector to obtain a feature vector set.

[0041] Specifically, step (3) is repeatedly executed to continuously implement "sample expansion" until the predetermined data quantity requirement is met; in actual execution, when the total number of original feature vectors and newly added feature vectors included in the feature vector set is greater than or equal to 2000, step (3) is stopped.

[0042] In actual operation, in order to verify the influence of different sample expansion methods on the thermal fatigue prediction model and further prove that the sample expansion method provided in the embodiment of the present application can improve the classification accuracy of the model, a comparison of the classification accuracy of rigging fatigue prediction is now carried out. Figure 4 As shown, Figure 4 A schematic diagram showing a comparison of the prediction capabilities of different sample expansion methods provided in the embodiments of the present application. Figure 4 The horizontal axis indicates different data augmentation methods (i.e., sample augmentation methods). Figure 4 The vertical axis is the classification accuracy. The experimental results show that the newly added feature vectors obtained by the sample expansion method provided in the embodiment of the present application are closer to the actual working conditions, so that the thermal fatigue prediction model can better distinguish different fatigue states.

[0043] In actual operation, after sample expansion, it is also necessary to preprocess the expanded feature vectors to provide structured and high-quality input data for the thermal fatigue prediction model; among them, the preprocessing includes at least: data denoising, outlier detection and data normalization operations. Data denoising is used to eliminate environmental interference, outlier detection ensures data integrity, and data normalization optimizes data distribution.

[0044] In actual operation, after preprocessing, key indicators reflecting the thermal fatigue characteristics of hook rigging can be extracted and constructed from the feature data obtained after preprocessing, and high-quality feature vectors can be constructed to improve the classification accuracy of the thermal fatigue prediction model, which can be used to reduce the feature dimension through feature dimensionality reduction technology to prevent overfitting; in addition, an intelligent evaluation model can be constructed for the preprocessed feature data to achieve accurate prediction and risk warning of the thermal fatigue prediction model.

[0045] Step (5): Based on the feature vector set, iteratively train the neural network model to obtain a thermal fatigue prediction model.

[0046] Specifically, during the iterative training process, when the training stop condition is reached, the thermal fatigue prediction model can be obtained.

[0047] In one implementation, step (5) includes: step (5.1) to step (5.2), the details of which are as follows: Step (5.1): According to the feature vector set, reconstruct feature data is determined through a feature dimensionality reduction model; wherein the neural network structure of the feature dimensionality reduction model is an autoencoder; the autoencoder comprises: an encoder, a quantum state network layer and a decoder connected in sequence, the quantum state network layer is used to compress the low-dimensional feature data output by the encoder to obtain compressed feature data, the compressed feature data is used to input into the decoder, and the decoder is made to decode the compressed feature data to obtain reconstructed feature data.

[0048] The autoencoder in the prior art only includes: an encoder and a decoder, the encoder E is used to convert feature data with higher dimensions into feature data with lower dimensions, and the decoder D is used to decode the low-dimensional feature representation to obtain reconstructed feature data; the embodiment of the present application inserts a quantum state network layer between the encoder E and the decoder D of the autoencoder to process and store the quantum state information of the low-dimensional feature data, thereby improving the efficiency and quality of feature compression.

[0049] In practical applications, the feature dimension reduction model is obtained by iterative training of the autoencoder; in actual implementation, when the number of iterative training rounds reaches 1000, the training is stopped.

[0050] In one implementation, during the iterative training of the autoencoder, the network parameters of the encoder E and the decoder D need to be initialized first, as shown below: ; ; In the formula, Represents the initial weight matrix of encoder E and decoder D; Indicates the initial bias. In actual implementation, The value of is set to 0; and They represent the number of neurons in the input layer and output layer of the autoencoder respectively.

[0051] After the network parameters are initialized, the weight matrix and bias of the encoder E are updated according to their corresponding parameter update formulas during the iterative training process. In actual operation, after the network parameters are initialized, different types of data can have a relatively balanced activation distribution when input into the autoencoder, avoiding the vanishing or exploding gradient caused by differences in the numerical ranges of different data sources. For example, the values ​​of conventional fatigue test data may be large, while the historical monitoring data is relatively stable. Initialization helps to balance the data distribution of the above data.

