Data Processing Method and Device for Predicting the Lubricating Oil Change Cycle of a Vehicle

Abstract

The present invention provides a data processing method and device for predicting the replacement cycle of vehicle lubricating oil, which relates to the technical field of data processing. After processing the vehicle maintenance records and vehicle monitoring parameters through a feature extraction model and a feature dimensionality reduction model, a high-order neural network is used as a classification algorithm to determine the replacement attribute of the lubricating oil of the target vehicle, so as to predict the lubricating oil replacement cycle. Among them, the feature extraction model dynamically adjusts the key parameters in the search strategy by fusing historical iteration information, determines the number of network nodes and layer weights of the feature dimensionality reduction model based on the dynamic topology structure adjustment mechanism, and the weight gradient of the high-order neural network is adjusted by using a high-order training mechanism with gradient penalty based on the mutual information between the input and the output. It can enhance the recognition ability of key features, improve the classification accuracy, and better process unbalanced data to enhance the classification effect.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a data processing method and device for predicting the replacement cycle of vehicle lubricating oil. Background Art

[0002] In existing vehicle maintenance practices, the replacement of lubricating oil is usually carried out according to the standard cycle recommended by the manufacturer or based on the experience of vehicle operators. This method ignores the actual impact of vehicle operating conditions on the performance of lubricating oil, which may result in the replacement of lubricating oil before it reaches its optimal service life or failure to replace it in a timely manner when it should be replaced, thus affecting vehicle performance and increasing maintenance costs. If the timing of lubricating oil replacement is not properly grasped, it will not only reduce vehicle performance but also may cause problems such as accelerated component wear. With the development of data mining technology and machine learning, it has become possible to use vehicle operation data to predict the optimal replacement time of lubricating oil. This method can more scientifically evaluate the usage status and replacement requirements of lubricating oil and achieve personalized maintenance based on the actual vehicle condition.

[0003] However, vehicle operation data such as engine speed and vehicle speed are basically collected through sensors, which have the characteristics of high dimensionality, time series, heterogeneity, noise and uncertainty. In addition, the relationship between vehicle operating conditions and the performance of lubricating oil is highly non-linear, and simple linear models or traditional algorithms may be difficult to capture these complex associations. Moreover, vehicle maintenance decisions require immediate feedback, which means that data analysis algorithms need to have the ability to efficiently process a large amount of real-time data.

[0004] However, the existing technology lacks in processing speed and timeliness. When dealing with large-scale vehicle operation data, it is often difficult to extract the key information affecting the performance of lubricating oil through effective feature extraction means, reducing the prediction reliability of the model. Traditional classification algorithms are difficult to accurately capture the core features of highly non-linear data when dealing with complex lubricating oil state prediction tasks, reducing the classification effect and practical value. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a data processing method and device for predicting the replacement cycle of vehicle lubricating oil, which can reduce the data dimension, improve the data processing speed, and enhance the ability to identify key features and improve the classification accuracy.

[0006] First aspect, an embodiment of the present invention provides a data processing method for predicting the lubricating oil replacement cycle of a vehicle. The method includes: obtaining the maintenance records and vehicle monitoring parameters of the target vehicle, and assembling the maintenance records and vehicle monitoring parameters into vector data; inputting the vector data into a pre-constructed feature extraction model to output the key features of the vector data; wherein, the feature extraction model updates the neural network parameters using a preset search strategy, and dynamically adjusts the key parameters in the preset search strategy by fusing historical iteration information; performing dimensionality reduction processing on the key parameters through a pre-constructed feature dimensionality reduction model; wherein, the number of network nodes and layer weights of the feature dimensionality reduction model are determined based on a dynamic topology adjustment mechanism; using a high-order neural network as a classification algorithm to classify the dimensionality-reduced key features and output a classification result; wherein, the classification result is used to characterize the replacement attribute of the lubricating oil of the target vehicle; the weight gradient of the high-order neural network is adjusted using a high-order training mechanism with gradient penalty based on the mutual information between the input and output; predicting the lubricating oil replacement cycle of the target vehicle based on the classification result.

[0007] In combination with the first aspect, an embodiment of the present invention further provides a first implementation manner of the first aspect. The classification result includes normal replacement and early replacement. The step of predicting the lubricating oil replacement cycle of the target vehicle based on the classification result includes: determining the classification result as the prediction result of the lubricating oil replacement cycle of the target vehicle.

[0008] In combination with the first aspect, an embodiment of the present invention further provides a second implementation manner of the first aspect. The step of adjusting the weight gradient of the high-order neural network using a high-order training mechanism with gradient penalty based on the mutual information between the input and output includes: obtaining the mutual information between the input and output of the high-order neural network, and solving the weight gradient of the high-order neural network using the chain rule; updating the weight gradient based on the mutual information and a preset gradient penalty term.

[0009] In combination with the first aspect, an embodiment of the present invention further provides a third implementation manner of the first aspect. The method further includes: updating the model parameters of the high-order neural network using a gradient-based method.

[0010] In combination with the first aspect, an embodiment of the present invention further provides a fourth implementation manner of the first aspect. The step of the feature extraction model updating the neural network parameters using a preset search strategy and dynamically adjusting the key parameters in the preset search strategy by fusing historical iteration information includes: using a preset exponential decay term as the weight to calculate the parameter adjustment factor corresponding to the loss function value of the feature extraction model; adjusting the search step of the search strategy based on the parameter adjustment factor and the iteration situation of the preset search strategy to adjust the key parameters in the search strategy.

[0011] In combination with the first aspect, an embodiment of the present invention further provides a fifth implementation manner of the first aspect, wherein the search strategy includes a grey wolf optimization algorithm; wherein the key parameters of the grey wolf optimization algorithm include a global search parameter and a local search parameter.

[0012] In combination with the first aspect, an embodiment of the present invention further provides a sixth implementation manner of the first aspect, wherein the feature dimensionality reduction model includes an autoencoder; the steps of determining the number of network nodes and layer weights of the feature dimensionality reduction model based on the dynamic topology structure adjustment mechanism include: training the autoencoder with a preset training sample set, and evaluating the importance of each feature in the training sample set during the decoding process through a sensitivity scoring function; adjusting the weight importance of each layer of the autoencoder based on the importance; using an approximation method to calculate the gradient of the loss function of the autoencoder with respect to the number of network nodes; and adjusting the number of network nodes of the autoencoder based on the gradient, as well as the training progress and reconstruction error of the autoencoder.

[0013] In combination with the first aspect, an embodiment of the present invention further provides a seventh implementation manner of the first aspect, wherein the method further includes: obtaining a pre-constructed initial sample set; performing feature distillation processing on the initial sample set using the attention mechanism of deep learning to extract the key feature vectors in the initial sample set; and constructing a training sample set by data augmentation based on the key feature vectors.

