Equipment health level assessment method and device based on model driving

Abstract

The invention provides an equipment health level assessment method and device based on model driving, and relates to the technical field of equipment health level assessment. Feature extraction is performed on a to-be-tested sample corresponding to operation monitoring data of preset equipment through a pre-constructed feature extraction model; and performing classification prediction by a classifier model based on a deep neural decision tree algorithm, and determining the health state of the equipment. Wherein network parameters of the feature extraction model are updated based on a competition strategy of species in the distributed ecological niche simulation ecological system. A decision branch of a decision tree of the deep neural decision tree algorithm is determined based on the state of a quantum bit of a node of the decision tree, and the state of the quantum bit is calculated by simulating information gain in a decision tree splitting process. Based on the method, the model parameters can be adaptively adjusted according to the change of data distribution, the complex dependency relationship between the features is accurately captured, and the classification precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment health level assessment, and in particular to a model-driven equipment health level assessment method and device. Background Art

[0002] As the scale and complexity of industrial equipment continue to increase, the difficulty of equipment maintenance and management also increases. The normal operation of large-scale equipment is crucial to the stability and efficiency of industrial production. However, equipment failures are often hidden and cause serious economic losses when they occur. In order to avoid the adverse effects of sudden equipment failures, preventive maintenance and equipment health assessment have become research hotspots in the industrial field. Traditional equipment health assessment mainly relies on expert experience or simple threshold judgments, but as the complexity of equipment operation data increases, these methods seem to be unable to cope with it and it is difficult to accurately predict the health status of the equipment.

[0003] Traditional methods usually rely on predefined feature extraction methods when processing device operation data, which lacks adaptability and flexibility. This results in less than ideal feature extraction results when processing different data distributions. Traditional feature extraction methods are also prone to falling into local optimal solutions and are unable to globally optimize feature representation. In addition, when processing complex data structures, traditional classifiers are limited in their processing and expression capabilities and are unable to fully capture the complex relationships in the data. Based on this, the classification accuracy of traditional models on complex data is often not high. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a model-driven equipment health level assessment method and device, which can adaptively adjust model parameters according to changes in data distribution, accurately capture the complex dependencies between features, and improve classification accuracy.

[0005] In a first aspect, an embodiment of the present invention provides a model-driven device health level assessment method, the method comprising: obtaining operation monitoring data of a preset device, converting text data in the operation monitoring data into a vector format, and generating a sample to be tested of the preset device; the operation monitoring data is collected from the operation log and sensor data of the preset device; extracting features of the sample to be tested by a pre-constructed feature extraction model to obtain target features corresponding to the sample to be tested; wherein the network parameters of the feature extraction model are updated based on the assigned ecological niche, and the ecological niche of the network parameters is determined by simulating the competition strategy of species in the ecosystem; inputting the target features into a preset classifier model, performing classification decisions on the target features through the classifier model, and outputting decision results; wherein the classifier model is based on a deep neural decision tree algorithm as a classifier algorithm, the nodes of the decision tree of the deep neural decision tree algorithm are initialized as quantum bits, and the decision branches of the decision tree are determined based on the state of the quantum bits, and the state of the quantum bits is determined by simulating the information gain calculation in the decision tree splitting process; based on the decision result, evaluating the health status of the preset device.

[0006] In combination with the first aspect, an embodiment of the present invention also provides a first implementation of the first aspect, wherein the step of evaluating the health status of a preset device based on the decision result includes: obtaining a decision label indicated by the decision result; and determining the device status indicated by the decision label as the health status of the preset device.

[0007] In combination with the first aspect, an embodiment of the present invention also provides a second implementation of the first aspect, wherein the step of simulating the information gain calculation in the decision tree splitting process to determine the state of the quantum bit includes: initializing the nodes of the decision tree of the preset deep neural decision tree algorithm as quantum bits; training the deep neural decision tree algorithm with a preset training sample set, and determining the rotation parameters of the custom rotation gate corresponding to the training sample set through exponential operations in the complex field; adjusting the entanglement and phase between the quantum bits based on the rotation parameters and the preset quantum logic gate operation; performing quantum measurement on the quantum bit based on the adjusted entanglement and phase between the quantum bits, and determining the probability distribution corresponding to the quantum bit to evaluate the state of the quantum bit.