[0052] The parameter update formula corresponding to the weight matrix and bias of encoder E is as follows: ; ;

[0053] ; ; In the formula, Indicated in The weight matrix used by encoder E in this training; Indicated in The weight matrix used by encoder E in this training; Indicated in The bias used by encoder E in this training; Indicated in The bias used by encoder E in this training; Represents the learning rate of the preset encoder E; Indicates that based on The loss function value determined by training the encoder once; Indicated in The first input to the encoder E in the training feature vectors; Indicated in The decoder D outputs the corresponding Reconstructed feature data; Represents the preset regularization coefficient. In actual implementation, The value of is set to 0.3; represents the weight coefficient of the preset quantum fidelity term. In actual implementation, The value of is set to 0.4; Representing quantum state data and reconstructed state data The fidelity between the quantum state data For the In the training, the quantum state network layer obtains the data after quantum encoding the low-dimensional feature data, and reconstructs the state data For the In this training, the quantum state network layer targets Compressed feature data obtained by performing quantum decoding; Represents the Sigmoid activation function; Indicated in In the training, the decoder D The weight matrix used by the network layers, the number of network layers included in the decoder D is ; Indicated in In the training, the decoder D The bias used by each network layer; Indicated in In the training, the decoder D The output data output by each network layer; Represents multiplication between elements; Indicated in The feature importance coefficient used in training, indicating The importance of the features included in In actual operation, historical monitoring data may be affected by environmental noise, resulting in large fluctuations in some data. The embodiment of the present application uses a gradient descent method to optimize encoder parameters to reduce data errors caused by dimensionality reduction.

[0054] In actual operation, since the monitoring data involves sensor time series data (such as stress, tension), material test data (such as fatigue strength), and accelerated test data (such as wear rate under high temperature environment), the Sigmoid activation function can ensure that the data is mapped within a controllable range to avoid network training instability caused by extreme values. For example, the tension test data of the sample hook rigging may have mutation points. The Sigmoid function can smooth the changes of the above data, making the features after data dimensionality reduction more stable.

[0055] The calculation formula is as follows: ; In the formula, Represents the Softmax function; Indicated in In the training, the encoder E The output data output by the network layers (i.e. the low-dimensional feature data mentioned above), the number of network layers included in the encoder E is ; and Respectively expressed in In the training, the encoder E The weight matrices and biases used by each network layer.

[0056] In one implementation, the quantum state network layer The process of processing the low-dimensional feature data output by the encoder E in the training includes the following steps: 1) The quantum encoding function in the quantum state network layer encodes the low-dimensional feature data into the quantum bit to obtain the quantum state data ;The quantum encoding function is as follows: ; In the formula, represents the quantum coding function; Represents a superposition of quantum states; Represents the number of qubits in the quantum state network layer; Indicates In the first training, the encoder E Low-dimensional feature data output by the network layer The elements; represents the quantum encoding unit gate applied to each low-dimensional feature data. In actual implementation, It is realized by combining Hadamard gate and Pauli gate; Among them, in the process of encoder E encoding the feature vector, the process of processing the input data in each network layer is as follows:

[0057] In the formula, Indicated in In the training, the encoder E The output data output by each network layer; Indicated in In the training, the encoder E The output data output by the network layer, ; and Respectively expressed in In the training, the encoder E The weight matrices and biases used by each network layer.

[0058] In actual operation, since the monitoring data includes multimodal information (such as multidimensional data from different sensors), conventional dimensionality reduction methods may lose the key correlation of some data. For example, in conventional fatigue test data, there may be a complex nonlinear relationship between the stress deformation data of metal materials and the ambient temperature and humidity. Quantum state multiplexing can use the quantum superposition characteristics to enable multiple characteristic information to be encoded into the quantum state at the same time, thereby retaining key information during the compression process.

[0059] 2) The quantum decoding function in the quantum state network layer converts the quantum state data Decode to obtain compressed feature data; the quantum decoding function is as follows: = ; In the formula, represents the quantum decoding function; Indicated in The input data of the first network layer in the decoder D in the training, that is, the compressed feature data; Indicated in Reconstructed state data in training Middle qubits of data; Represents the unit gate in the decoder D, which is Set accordingly. Used to convert reconstructed data into compressed feature data.