[0014] In combination with the first aspect, an embodiment of the present invention further provides an eighth implementation manner of the first aspect, wherein data augmentation is performed through a generative adversarial network, and the total loss function of the generative adversarial network includes a generative adversarial loss, a feature matching loss, and a quantum state fidelity loss; the method further includes: performing a quantum gate operation on the parameters of the generative adversarial network based on preset Euler angle parameters to initialize the parameters of the generative adversarial network.

[0015] Second aspect, an embodiment of the present invention provides a data processing device for predicting the replacement cycle of vehicle lubricating oil. The device includes: a data acquisition module, configured to acquire the maintenance records and vehicle monitoring parameters of a target vehicle, and assemble the maintenance records and vehicle monitoring parameters into vector data; a feature extraction module, configured to input the vector data into a pre-constructed feature extraction model, and output the key features of the vector data; wherein, the feature extraction model updates the neural network parameters by using a preset search strategy, and dynamically adjusts the key parameters in the preset search strategy by fusing historical iteration information; a feature dimensionality reduction module, configured to perform dimensionality reduction processing on the key features through a pre-constructed feature dimensionality reduction model; wherein, the number of network nodes and layer weights of the feature dimensionality reduction model are determined based on a dynamic topology adjustment mechanism; an execution module, configured to use a high-order neural network as a classification algorithm to classify the dimensionality-reduced key features, and output a classification result; wherein, the classification result is used to characterize the replacement attribute of the lubricating oil of the target vehicle; the weight gradient of the high-order neural network is adjusted by using a high-order training mechanism with gradient penalty based on the mutual information between the input and the output; an output module, configured to predict the replacement cycle of the lubricating oil of the target vehicle based on the classification result.

[0016] The embodiments of the present invention bring the following beneficial effects: A data processing method and device for predicting the replacement cycle of vehicle lubricating oil provided by the embodiments of the present invention can reduce the data dimension, improve the data processing speed, and can enhance the recognition ability of key features, improve the classification accuracy, enhance the model stability, improve the generalization ability, and better process unbalanced data, thereby enhancing the classification effect and practical value.

[0017] Other features and advantages of the present invention will be described in the following specification, and part of them will become obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification, claims and drawings.

[0018] To make the above objectives, features and advantages of the present invention more obvious and understandable, the following specific preferred embodiments are given, and in conjunction with the accompanying drawings, the detailed description is as follows. Description of the Drawings

[0019] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1Flow chart of a data processing method for predicting the replacement cycle of vehicle lubricating oil provided by an embodiment of the present invention;

[0021] Figure 2 Flow chart of another data processing method for predicting the replacement cycle of vehicle lubricating oil provided by an embodiment of the present invention;

[0022] Figure 3 Flow chart of a method for constructing a training sample set provided by an embodiment of the present invention;

[0023] Figure 4 Schematic structural diagram of a data processing device for predicting the replacement cycle of vehicle lubricating oil provided by an embodiment of the present invention;

[0024] Figure 5 Schematic structural diagram of another data processing device for predicting the replacement cycle of vehicle lubricating oil provided by an embodiment of the present invention;

[0025] Figure 6 Schematic structural diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] An embodiment of the present invention provides a data processing method and device for predicting the replacement cycle of vehicle lubricating oil, which can reduce the data dimension, improve the data processing speed, and enhance the recognition ability of key features and improve the classification accuracy.

[0028] For ease of understanding, first, a data processing method for predicting the replacement cycle of vehicle lubricating oil provided by an embodiment of the present invention will be described. Among them, Figure 1 shows a flow chart of a data processing method for predicting the replacement cycle of vehicle lubricating oil provided by an embodiment of the present invention, as Figure 1 shown. The method includes the following steps:

[0029] Step S102, obtain the maintenance records and vehicle monitoring operation parameters of the target vehicle, and assemble the maintenance records and vehicle operation monitoring parameters into vector data.

[0030] First, in the embodiments of the present invention, by collecting relevant data of the target vehicle, assembling the data into data that can be recognized by the model, and then using the pre-constructed model for data processing, the lubricating oil replacement cycle can be evaluated and predicted. The following method can be used for assembly: Vectorization of maintenance records Assume that each maintenance record contains the following information: the time (date) of lubricating oil replacement, the type of lubricating oil used, and the remarks of the maintenance technician. Convert this information into numerical features. For example: Time: It can be converted into the number of days since the last lubricating oil replacement. Lubricating oil type: It can be represented using one-hot encoding. Remarks of the maintenance technician: Text processing can be performed to extract keywords and convert them into numerical features. In specific implementation, the data collection sources of the present invention include the vehicle's internal sensors and maintenance records, and the collected data is stored in the JSON format. The maintenance records include the time of each lubricating oil replacement, the type of lubricating oil used, the remarks of the maintenance technician, etc. Vehicle operation monitoring parameters: including driving mileage, engine working time, oil temperature, oil pressure, engine speed, vehicle load, etc. In one embodiment, the attributes of the data include engine speed (RPM) represented as R a1 、vehicle speed (km / h) represented as R a2 、oil temperature (°C) represented as R a3 、driving mileage (km) represented as R a4 、fuel consumption (L / 100km) represented as R a5 、engine load (%) represented as R a6 、lubricating oil viscosity (cP) represented as R a7 、lubricating oil acid value (mgKOH / g) represented as R a8 、particle count (per 100ml) represented as R a9 、moisture content (%) represented as R a10 .

[0031] It should be noted that this embodiment is only to illustrate a data format and type of the present invention. In actual applications, the attributes of the data are usually more than 10, and the number of data attributes may reach dozens or even hundreds. Further, the collected data is also preprocessed, including: Cleaning: Removing missing values, outliers, and duplicate records. Standardization: Standardizing data with different units and magnitudes to make them comparable.

[0032] Step S104: Input the vector data into the pre-constructed feature extraction model to output the key features of the vector data.

[0033] Step S106: Perform dimensionality reduction processing on the key parameters through the pre-constructed feature dimensionality reduction model.

[0034] Feature extraction is to extract the most useful information from the original data to reduce the data dimension and improve the performance of the model. Feature dimensionality reduction is to further reduce the number of features on the basis of feature extraction to reduce the computational complexity and improve the generalization ability of the model. Existing feature extraction techniques usually rely on fixed parameter settings, which may perform poorly when dealing with complex or changing data. Even when parameter search strategies are involved to determine their network parameters, the historical iteration impact of the training process is not considered in the parameter tuning, resulting in poor feature extraction ability of the model to extract key information from a large amount of data, making the model unable to accurately reflect the actual situation of the data and affecting the reliability of the classification results. Correspondingly, the present invention introduces an adaptive feature extraction method, updates the neural network parameters by using a preset search strategy, and dynamically adjusts the key parameters in the search strategy by fusing historical iteration information. This method combines a parameter optimization strategy, historical iteration information, and an adaptive adjustment mechanism, which helps to improve the accuracy and reliability of the model and can significantly enhance the adaptability and performance of the model.