[0008] In combination with the first aspect, an embodiment of the present invention also provides a third implementation of the first aspect, wherein the step of generating a decision branch of a decision tree based on the state of a quantum bit includes: calculating the decision condition corresponding to the training sample set according to the state of the quantum bit; based on the decision condition, controlling the node splitting operation of the decision tree through a preset decision function to generate a decision branch of the decision tree.

[0009] In combination with the first aspect, an embodiment of the present invention also provides a fourth implementation of the first aspect, wherein an ecological niche is determined based on the competition strategy of species in a simulated ecosystem, and the step of updating the network parameters of a feature extraction model based on the assigned ecological niche includes: training a preset feature extraction model through a preset training sample set; wherein the network parameters of the feature extraction model are assigned an initial ecological niche; calculating the loss function value corresponding to the network parameters of the feature extraction model, using the loss function value as the fitness score corresponding to the network parameters, and calculating the energy regulation factor corresponding to the network parameters; based on the energy regulation factor, calculating the dynamic mutation rate corresponding to the network parameters; migrating the initial ecological niche based on the dynamic mutation rate, thereby updating the network parameters.

[0010] In combination with the first aspect, an embodiment of the present invention also provides a fifth implementation of the first aspect, wherein the step of calculating the dynamic mutation rate corresponding to the network parameters based on the energy adjustment factor includes: calculating the mutation energy corresponding to the network parameters according to the energy adjustment factor and the preset maximum energy input; calculating the adaptability score based on the mutation energy and the preset migration and mutation adjustment coefficients, and determining the dynamic mutation rate corresponding to the network parameters.

[0011] In combination with the first aspect, an embodiment of the present invention also provides a sixth implementation of the first aspect, wherein the method also includes: reducing the dimension of the target feature through a pre-constructed dimensionality reduction model; wherein the method of constructing the dimensionality reduction model includes: inputting a preset training sample set into a preset autoencoder neural network, mapping the training sample set to an initial low-dimensional feature space through an encoder of the autoencoder neural network to obtain an initial low-dimensional feature; determining the feature importance of the initial low-dimensional feature based on the loss function value corresponding to the initial low-dimensional feature; recursively optimizing the feature weight vector of the autoencoder neural network based on the feature importance; determining a new low-dimensional feature based on the feature weight vector, and reconstructing the new low-dimensional feature through a decoder; until the autoencoder neural network meets the preset iteration conditions, a dimensionality reduction model is constructed based on the autoencoder neural network.

[0012] In combination with the first aspect, an embodiment of the present invention also provides a seventh implementation method of the first aspect, wherein the method also includes: obtaining pre-collected equipment operation monitoring samples; the equipment operation monitoring samples include operation logs and sensor data of preset equipment; based on the equipment health status represented by the operation monitoring samples, the operation monitoring samples are labeled, and the text data in the operation monitoring samples are converted into a vector format to construct an initial sample set; the initial sample set is expanded using a pre-built data expansion model to generate expanded samples of the initial sample set; wherein the data expansion model includes a generative adversarial network algorithm, and the generator parameters and discriminator parameters of the generative adversarial network algorithm are adjusted based on the dynamic strategy of game theory; a training sample set is constructed based on the expanded samples and the initial sample set.

[0013] In combination with the first aspect, an embodiment of the present invention also provides an eighth implementation of the first aspect, wherein the step of adjusting the generator parameters and discriminator parameters of the generative adversarial network algorithm based on a dynamic strategy of game theory includes: adjusting the feature weights of a preset training sample set through an adaptive feature transformation strategy, and performing data expansion on the training sample set after the feature weights are adjusted through the generator of the generative adversarial network algorithm to generate initial expanded samples; evaluating the initial expanded samples through the discriminator of the generative adversarial network algorithm to determine the initial loss value of the generator; calculating the loss function of the generator based on the initial loss value and the entropy of the initial expanded samples; and calculating the loss function of the discriminator based on a preset regularization term; adjusting the generator parameters of the generator and the discriminator parameters of the discriminator based on the loss function of the generator and the loss function of the discriminator, as well as the preset dynamic game strategy.