[0060] In actual operation, since there is inevitable information loss in the dimensionality reduction process of monitoring data, quantum decoding recovers data through reverse quantum gate operations and realizes data conversion in combination with measurement operations. For real-time monitoring data on site, such as vibration signals and stress signals, the decoding stage further screens out the most representative features through feature importance calculation, thereby reducing redundant features and improving data reconstruction quality.

[0061] Step (5.2): Based on the reconstructed feature data, iterative training is performed on the neural network model to obtain a thermal fatigue prediction model.

[0062] In one implementation, the neural network structure of the thermal fatigue prediction model is an extreme learning machine, and the network parameters updated during the iterative training process of the extreme learning machine include: the number of neurons included in the hidden layer and the weight matrix of the output layer.

[0063] In practical applications, the input data used to train the extreme learning machine is the reconstructed feature data output by the trained feature dimensionality reduction model.

[0064] In the process of iterative training for the extreme learning machine, it is necessary to first initialize the number of neurons included in the hidden layer of the extreme learning machine and the weight matrix of the output layer (not shown below), as shown below: ; In the formula, Represents the number of neurons included in the hidden layer of the extreme learning machine; Represents the minimum value function; Represents the dimension of the reconstructed feature data.

[0065] After the number of neurons included in the hidden layer of the extreme learning machine and the weight matrix of the output layer are initialized, the number of neurons included in the hidden layer of the extreme learning machine is updated according to the first parameter update formula during the training process; the weight matrix of the output layer of the extreme learning machine is updated according to the second parameter update formula.

[0066] In actual operation, the dimension of the input data directly affects the initial number of neurons in the hidden layer, and the prediction of the thermal fatigue prediction model is usually affected by multiple factors including stress, temperature and material properties. Therefore, the key fatigue influencing factors will be retained during feature dimensionality reduction to ensure that the extreme learning machine can effectively learn fatigue damage patterns.

[0067] The first parameter update formula is as follows: ; In the formula, Indicated in The number of neurons included in the hidden layer of the extreme learning machine during training; Indicated in The number of neurons included in the hidden layer of the extreme learning machine during training; Indicates the number of neurons included in the hidden layer of the preset extreme learning machine that can be increased during each training process. In actual implementation, The value of is set to 5, that is, if the prediction accuracy of the extreme learning machine does not reach the target accuracy, 5 neurons will be automatically added to the hidden layer of the extreme learning machine until the prediction accuracy of the extreme learning machine tends to be stable or reaches the quantity threshold. The quantity threshold can be set to 50; Indicates that based on The prediction accuracy of the extreme learning machine determined by the training, represents the target accuracy of the extreme learning machine; represents the indicator function. When the prediction accuracy of the extreme learning machine does not reach the target accuracy, its value is 1, otherwise, its value is 0; In actual operation, if the prediction accuracy of the thermal fatigue prediction model is low, additional nodes can be added to improve the classification ability of the thermal fatigue prediction model for this category, which helps to avoid a few categories being ignored by the model and improve the overall classification performance.

[0068] The second parameter update formula is as follows: ; In the formula, Indicated in The weight matrix of the output layer of the extreme learning machine in this training; represents the output matrix of the hidden layer of the extreme learning machine; represents the transpose of the output matrix of the hidden layer of the extreme learning machine; represents the target output matrix; Represents the preset regularization parameter. In actual implementation, The value of is set to 0.3; Represents the identity matrix.

[0069] The loss function used by the thermal fatigue prediction model during iterative training is as follows: ; In the formula, Indicates that based on The loss function value of the extreme learning machine determined by the training; Represents the preset regularization parameter. In actual implementation, The value of is set to 0.3; Indicated in The weight matrix of the output layer of the extreme learning machine in this training.

[0070] In actual implementation, if based on The loss function value of the extreme learning machine determined by the training is less than the preset loss function threshold. Keep it, indicating that the training of the output layer of the extreme learning machine is over, otherwise proceed to the training session.