[0035] Furthermore, the feature dimensionality reduction processing of the prior art (such as PCA, LDA, etc.) projects high-dimensional data into a low-dimensional space through linear or non-linear mapping. In this process, some feature information important for the task may be lost, resulting in the failure to retain key information while reducing the data dimension, causing information loss and affecting the execution efficiency and accuracy of the model. Moreover, many existing dimensionality reduction models adopt fixed network structures and parameter settings and cannot be dynamically adjusted according to the changes in the data, thus not being able to well adapt to different data distributions. If the dimensionality reduction model is too simple, it may not be able to capture the complex patterns in the data; if the model is too complex, it may lead to overfitting, that is, performing well on the training data but poorly on the unseen data.

[0036] In response to this, the feature dimensionality reduction model proposed in the embodiment of the present invention solves the above problems by introducing a dynamic topology structure adjustment mechanism, wherein the number of network nodes and layer weights of the feature dimensionality reduction model are determined based on the dynamic topology structure adjustment mechanism. The model can dynamically adjust the number of network nodes and layer weights according to the characteristics of the input data. This adaptability enables the model to find the optimal network structure on different data sets, thereby better retaining key information.

[0037] Step S108: Use a high-order neural network as a classification algorithm to classify the key features after dimensionality reduction and output a classification result.

[0038] Step S110: Predict the lubricating oil replacement cycle of the target vehicle based on the classification result.

[0039] Among them, in the embodiment of the present invention, a trained high-order neural network is used as a classification algorithm, which can classify new dimensionality-reduced features and output classification results. The classification results are used to characterize the replacement attribute of the lubricating oil of the target vehicle, which can be binary classification (need to be replaced / do not need to be replaced) or multi-classification (different levels of replacement requirements). Further, the current lubricating oil state of the target vehicle is determined according to the classification results. For example, if the classification result is "need to be replaced", it indicates that the current lubricating oil state is not good and needs to be replaced as soon as possible.

[0040] There are situations of gradient disappearance or gradient explosion in traditional classifiers, and the training process is unstable. In addition, in high-dimensional data, it is difficult to find the most discriminative feature combinations, and the key features in the data cannot be accurately identified and utilized, resulting in poor classification effects in practical applications and unable to meet the requirements of practical applications. In this regard, the weight gradient of the high-order neural network of the present invention is adjusted based on the mutual information between the key features and the output, and a high-order training mechanism with gradient penalty is adopted. By measuring the metric of the dependence relationship between two random variables, the key features that are most helpful for the classification task can be selected, reducing the interference of irrelevant or redundant features and improving the classification accuracy of the model. This helps the model better capture the important patterns in the data. Moreover, by adding a gradient-related penalty term to the loss function, the model is encouraged to generate smoother gradients, improving the stability and generalization ability of the model.

[0041] In summary, a data processing method for predicting the replacement cycle of vehicle lubricating oil provided by the embodiment of the present invention can reduce the data dimension, improve the data processing speed, and can enhance the recognition ability of key features, improve the classification accuracy, enhance the model stability, improve the generalization ability, and better process unbalanced data, enhancing the classification effect and practical value.

[0042] Further, on the basis of the above embodiment, the embodiment of the present invention also provides another data processing method for predicting the replacement cycle of vehicle lubricating oil. Figure 2 The flowchart of another data processing method for predicting the replacement cycle of vehicle lubricating oil provided by the embodiment of the present invention is shown. Referring to Figure 2 , the method includes the following steps:

[0043] Step S202, obtain the maintenance records and vehicle monitoring operation parameters of the target vehicle, and assemble the maintenance records and vehicle operation monitoring parameters into vector data.

[0044] Step S204, input the vector data into a pre-constructed feature extraction model, and output the key features of the vector data.

[0045] The present invention uses a three-layer fully connected neural network for feature extraction. In the prior art, some solutions use neural networks for feature extraction. In some neural network structures, problems such as gradient disappearance, gradient explosion, or getting stuck in local optimal solutions may be encountered, which affect the stability of training and the performance of the model. The present invention uses a neural network based on the grey wolf optimization algorithm with historical iteration influence. Based on the traditional grey wolf optimization algorithm, an adaptive mechanism is adopted to adjust the search strategy according to the optimization results of the previous stage, improving the search efficiency and accuracy of the algorithm in the high-dimensional data space.

[0046] Among them, the feature extraction model of the embodiment of the present invention is trained through the following steps:

[0047] 1) Initialize the parameters of the neural network. The parameters of the neural network include weights and biases. In one embodiment, the initialization method is expressed as:

[0048]

[0049] In the formula, represents the initial value of the neural network parameters; is the initialization standard deviation; G t (0, I) represents a Gaussian distribution with a mean of 0 and a covariance of the identity matrix. Preferably, is set to 0.01.

[0050] 2) Update the neural network parameters using a preset search strategy, and dynamically adjust the key parameters in the search strategy by fusing historical iteration information.

[0051] Among them, the search strategy includes the grey wolf optimization algorithm. The grey wolf optimization algorithm (GWO, Grey Wolf Optimizer) is a meta-heuristic optimization algorithm based on the predation behavior of grey wolves in nature. In the embodiment of the present invention, the key parameters of the grey wolf optimization algorithm include global search parameters and local search parameters. The grey wolf optimization algorithm with historical iteration influence adjusts the parameters of the neural network by simulating the social behavior of grey wolves. Each wolf represents a set of potential solutions of the neural network, and the social behavior of grey wolves helps to explore the solution space and find the optimal solution. The way to update the position of grey wolves is expressed as:

[0052]

[0053] In the formula, and are the positions of the wolves in the t-th and (t + 1)-th iterations respectively; is the individual best position in the grey wolves; is the global best position; and are parameters for weighing exploration and exploitation. Preferably, and They are set to 0.3 and 0.5 respectively.

[0054] In the embodiment of the present invention, by introducing parameters and to balance the exploration and exploitation behaviors in the algorithm. It allows the algorithm to flexibly switch between global search and local search according to the characteristics of the problem, thereby improving the ability to solve complex optimization problems. Especially when they are set to 0.3 and 0.5 respectively, it tends to make more use of the information of the found good solutions for more detailed search. Compared with the traditional Grey Wolf Optimization algorithm, in the embodiment of the present invention, not only fixed values are defined as the weights of exploration and exploitation, but also a method is proposed to dynamically adjust one of these weights according to the historical iteration performance (i.e., ), which can enhance the adaptability of the algorithm to different optimization scenarios. In specific implementation, a preset exponential decay term is used as the weight to calculate the parameter adjustment factor corresponding to the loss function value of the feature extraction model; based on the parameter adjustment factor and the iteration situation of the preset search strategy, the search step size of the search strategy is adjusted to adjust the key parameters in the search strategy.