[0071] In actual operation, the embodiment of the present application uses a quantum fidelity term to measure the degree of information loss in the dimensionality reduction process. Since accelerated fatigue test data often contain key failure modes, if dimensionality reduction leads to the loss of important modes, it will affect the accuracy of equipment life prediction. By utilizing the quantum fidelity term, the similarity between the reconstructed data and the original data can be improved during the training process. For example, in the dimensionality reduction task of accelerated fatigue test data, this optimization strategy can ensure that the main change trend of the material under different stress states is still retained after data compression, thereby improving the accuracy of fault diagnosis.

[0072] In actual operation, in actual application, if the label of the thermal fatigue prediction model is based on 0-1 binary classification (low fatigue vs. high fatigue), then the training process of the thermal fatigue prediction model can adopt a smaller regularization parameter (such as 0.1-0.3) to ensure the clarity of the classification boundary. If it is a multi-level fatigue classification (such as low, medium, and high fatigue damage), it is necessary to appropriately increase the regularization parameter (such as 0.5) to enhance the generalization ability of the thermal fatigue prediction model and reduce overfitting.

[0073] In one implementation, since the dimension reduction data used for training the extreme learning machine has been processed by the feature dimension reduction model, before using the dimension reduction data for training the extreme learning machine, it is necessary to supplement the dimension reduction data with "technical features"; the supplement of the "technical features" is determined according to the feature construction formula; The feature construction formula is as follows: ; ; In the formula, A vector representing the reconstructed feature data; Indicates the number of cells to be added to each The eigenvector of It means feature extraction based on finite element analysis; represents the feature vector merging operation; , and It represents the result of dimensionality reduction based on different dimensional benchmarks based on principal component analysis. represents the first finite element vector, represents the second finite element vector, represents the third finite element vector.

[0074] In actual operation, finite element analysis can simulate the stress distribution and deformation of rigging in long-term work, and extract the most representative fatigue characteristics after dimensionality reduction through principal component analysis. For example, the residual stress distribution of hook rigging under different loads can be generated through finite element analysis and added to the extreme learning machine model as an additional feature, so that the thermal fatigue prediction model can more accurately distinguish between high fatigue and low fatigue categories.

[0075] In actual operation, the performance of the thermal fatigue prediction model can also be verified and continuously improved to ensure that the model has high accuracy and good generalization ability in practical applications; among them, performance verification and continuous improvement include cross-validation, performance index calculation, error analysis and online update mechanism, among which cross-validation is used to evaluate the stability of the thermal fatigue prediction model, performance index calculation provides a quantitative evaluation basis, error analysis assists in locating model deficiencies, and the online update mechanism enables the thermal fatigue prediction model to dynamically adjust to new data; in addition, the evaluation results and model prediction results of the thermal fatigue prediction model can be presented to decision makers and maintenance personnel in an intuitive way to support fire safety management and preventive maintenance.

[0076] In summary, the technical solution in the embodiment of the present application has the following advantages compared with the prior art: 1) Using a multilinear interpolation algorithm embedded with symmetric polynomials, the physical, monitoring and simulation data of the rigging are fully utilized to achieve high-quality sample expansion, improve the generalization ability of the thermal fatigue prediction model, and reduce the overfitting problem of the thermal fatigue prediction model; 2) An autoencoding neural network based on quantum state reuse is adopted, and the quantum state reuse layer is used to reduce information loss and improve the compression efficiency and reconstruction quality of the thermal fatigue characteristics of the rigging.

[0077] 3) The extreme learning machine algorithm based on dynamic hidden layer node allocation is adopted to dynamically adjust the structure of the classification model according to the complexity of the state data, improve the classification accuracy, and adapt it to the fatigue assessment needs under different working conditions.

[0078] Second, the present application provides an artificial intelligence-based safety hook rigging thermal fatigue assessment device, such as Figure 5 As shown, Figure 5 A schematic diagram of the structure of a safety hook rigging thermal fatigue assessment device based on artificial intelligence provided in an embodiment of the present application, the device includes: a data module 310 and a prediction module 320; The data module 310 is used to obtain the current monitoring data of the hook rigging to be evaluated; A prediction module 320, for obtaining a thermal fatigue prediction result of the hook rigging to be evaluated through a thermal fatigue prediction model according to current monitoring data; Among them, the thermal fatigue prediction model is obtained by training on the historical monitoring data and fatigue test data of the sample hook rigging; the historical monitoring data is the status data of the sample hook rigging during actual use; the fatigue test data is the status data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be evaluated.