[0055] Specifically, in each algorithm iteration, the search strategy is adjusted according to the optimization result of the previous generation, and the adaptive adjustment method is expressed as:

[0056]

[0057] In the formula, is the updated ; is the adjustment factor used to adjust the size; represents the Euclidean distance, that is, the L2 norm.

[0058] In one embodiment, the adjustment factor is adjusted based on the feedback of historical optimization performance, and the calculation method is expressed as:

[0059]

[0060] In the formula, T ve is the number of historical iterations considered, indicating how many generations of performance affect the current adjustment; is the decay factor, is the decay rate; is the loss function value at the t-th iteration; is the neural network parameter at the t-th iteration, that is, the individual best position x of the Grey Wolf at the t-th iteration cp corresponding neural network parameters. Preferably, is set to 3.

[0061] In summary, in the embodiments of the present invention, the best position x of the previous generations cp and the global best position x cg , as well as the position of the current individual are used to calculate the new position, and this method effectively integrates past successful experiences and current state information. In addition, the calculation of the adjustment factor takes into account the change of the loss function value within the historical iteration times, further improving the accuracy of the selection of the search direction.

[0062] Moreover, the embodiments of the present invention consider the influences of both the individual best position and the global best position and assign different weights. It not only retains the advantages of the traditional grey wolf optimization algorithm and can quickly locate the optimal region, but also strengthens the directivity towards the specific target point, enabling the algorithm to perform more precise fine-tuning when approaching the optimal solution. When calculating the adjustment factor, an exponential decay term is adopted, which means that as time goes by (i.e., the number of iterations increases), the importance of the early iteration results gradually decreases, helping the algorithm to respond faster to the newly discovered good solutions and avoiding search stagnation caused by over-reliance on the information in the earlier stage. Update the weights and biases of the neural network, and the update method is expressed as:

[0063]

[0064] In the formula, and are the neural network parameters of the t-th iteration and the (t + 1)-th iteration respectively; is the learning rate of the neural network; is the gradient of the loss function at the parameter . Preferably, is set to 0.01.

[0065] Furthermore, the calculation method of the gradient is expressed as:

[0066]

[0067] In the formula, is the number of samples in each batch of training data; is the input data of the -th sample; is the true label of the -th sample; is the cross-entropy loss function; is the predicted output of the neural network model.

[0068] Repeat the above steps iteratively until the preset stop iteration condition is met, which indicates that the model training is completed. In one embodiment, the preset stop iteration condition is to reach the preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times.

[0069] The embodiment of the present invention uses a three-layer fully connected neural network for feature extraction. By improving the performance of the Grey Wolf Optimization algorithm in dealing with complex, high-dimensional or multimodal optimization problems, it more intelligently controls the exploration and exploitation ratio in the search process, uses historical information to guide future decisions, etc., in order to achieve a better solution, enabling the model to extract key information from the data and overcoming the low performance of previous models due to insufficient feature extraction.

[0070] Step S206, perform dimensionality reduction processing on the key parameters through a pre-constructed feature dimensionality reduction model.

[0071] The present invention uses an autoencoder algorithm for feature dimensionality reduction. The autoencoder algorithm consists of two parts: an encoder and a decoder. The encoder is responsible for converting high-dimensional input data into low-dimensional feature representations, and the decoder then attempts to reconstruct the original input from these low-dimensional features, ensuring that as little information as possible is lost during the dimensionality reduction process. The present invention adopts a dynamic topology adjustment mechanism, which enables the network to automatically adjust its internal structure according to the complexity of the input data, such as increasing or decreasing the number of neural network layers or the number of neurons, optimizing the sparse representation ability of the network, and improving the efficiency and effect of feature dimensionality reduction.

[0072] In specific implementation, the training steps are as follows:

[0073] 1) Initialize the network parameters of the autoencoder. In one embodiment, the network parameters are initialized using the He initialization method, and the initialization method is expressed as:

[0074]

[0075] In the formula, represents the weight of the l-th layer of the autoencoder; represents the number of nodes in the (l - 1)-th layer; represents a random value drawn from the standard normal distribution

[0076] 2) Determine the number of network nodes and layer weights of the feature dimensionality reduction model based on the dynamic topology adjustment mechanism.

[0077] a - Train the autoencoder using a preset training sample set.

[0078] In the forward propagation process of the data, the data is compressed into low-dimensional features by the encoder and then reconstructed by the decoder. The output of the encoder ​Reconstruction of the decoder is calculated as follows:

[0079]

[0080]

[0081] In the formula, represents the ReLU activation function; , respectively represent the weights of the encoder and the decoder; , respectively represent the biases of the encoder and the decoder; represents the input data.

[0082] b - Evaluate the importance of each feature in the training sample set during the decoding process through the sensitivity scoring function.

[0083] c - Adjust the weight importance of each layer of the autoencoder based on the importance.

[0084] In one embodiment, an adaptive feature feedback adjustment mechanism is used to calculate the adaptive feature feedback adjustment increment of the encoder, which is used to adjust the weights of the feature representation in real time, and dynamically adjust the weight importance of each layer of the encoder according to the error output by the decoder. The calculation method is expressed as:

[0085]

[0086] In the formula, represents the learning rate of the adaptive feature feedback adjustment mechanism; represents the loss function of the decoder; represents the partial derivative of the loss function with respect to the output of the encoder; is the sensitivity scoring function; represents element-wise multiplication. Preferably, is set to 0.01.

[0087] Moreover, in this embodiment, the sensitivity scoring function is used to evaluate the importance of each feature during the decoding process. The calculation method is expressed as:

[0088]

[0089] Use the gradient descent method to update the network parameters to minimize the loss function. The update method is expressed as:

[0090]

[0091] In the formula, represents the learning rate; represents the partial derivative of the loss function with respect to the weight; Represents the incremental adaptive feature feedback adjustment of the encoder.

[0092] d - Use an approximation method to calculate the gradient of the loss function of the auto - encoder with respect to the number of network nodes.

[0093] In the initial stage of training, automatically adjust the network hierarchy according to the training progress and reconstruction error of the network. The calculation method of the adjustment function is expressed as:

[0094]

[0095] In the formula, is the adjustment function for adjusting the number of network nodes. If its value exceeds the threshold, increase the number of neurons in the middle layer; if its value does not exceed the threshold, reduce the number of neurons in the middle layer; represents the number of nodes in the l - th layer; represents the structural adjustment step size; represents the current loss function; represents the gradient of the loss function with respect to the number of nodes in the l - th layer. Further, considering that the number of nodes directly affects the output dimension of the network layer, an approximation method is used to calculate the loss function with respect to the number of nodes The calculation method of the gradient is expressed as:

[0096]

[0097] In the formula, represents a random value drawn from the standard normal distribution

[0098] e - Based on the gradient, as well as the training progress and reconstruction error of the auto - encoder, adjust the number of network nodes of the auto - encoder.