[0079] In one implementation, the fatigue test data includes: conventional fatigue test data and accelerated fatigue test data; the prediction module 320 is further used to obtain a training sample set including historical monitoring data, conventional fatigue test data and accelerated fatigue test data about the sample hook rigging; Among them, the conventional fatigue test data is the status data obtained by performing conventional fatigue tests on the sample hook rigging, and the conventional fatigue test is a test process that simulates the actual use process of the sample hook rigging; the accelerated fatigue test data is the status data obtained by performing accelerated fatigue tests on the sample hook rigging; The prediction module 320 is further used to determine the original feature vector of each original sample in the plurality of original samples included in the training sample set; The prediction module 320 is further used to construct a new feature vector according to the original feature vector and the linear interpolation formula; The prediction module 320 is further used to combine the original feature vector and the newly added feature vector to obtain a feature vector set; The prediction module 320 is further used to train a thermal fatigue prediction model according to the feature vector set.

[0080] In one implementation, the linear difference formula is as follows: ; ; In the formula, represents the original feature vector; represents the newly added feature vector; express Symmetric polynomial of ; Indicates interpolation coefficients; Indicates interpolation matrix; Represents the number of interpolation coefficients and interpolation matrices; represents a set of vectors selected from multiple original feature vectors; Indicates converting a vector into a diagonal matrix; represents the natural exponential function; represents the L2 norm.

[0081] In one implementation, is determined according to the symmetric polynomial formula; the symmetric polynomial formula is as follows: ; ; In the formula, Symmetric polynomial functions; express Middle The weight of the item, express The number of items; express The corresponding A vector of features; express the number of eigenvectors included; express Corresponding to The power of the vector of features, , Indicates the highest order; express Corresponding to The vector of features Power.

[0082] In one implementation, the prediction module 320 is further used to determine the reconstructed feature data through the feature dimension reduction model according to the feature vector set; Among them, the neural network structure of the feature dimensionality reduction model is an autoencoder; the autoencoder includes: an encoder, a quantum state network layer and a decoder connected in sequence, the quantum state network layer is used to compress the low-dimensional feature data output by the encoder to obtain compressed feature data, and the compressed feature data is used to input into the decoder, so that the decoder decodes the compressed feature data to obtain reconstructed feature data; The prediction module 320 is further used to perform iterative training on the neural network model according to the reconstructed feature data to obtain a thermal fatigue prediction model.

[0083] In one implementation, the weight matrix and bias of the encoder are updated according to their corresponding parameter update formulas during the iterative training process. The parameter update formulas corresponding to the weight matrix and bias of the encoder are as follows: ; ; ; ; In the formula, Indicated in The weight matrix used by the encoder in this training; Indicated in The weight matrix used by the encoder in this training; Indicated in The bias used by the encoder in this training; Indicated in The bias used by the encoder in this training; Indicates the preset encoder learning rate; Indicates that based on The loss function value determined by training the encoder once; Indicated in The first input to the encoder in the training feature vectors; Indicated in The decoder output in the training corresponds to Reconstructed feature data; Represents the preset regularization coefficient; Represents the weight coefficient of the preset quantum fidelity term; Representing quantum state data and reconstructed state data The fidelity between the quantum state data For the In the training, the quantum state network layer obtains the data after quantum encoding the low-dimensional feature data, and reconstructs the state data For the In this training, the quantum state network layer targets The compressed feature data obtained by quantum decoding, is the sample size, represents the F-norm; Represents the Sigmoid activation function; Indicated in In the decoder of the training The weight matrix used by the network layers, the number of network layers included in the decoder is ; Indicated in In the decoder of the training The bias used by each network layer; Indicated in In the decoder of the training The output data output by each network layer; Represents multiplication between elements; Indicated in The feature importance coefficient used in training, indicating The importance of the features contained in .