[0099] Specifically, to calculate the loss function of the auto - encoder, the present invention uses the mean - square error loss based on L1 regularization as the loss function of the auto - encoder. The calculation method is expressed as:

[0100]

[0101] In the formula, is the loss function of the auto - encoder; represents the reconstruction error; represents the regularization coefficient; represents the L1 regularization term of the weight. Preferably, is set to 0.3.

[0102] Further, the calculation method of the reconstruction error is expressed as:

[0103] ​

[0104] Where \(m\) is the number of samples input to the autoencoder in the current batch, and are the \(k\)-th elements of the actual input and the reconstructed output, respectively.

[0105] And the L1 regularization term is used to increase the sparsity of the weights, and its calculation method is expressed as:

[0106]

[0107] Here, \(n\) r is the dimension of the weight vector, is the \(i\)-th element of the weight vector.

[0108] Repeat the above steps iteratively until the preset stop iteration condition is satisfied, which indicates that the model training is completed. In one embodiment, the preset stop iteration condition is to reach the preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times. In summary, by using the autoencoder algorithm of the embodiment of the present invention for feature dimensionality reduction, the high-dimensional data is converted into low-dimensional data through the encoder while maintaining the important information of the data, thereby optimizing the processing speed and model efficiency.

[0109] Step S208: Use a high-order neural network as the classification algorithm to classify the key features after dimensionality reduction and output the classification result.

[0110] The present invention uses a high-order neural network as the classification algorithm and adopts a high-order training mechanism with gradient penalty to adjust the weight gradient of each layer during the training process, ensuring that the network pays more attention to the features that have the greatest impact on the final classification decision during the learning process, thereby improving the classification accuracy.

[0111] Specifically, the training process of the high-order neural network classification algorithm is as follows:

[0112] 1) Initialize the parameters of the high-order neural network. In one embodiment, the weights and biases of the high-order neural network are initialized in the He initialization manner, which is expressed as:

[0113]

[0114]

[0115] Where represents the weights of the \(l\)-th layer of the high-order neural network; represents the biases of the \(l\)-th layer of the high-order neural network; represents the number of nodes in the \((l - 1)\)-th layer of the high-order neural network; Represents a random value drawn from a normal distribution with a mean of 0 and a standard deviation of 1.

[0116] 2) Based on the mutual information between the input and the output, a high-order training mechanism with gradient penalty is used to adjust the weight gradients of the high-order neural network.

[0117] Specifically, obtain the mutual information between the input and the output of the high-order neural network, and use the chain rule to solve the weight gradients of the high-order neural network; update the weight gradients based on the mutual information and a preset gradient penalty term.

[0118] During the forward propagation process, the dimension-reduced input data passes through a multi-layer neural network. Each layer uses the ReLU activation function to transform the linearly weighted input, and the calculation of the output of each layer is expressed as:

[0119]

[0120] In the formula, Represents the output of the l-th layer; Represents the output of the (l - 1)-th layer. For the first layer, Is the input feature; Represents the ReLU activation function.

[0121] Use the loss function to calculate the difference between the model output and the true label, and the calculation method is expressed as:

[0122]

[0123] In the formula, Represents the total number of categories; Represents the i-th element in the one-hot encoding of the true label; Is the probability of the i-th element after the Softmax function processing of the model output, where the purpose of the Softmax function is to convert the output into a probability distribution.

[0124] Calculate the gradient of the loss function with respect to each parameter, and adjust the gradients of those parameters that are sensitive to the loss change through the gradient penalty mechanism, so that the model training pays more attention to important features. Then the calculation method of the weight update increment is expressed as:

[0125]

[0126] In the formula, Is the input feature And the output Between the mutual information; Represents the learning rate of the high-order neural network; Represents the gradient penalty coefficient; Is the sign function of the weight. Preferably, Set to 0.01, Set to 3.

[0127] Furthermore, the gradient is calculated using the chain rule, and the calculation method is expressed as:

[0128]

[0129] In the formula, is the partial derivative of the loss function with respect to the output of the last layer of the network, is the partial derivative of the output of the last layer with respect to the weight.

[0130] Furthermore, an adaptive feature correlation feedback mechanism is adopted to dynamically adjust the connection weights between layers in the neural network, and the adjustment is based on the mutual information between features, so as to improve the model's recognition ability of the complex relationship between the influencing factors of the lubricating oil replacement cycle. Then the input feature and the output The calculation method of the mutual information between them is expressed as:

[0131]

[0132] In the formula, is the mutual information between the input feature and the output , is the joint probability density function, and are the marginal probability density functions.

[0133] Furthermore, the embodiments of the present invention adopt a gradient-based method to update the model parameters of the high-order neural network to improve the optimization efficiency and convergence speed. The update method is expressed as:

[0134]

[0135]

[0136] In the formula, is the gradient of the bias with respect to the loss. Repeat the above steps iteratively until the preset stop iteration condition is satisfied, which means the model training is completed. In one embodiment, the preset stop iteration condition is to reach the preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times. In summary, the embodiments of the present invention adopt a high-order neural network and a gradient penalty mechanism to train the classifier model, strengthen the model's learning of key features, and improve the accuracy and reliability of classification.

[0137] Step S210, determine the classification result as the prediction result of the lubricating oil replacement cycle of the target vehicle.

[0138] The classification results include normal replacement and early replacement. Among them, "normal replacement" refers to the replacement according to the manufacturer's recommendation and the standard replacement cycle under operating conditions, and "early replacement" refers to the need for early lubricating oil replacement due to harsh operating conditions (such as high load, high temperature).

[0139] Furthermore, based on the above embodiments, the embodiments of the present invention also design a training sample set. Correspondingly, the embodiments of the present invention also provide another data processing method for predicting the lubricating oil replacement cycle of vehicles. This embodiment mainly describes the construction method of the training sample set. Refer to Figure 3 This method includes the following steps:

[0140] Step S10, obtain the pre-constructed initial sample set.

[0141] Among them, the initial sample set refers to the maintenance records and vehicle monitoring parameters of the above embodiments, which will not be elaborated here. During the construction of the training sample set, the embodiments of the present invention also prepare a label for each sample to label the collected data. This label indicates whether the lubricating oil needs to be replaced. In one embodiment, the labeled categories include: "normal replacement" and "early replacement". For example, "needs to be replaced" can be marked as 1, and "early replacement" can be marked as 0. Among them, the labels can be generated by manual annotation. It should be noted that "normal replacement" refers to the replacement according to the manufacturer's recommendation and the standard replacement cycle under operating conditions, and "early replacement" refers to the need for early lubricating oil replacement due to harsh operating conditions (such as high load, high temperature).

[0142] Step S11, use the attention mechanism of deep learning to perform feature distillation processing on the initial sample set, and extract the key feature vectors in the initial sample set.