[0084] In one implementation, the neural network structure of the thermal fatigue prediction model is an extreme learning machine; the number of neurons included in the hidden layer of the extreme learning machine is updated according to a first parameter update formula during the iterative training process, and the first parameter update formula is as follows: ; In the formula, Indicated in The number of neurons included in the hidden layer of the extreme learning machine during training; Indicated in The number of neurons included in the hidden layer of the extreme learning machine during training; Indicates the number of neurons included in the hidden layer of the preset extreme learning machine that can be increased during each training process; Indicates that based on The prediction accuracy of the extreme learning machine determined by the training; represents the target accuracy of the extreme learning machine; Represents the indicator function, which has a value of 1 when the prediction accuracy of the extreme learning machine does not reach the target accuracy, otherwise it has a value of 0.

[0085] In one implementation, the loss function used by the thermal fatigue prediction model during iterative training is as follows: ； In the formula, Indicates that based on The loss function value of the extreme learning machine determined by the training; represents the preset regularization parameter; Indicated in The weight matrix of the output layer of the extreme learning machine in this training.

[0086] Third, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, steps S110 to S120 provided in the above embodiment are implemented.

[0087] Fourth, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, steps S110 to S120 of the above embodiment are executed.

[0088] Fifth, the computer program product provided in the present application includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method in the previous method embodiment. The specific implementation can refer to steps S110~S120 of the method embodiment, which will not be repeated here.

[0089] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0090] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0091] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0092] It should be noted that if the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0093] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0094] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A thermal fatigue assessment method for safety hook rigging based on artificial intelligence, characterized in that: The method comprises: Obtain current monitoring data for the hook rigging to be evaluated; According to the current monitoring data, a thermal fatigue prediction result of the hook rigging to be evaluated is obtained through a thermal fatigue prediction model; Among them, the thermal fatigue prediction model is obtained by training on historical monitoring data and fatigue test data of sample hook rigging; the historical monitoring data is the status data of the sample hook rigging during actual use; the fatigue test data is the status data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be evaluated.

2. The method according to claim 1, characterized in that The fatigue test data includes: conventional fatigue test data and accelerated fatigue test data; before obtaining the current monitoring data of the hook rigging to be evaluated, the method further includes: Acquire a training sample set including the historical monitoring data, the conventional fatigue test data and the accelerated fatigue test data about the sample hook rigging; The conventional fatigue test data is the state data obtained by performing a conventional fatigue test on the sample hook rigging, and the conventional fatigue test is a test process that simulates the actual use process of the sample hook rigging; the accelerated fatigue test data is the state data obtained by performing an accelerated fatigue test on the sample hook rigging; Determining an original feature vector of each of the original samples in the plurality of original samples included in the training sample set; Constructing a new feature vector according to the original feature vector and the linear interpolation formula; Combining the original feature vector and the newly added feature vector to obtain a feature vector set; The thermal fatigue prediction model is trained according to the feature vector set.

3. The method according to claim 2, characterized in that The linear difference formula is as follows: ； ； In the formula, represents the original feature vector; represents the newly added feature vector; express Symmetric polynomial of ; Indicates interpolation coefficients; Indicates interpolation matrix; Represents the number of interpolation coefficients and interpolation matrices; represents a vector set selected from the plurality of original feature vectors; Indicates converting a vector into a diagonal matrix; represents the natural exponential function; represents the L2 norm.

4. The method according to claim 3, characterized in that: Said is determined according to a symmetric polynomial formula; the symmetric polynomial formula is as follows: ； ； In the formula, Symmetric polynomial functions; express Middle The weight of the item, express The number of items; express The corresponding A vector of features; express the number of eigenvectors included; express Corresponding to The power of the vector of features, , Indicates the highest order; express Corresponding to The vector of features Power.

5. The method according to claim 2, characterized in that: The step of training the thermal fatigue prediction model according to the feature vector set includes: Determining, according to the feature vector set, reconstructing feature data through a feature dimension reduction model; The neural network structure of the feature dimensionality reduction model is an autoencoder; the autoencoder comprises: an encoder, a quantum state network layer and a decoder connected in sequence, the quantum state network layer is used to compress the low-dimensional feature data output by the encoder to obtain compressed feature data, the compressed feature data is used to input into the decoder, and the decoder is made to decode the compressed feature data to obtain reconstructed feature data; According to the reconstructed characteristic data, an iterative training is performed on a neural network model to obtain the thermal fatigue prediction model.