[0143] In the task of the present invention, the acquisition, annotation, and preprocessing of training data are time-consuming and laborious, and insufficient training samples are likely to lead to poor generalization ability of the model and affect the accuracy of the model at the same time. The embodiments of the present invention use an adaptive feature distillation-based data augmentation network algorithm to generate samples. Before the start of each training cycle, key feature vectors are extracted from the real dataset through the feature distillation unit. Further, these feature vectors are used as additional guiding information to enable the data augmentation algorithm to generate augmented samples, assisting the data augmentation algorithm to more accurately capture and simulate the real data distribution, expressed as:

[0144]

[0145]

[0146] In the formula, is the distilled feature vector; is the real dataset; is the fake data generated by the generator; represents the feature distillation function; represents the function for generating fake data.

[0147] In one embodiment, the feature distillation process uses the attention mechanism in deep learning to extract important features, and the calculation method is expressed as:

[0148]

[0149] In the formula, Q c , K c and V c are the query matrix, key matrix, and value matrix learned from the input data , is the Softmax function, d k is the dimension of the key vector. Preferably, the dimension d k of the key vector is set to 64.

[0150] Furthermore, the query, key, and value matrices are obtained through linear transformation, and the calculation method is expressed as:

[0151]

[0152]

[0153]

[0154] In the formula, is the query weight matrix, is the key weight matrix, is the value weight matrix. Moreover, the query weight matrix, key weight matrix, and value weight matrix are determined by a learning method. In this embodiment, the learning method is to perform backpropagation calculation through gradient descent.

[0155] Step S12: Based on the key feature vectors, perform data augmentation on the initial sample set to construct a training sample set.

[0156] In one implementation, the present invention performs data augmentation through a generative adversarial network. The generative adversarial network algorithm based on adaptive feature distillation includes two parts: a generator ( ) and a discriminator ( ), the generator is responsible for generating new data instances, and the discriminator is responsible for evaluating the authenticity of these data instances. To enhance the data generation ability of the model and optimize the diversity and authenticity of the generated data, the present invention adopts a feature distillation unit to learn and distill key features from the real dataset, guiding the generator to more accurately simulate the data distribution, thereby enhancing the exploration ability of the generated samples and achieving a wider coverage of the data space. In a specific implementation, the embodiments of the present invention perform quantum gate operations on the parameters of the generative adversarial network based on preset Euler angle parameters to initialize the parameters (the generator G c and the discriminator D c parameters) of the generative adversarial network. In one embodiment, the parameters of the generator and the parameters of the discriminator are initialized using an initialization function of quantum gate transformation, enabling more flexible adjustment space for the parameters during initialization. Taking the generator as an example, when the initialization function of quantum gate transformation initializes the parameters of the generator, the parameter initialization method is expressed as:

[0157]

[0158] In the formula, represents a quantum gate operation parameterized by the Euler angles , and parameters, is the first Euler angle parameter, is the second Euler angle parameter, is the third Euler angle parameter.

[0159] Furthermore, the Euler angle parameters intuitively describe the rotation of an object in three-dimensional space, determining the initialization direction of the quantum bits. When initializing the parameters of the generative network, the randomness of parameter initialization is increased, and multiple angle parameters are introduced to enable more flexible adjustment directions for the parameters. The calculation method is expressed as:

[0160]

[0161]

[0162]

[0163] In the formula, rand(0, 1) is a function that generates random numbers uniformly distributed between 0 and 1.

[0164] Furthermore, the generator uses the current quantum state and the distilled feature vector to generate new data instances, and the discriminator evaluates the similarity between the generated data and the real data and feeds back to the generator to adjust the parameters. The rule for updating the parameters through the alternating training of the generator and the discriminator is expressed as:

[0165]

[0166]

[0167] In the formula, and are the learning rates of the generator and the discriminator, respectively; is the total loss function of the generative adversarial network; and represent the gradients of the corresponding parameters, respectively. Preferably, and are both set in an adaptive adjustment manner. The initial value of is set to 0.01, and the initial value of and are defined as the loss change rates of the generator and the discriminator, respectively, and the calculation method is expressed as:

[0168]

[0169]

[0170] In the formula, and represent the derivatives of the loss functions of the generator and the discriminator with respect to time, and are used to detect the change speed of the loss.

[0171] Moreover, in this embodiment, and the adaptive setting method is expressed as:

[0172]

[0173]

[0174] In the formula, and are the learning rates of the generator and the discriminator at the t-th iteration, respectively. and are the initial learning rates of the generator and the discriminator; and are the first adjustment coefficient and the second adjustment coefficient, respectively. Preferably, and are set to 0.01 and 0.03, respectively, and are set to 0.3 and 0.7, respectively.

[0175] Furthermore, the total loss function of the generative adversarial network includes generative adversarial loss, feature matching loss, and quantum state fidelity loss. Calculate the total loss function of the generative adversarial network and continuously optimize the performance of the generator and discriminator. The calculation method of the total loss function of the generative adversarial network is expressed as:

[0176]

[0177]

[0178]

[0179]

[0180] In the formula, is the L2 norm, , , are the weight coefficients of the generative adversarial loss, feature matching loss, and quantum state fidelity loss respectively; , , are the generative adversarial loss, feature matching loss, and quantum state fidelity loss respectively; represents the feature extraction function; represents the inner product of quantum states; is the absolute value symbol. Preferably, , , are set to 1.0, 0.5, 0.5 respectively.

[0181] In one embodiment, the generative adversarial loss is represented by a logarithmic likelihood function, and its calculation method is expressed as:

[0182]

[0183] In the formula, x i represents the i-th sample input to the generative adversarial network, specifically a sample from the real data set or generated data; nc represents the number of samples generated by the current batch input to the generative adversarial network, and y i is the label of the i-th sample input to the generative adversarial network (1 for real data and 0 for generated data), is the output of the discriminator.

[0184] Repeat the above steps iteratively until the preset iteration stop condition is met, which indicates that the model training is completed. In one embodiment, the preset iteration stop condition is to reach the preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times. After the data augmentation model training is completed, the trained data augmentation model is used to increase the number of samples. In one embodiment, if the original collected samples are 800 and the data augmentation model generates 200 samples, the augmented dataset contains 1000 samples. Among them, the training sample set can be used to train the feature extraction model, the feature dimensionality reduction model, and the high-order neural network. In one implementation, the feature extraction model can be used to extract features from the training sample set, and then the feature dimensionality reduction model is trained. Further, the high-order neural network can also be trained after the feature dimensionality reduction model reduces the dimensionality of its training samples. In summary, the embodiment of the present invention effectively solves the problem of insufficient training data volume by adopting the adaptive feature distillation generative adversarial network, learns and distills key features from the real dataset to guide the generator to more accurately simulate the data distribution.

[0185] Furthermore, the embodiment of the present invention also provides a data processing device for predicting the lubricating oil replacement cycle of a vehicle. Figure 4 shows a schematic structural diagram of a data processing device for predicting the lubricating oil replacement cycle of a vehicle provided by the embodiment of the present invention. Refer to Figure 4 The device includes: a data acquisition module 100, configured to acquire the maintenance records and vehicle monitoring parameters of the target vehicle, and assemble the maintenance records and vehicle monitoring parameters into vector data; a feature extraction module 200, configured to input the vector data into a pre-constructed feature extraction model and output the key features of the vector data; wherein, the feature extraction model updates the neural network parameters using a preset search strategy, and dynamically adjusts the key parameters in the search strategy by fusing historical iteration information; a feature dimensionality reduction module 300, configured to perform dimensionality reduction processing on the key parameters through a pre-constructed feature dimensionality reduction model; wherein, the number of network nodes and layer weights of the feature dimensionality reduction model are determined based on a dynamic topology adjustment mechanism; an execution module 400, configured to use a high-order neural network as a classification algorithm to classify the dimensionality-reduced key features and output a classification result; wherein, the classification result is used to characterize the replacement attribute of the lubricating oil of the target vehicle; the weight gradient of the high-order neural network is adjusted using a high-order training mechanism with gradient penalty based on the mutual information between the input and the output; an output module 500, configured to predict the lubricating oil replacement cycle of the target vehicle based on the classification result.

[0186] The data processing device for predicting the lubricating oil replacement cycle of a vehicle provided by the embodiment of the present invention has the same implementation principle and the same technical effects as those of the foregoing method embodiment. For the sake of brief description, for the parts not mentioned in the device embodiment, reference can be made to the corresponding content in the foregoing method embodiment.

[0187] Further, based on the above embodiments, an embodiment of the present invention further provides another data processing device for predicting the lubricating oil replacement cycle of a vehicle. Figure 5 FIG. shows a schematic structural diagram of another data processing device for predicting the lubricating oil replacement cycle of a vehicle provided by an embodiment of the present invention. Among them, the classification result includes normal replacement and early replacement; the above output module 500 is further configured to determine the classification result as the prediction result of the lubricating oil replacement cycle of the target vehicle. The above execution module 400 is further configured to obtain the mutual information between the input and output of the high-order neural network, and solve the weight gradient of the high-order neural network by using the chain rule; update the weight gradient based on the mutual information and a preset gradient penalty term. The above execution module 400 is further configured to update the model parameters of the high-order neural network by using a gradient-based method.

[0188] The above feature extraction module 200 is further configured to use a preset exponential decay term as a weight to calculate a parameter adjustment factor corresponding to the loss function value of the feature extraction model; adjust the search step size of the search strategy based on the parameter adjustment factor and the iteration situation of the preset search strategy to adjust the key parameters in the search strategy. The search strategy includes a grey wolf optimization algorithm; the key parameters of the grey wolf optimization algorithm include a global search parameter and a local search parameter. The feature dimensionality reduction model includes an autoencoder; the above feature dimensionality reduction module 300 is further configured to train the autoencoder through a preset training sample set, evaluate the importance of each feature in the training sample set during the decoding process through a sensitivity scoring function; adjust the weight importance of each layer of the autoencoder based on the importance; calculate the gradient of the loss function of the autoencoder with respect to the number of network nodes by using an approximation method; adjust the number of network nodes of the autoencoder based on the gradient, as well as the training progress and reconstruction error of the autoencoder.

[0189] The device further includes a construction module 600, configured to obtain a pre-constructed initial sample set; perform feature distillation processing on the initial sample set by using the attention mechanism of deep learning to extract key feature vectors in the initial sample set; perform data augmentation on the initial sample set based on the key feature vectors to construct a training sample set. Among them, data augmentation is performed through a generative adversarial network, and the total loss function of the generative adversarial network includes a generative adversarial loss, a feature matching loss, and a quantum state fidelity loss; the above construction module 600 is further configured to perform quantum gate operations on the parameters of the generative adversarial network based on preset Euler angle parameters to initialize the parameters of the generative adversarial network.

[0190] An embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above is implemented. Figures 1 to 3Steps of any of the methods shown. An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the above-mentioned Figures 1 to 3 Steps of any of the methods shown. An embodiment of the present invention further provides a schematic structural diagram of an electronic device, as Figure 6 shown, which is the schematic structural diagram of the electronic device. Among them, the electronic device includes a processor 61 and a memory 60. The memory 60 stores computer-executable instructions that can be executed by the processor 61. The processor 61 executes the computer-executable instructions to implement the above-mentioned Figures 1 to 3 Steps of any of the methods. In Figure 6 the shown embodiment, the electronic device further includes a bus 62 and a communication interface 63. Among them, the processor 61, the communication interface 63 and the memory 60 are connected through the bus 62. Among them, the memory 60 may include a high-speed random access memory (RAM, Random Access Memory), and may also include a non-volatile memory, such as at least one disk memory. Through at least one communication interface 63 (which can be wired or wireless), a communication connection is realized between the system network element and at least one other network element. The Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 62 can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc., and can also be an AMBA (Advanced Microcontroller Bus Architecture) bus. Among them, AMBA defines three buses, including an APB (Advanced Peripheral Bus) bus, an AHB (Advanced High-performance Bus) bus and an AXI (Advanced eXtensible Interface) bus. The bus 62 can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 6 only a bidirectional arrow is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0191] The processor 61 may be an integrated circuit chip with the ability to process signals. In the implementation process, the steps of the above method can be completed by the integrated logic circuit of the hardware in the processor 61 or the instructions in the form of software. The above-mentioned processor 61 may be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it may also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field-programmable gate array (FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by the combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory, and the processor 61 reads the information in the memory and combines its hardware to complete the foregoing Figures 1 to 3 any of the illustrated methods.