6. The method according to claim 5, characterized in that The weight matrix and bias of the encoder are updated according to their corresponding parameter update formulas during the iterative training process. The parameter update formulas corresponding to the weight matrix and bias of the encoder are as follows: ； ； ； ； In the formula, Indicated in The weight matrix used by the encoder in this training; Indicated in The weight matrix used by the encoder in this training; Indicated in The bias used by the encoder in this training; Indicated in The bias used by the encoder in this training; Indicates the preset encoder learning rate; Indicates that based on The loss function value determined by training the encoder once; Indicated in The first input to the encoder in the training feature vectors; Indicated in The decoder output in the training corresponds to Reconstructed feature data; Represents the preset regularization coefficient; Represents the weight coefficient of the preset quantum fidelity term; Representing quantum state data and reconstructed state data The fidelity between the quantum state data For the In the training, the quantum state network layer obtains the data after quantum encoding the low-dimensional feature data, and reconstructs the state data For the In this training, the quantum state network layer targets The compressed feature data obtained by quantum decoding, is the sample size, represents the F-norm; Represents the Sigmoid activation function; Indicated in In the decoder of the training The weight matrix used by the network layers, the number of network layers included in the decoder is ; Indicated in In the decoder of the training The bias used by each network layer; Indicated in In the decoder of the training The output data output by each network layer; Represents multiplication between elements; Indicated in The feature importance coefficient used in training, indicating The importance of the features contained in .

7. The method according to claim 2, characterized in that The neural network structure of the thermal fatigue prediction model is an extreme learning machine; the number of neurons included in the hidden layer of the extreme learning machine is updated according to the first parameter update formula during the iterative training process, and the first parameter update formula is as follows: ； In the formula, Indicated in The number of neurons included in the hidden layer of the extreme learning machine during training; Indicated in The number of neurons included in the hidden layer of the extreme learning machine during training; Indicates the number of neurons included in the hidden layer of the preset extreme learning machine that can be increased during each training process; Indicates that based on The prediction accuracy of the extreme learning machine determined by the training; represents the target accuracy of the extreme learning machine; Represents the indicator function, which has a value of 1 when the prediction accuracy of the extreme learning machine does not reach the target accuracy, otherwise it has a value of 0.

8. The method according to claim 7, characterized in that The loss function used by the thermal fatigue prediction model in the iterative training process is as follows: ； In the formula, Indicates that based on The loss function value of the extreme learning machine determined by the training; represents the preset regularization parameter; Indicated in The weight matrix of the output layer of the extreme learning machine in this training.

9. An artificial intelligence-based safety hook rigging thermal fatigue assessment device, characterized in that: The device comprises: a data module and a prediction module; The data module is used to obtain current monitoring data of the hook rigging to be evaluated; The prediction module is used to obtain the thermal fatigue prediction result of the hook rigging to be evaluated through a thermal fatigue prediction model according to the current monitoring data; Among them, the thermal fatigue prediction model is obtained by training on historical monitoring data and fatigue test data of sample hook rigging; the historical monitoring data is the status data of the sample hook rigging during actual use; the fatigue test data is the status data obtained by fatigue testing the sample hook rigging; the thermal fatigue prediction result indicates the degree of thermal fatigue of the hook rigging to be evaluated.

10. The device according to claim 9, characterized in that The fatigue test data includes: conventional fatigue test data and accelerated fatigue test data; The prediction module is further used to obtain a training sample set including the historical monitoring data, the conventional fatigue test data and the accelerated fatigue test data about the sample hook rigging; The conventional fatigue test data is the state data obtained by performing a conventional fatigue test on the sample hook rigging, and the conventional fatigue test is a test process that simulates the actual use process of the sample hook rigging; the accelerated fatigue test data is the state data obtained by performing an accelerated fatigue test on the sample hook rigging; The prediction module is further used to determine the original feature vector of each of the multiple original samples included in the training sample set; The prediction module is further used to construct a newly added feature vector according to the original feature vector and a linear interpolation formula; The prediction module is further used to combine the original feature vector and the newly added feature vector to obtain a feature vector set; The prediction module is further used to train the thermal fatigue prediction model according to the feature vector set.

Images