[0192] A computer program product for a data processing method and device for predicting the replacement cycle of vehicle lubricating oil provided by an embodiment of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the foregoing method embodiments. For specific implementation, reference can be made to the method embodiments and will not be elaborated here. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process of the above-described system can refer to the corresponding process in the foregoing method embodiments and will not be elaborated here. If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program code. Finally, it should be noted that the above embodiments are only specific implementation manners of the present invention, used to illustrate the technical solutions of the present invention, and are not intended to limit them. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions described in the foregoing embodiments or easily conceive of changes, or equivalently replace some of the technical features; and these modifications, changes, or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention and should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A data processing method for predicting vehicle lubricant replacement cycle, characterized in that: The method comprises: Acquire maintenance records and vehicle monitoring parameters of a target vehicle, and assemble the maintenance records and the vehicle monitoring parameters into vector data; Input the vector data into a pre-built feature extraction model, and output key features of the vector data; wherein the feature extraction model uses a preset search strategy to update neural network parameters, and dynamically adjusts key parameters in the search strategy by fusing historical iteration information; The key parameters are subjected to dimensionality reduction processing by a pre-built feature dimensionality reduction model; wherein the number of network nodes and layer weights of the feature dimensionality reduction model are determined based on a dynamic topology structure adjustment mechanism; A high-order neural network is used as a classification algorithm to classify the key features after dimensionality reduction and output classification results; wherein the classification results are used to characterize the lubricant replacement attributes of the target vehicle; the weight gradient of the high-order neural network is adjusted based on the mutual information between the input and output, using a high-order training mechanism with gradient penalty; Based on the classification result, predicting the lubricant replacement cycle of the target vehicle; The feature dimensionality reduction model includes an autoencoder; the step of determining the number of network nodes and layer weights of the feature dimensionality reduction model based on a dynamic topology structure adjustment mechanism includes: The autoencoder is trained using a preset training sample set, and the importance of each feature of the training sample set in the decoding process is evaluated using a sensitivity scoring function; Based on the importance, adjusting the weight importance of each layer of the autoencoder network; Using an approximate method, the gradient of the loss function of the autoencoder with respect to the number of network nodes is calculated; Based on the gradient, and the training progress and reconstruction error of the autoencoder, adjusting the number of network nodes of the autoencoder; Among them, the calculation method of the sensitivity scoring function is expressed as: ； in, is the encoder output of the autoencoder, A reconstruction of the decoder of the autoencoder; The calculation method for adjusting the weight importance is expressed as: ； in, represents the learning rate of the adaptive feature feedback regulation mechanism; represents the loss function of the decoder; represents the partial derivative of the loss function with respect to the encoder output; is the sensitivity scoring function; Represents element-wise multiplication; The calculation method for adjusting the number of network nodes of the autoencoder is expressed as: ； In the formula, The adjustment function for adjusting the number of network nodes is characterized as follows: if its value exceeds the threshold, the number of neurons in the middle layer is increased; if its value does not exceed the threshold, the number of neurons in the middle layer is reduced; Indicates the number of nodes in the lth layer; represents the step length of structural adjustment; Represents the current loss function; Represents the gradient of the loss function with respect to the number of nodes in the lth layer; An approximate method is used to calculate the gradient of the loss function of the autoencoder with respect to the number of network nodes. The calculation method is expressed as: In the formula, Represents the standard normal distribution The random value to draw.

2. The method according to claim 1, characterized in that The classification results include normal replacement and early replacement; Based on the classification result, the step of predicting the lubricant replacement cycle of the target vehicle includes: The classification result is determined as a prediction result of a lubricant replacement period of the target vehicle.

3. The method according to claim 1, characterized in that The weight gradient of a high-order neural network is based on the mutual information between input and output, and the steps of adjusting it using a high-order training mechanism with gradient penalty include: Obtaining mutual information between input and output of the high-order neural network, and solving the weight gradient of the high-order neural network using the chain rule; The weight gradient is updated based on the mutual information and a preset gradient penalty term.

4. The method according to claim 3, characterized in that The method further comprises: A gradient-based method is used to update the model parameters of the high-order neural network.

5. The method according to claim 1, characterized in that The step of dynamically adjusting key parameters in the preset search strategy by fusing historical iteration information in the feature extraction model includes: Using a preset exponential decay term as a weight, calculating a parameter adjustment factor corresponding to the loss function value of the feature extraction model; Based on the parameter adjustment factor and the iteration status of the preset search strategy, the search step size of the search strategy is adjusted to adjust the key parameters in the search strategy.

6. The method according to claim 5, characterized in that The search strategy includes a gray wolf optimization algorithm; Among them, the key parameters of the gray wolf optimization algorithm include global search parameters and local search parameters.

7. The method according to claim 1, characterized in that The method further comprises: Get a pre-built initial sample set; Using a deep learning attention mechanism, the initial sample set is subjected to feature distillation processing to extract key feature vectors from the initial sample set; Based on the key feature vector, data expansion is performed on the initial sample set to construct a training sample set.

8. The method according to claim 7, characterized in that Data augmentation is performed by generating an adversarial network, wherein the total loss function of the generated adversarial network includes a generated adversarial loss, a feature matching loss, and a quantum state fidelity loss; the method further comprises: Based on the preset Euler angle parameters, quantum gate operations are performed on the parameters of the generative adversarial network to initialize the parameters of the generative adversarial network.

9. A data processing device for predicting the replacement cycle of vehicle lubricating oil, characterized in that: The device comprises: A data acquisition module, used to acquire the maintenance record and vehicle monitoring parameters of the target vehicle, and assemble the maintenance record and the vehicle monitoring parameters into vector data; A feature extraction module, used to input the vector data into a pre-built feature extraction model and output key features of the vector data; wherein the feature extraction model uses a preset search strategy to update neural network parameters, and dynamically adjusts key parameters in the search strategy by fusing historical iteration information; A feature dimension reduction module, used for reducing the dimension of the key parameters through a pre-built feature dimension reduction model; wherein the number of network nodes and layer weights of the feature dimension reduction model are determined based on a dynamic topology structure adjustment mechanism; An execution module is used to classify the key features after dimensionality reduction by using a high-order neural network as a classification algorithm, and output a classification result; wherein the classification result is used to characterize the replacement attribute of the lubricant oil of the target vehicle; the weight gradient of the high-order neural network is adjusted based on the mutual information between the input and the output, using a high-order training mechanism of gradient penalty; An output module, used for predicting a lubricant replacement cycle of the target vehicle based on the classification result; The feature dimension reduction model includes an autoencoder; the feature dimension reduction module is further used to: train the autoencoder through a preset training sample set, and evaluate the importance of each feature of the training sample set in the decoding process through a sensitivity scoring function; based on the importance, adjust the weight importance of each layer of the network of the autoencoder; use an approximate method to calculate the gradient of the loss function of the autoencoder with respect to the number of network nodes; based on the gradient, as well as the training progress and reconstruction error of the autoencoder, adjust the number of network nodes of the autoencoder; Among them, the calculation method of the sensitivity scoring function is expressed as: ； in, is the encoder output of the autoencoder, A reconstruction of the decoder of the autoencoder; The calculation method for adjusting the weight importance is expressed as: ； in, represents the learning rate of the adaptive feature feedback regulation mechanism; represents the loss function of the decoder; represents the partial derivative of the loss function with respect to the encoder output; is the sensitivity scoring function; Represents element-wise multiplication; The calculation method for adjusting the number of network nodes of the autoencoder is expressed as: ； In the formula, The adjustment function for adjusting the number of network nodes is characterized as follows: if its value exceeds the threshold, the number of neurons in the middle layer is increased; if its value does not exceed the threshold, the number of neurons in the middle layer is reduced; Indicates the number of nodes in the lth layer; represents the step length of structural adjustment; Represents the current loss function; Represents the gradient of the loss function with respect to the number of nodes in the lth layer; An approximate method is used to calculate the gradient of the loss function of the autoencoder with respect to the number of network nodes. The calculation method is expressed as: In the formula, Represents the standard normal distribution The random value to draw.

Images