[0014] In a second aspect, an embodiment of the present invention further provides a model-driven device for evaluating the health level of an equipment, the device comprising: a data acquisition module for acquiring operation monitoring data of a preset device, converting text data in the operation monitoring data into a vector format, and generating a sample to be tested of the preset device; the operation monitoring data is collected from the operation log and sensor data of the preset device; a data processing module for extracting features of the sample to be tested through a pre-built feature extraction model to obtain target features corresponding to the sample to be tested; wherein the network parameters of the feature extraction model are updated based on the assigned ecological niche, and the ecological niche of the network parameters is determined by simulating the competition strategy of species in the ecosystem; an execution module for inputting the target features into a preset classifier model, performing classification decisions on the target features through the classifier model, and outputting decision results; wherein the classifier model is based on a deep neural decision tree algorithm as a classifier algorithm, the nodes of the decision tree of the deep neural decision tree algorithm are initialized as quantum bits, and the decision branches of the decision tree are determined based on the state of the quantum bits, and the state of the quantum bits is determined by simulating the information gain calculation in the decision tree splitting process; an output module for evaluating the health status of the preset device based on the decision result.

[0015] The embodiments of the present invention bring the following beneficial effects: The embodiments of the present invention provide a model-driven device health level assessment method and apparatus, which extracts features from the test samples corresponding to the operation monitoring data of the preset device through a pre-built feature extraction model, and then performs classification prediction by a classifier model to determine the health status of the device. Among them, the network parameters of the feature extraction model are updated based on the assigned ecological niche, and the ecological niche of the network parameters is determined by simulating the competition strategy of species in the ecosystem. The classifier model is based on a deep neural decision tree algorithm as a classifier algorithm. The nodes of the decision tree of the deep neural decision tree algorithm are initialized as quantum bits, and the decision branches of the decision tree are determined based on the state of the quantum bit, and the state of the quantum bit is determined by simulating the information gain calculation in the decision tree splitting process. The embodiments of the present invention can adaptively adjust the model parameters according to changes in data distribution, accurately capture the complex dependencies between features, and improve classification accuracy.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of a model-driven device health level assessment method provided in an embodiment of the present invention; Figure 2 A flowchart of another model-driven device health level assessment method provided by an embodiment of the present invention; Figure 3 A schematic diagram of a training process of a dimensionality reduction model provided in an embodiment of the present invention; Figure 4 A flowchart of a method for constructing a training sample set provided by an embodiment of the present invention; Figure 5 A schematic diagram of the structure of a device health level assessment device based on model drive provided by an embodiment of the present invention; Figure 6 A schematic diagram of the structure of another device health level assessment device based on model drive provided by an embodiment of the present invention; Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the following embodiments of the present disclosure are described by specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0021] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described in the present invention may be embodied in a wide variety of forms, and any specific structure and / or function described in the present invention is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described in the present invention may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described in the present invention may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described in the present invention may be used to implement this device and / or practice this method.

[0022] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. The type, quantity and proportion of each component in actual implementation may be changed at will, and the component layout type may also be more complicated. In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the aspects described can be practiced without these specific details.

[0023] The embodiment of the present invention provides a model-driven equipment health level assessment method and device, which can adaptively adjust model parameters according to changes in data distribution, accurately capture the complex dependencies between features, and improve classification accuracy.

[0024] For ease of understanding, a model-driven device health level assessment method provided by an embodiment of the present invention is first described. Figure 1 A flow chart of a model-driven device health level assessment method provided by an embodiment of the present invention is shown. Figure 1 As shown, the method includes: Step S102, obtaining operation monitoring data of a preset device, converting text data in the operation monitoring data into a vector format, and generating a sample to be tested of the preset device.

[0025] In order to evaluate the health status of the equipment, the embodiment of the present invention obtains the operation monitoring data of the equipment and processes the data. Among them, the operation monitoring data of the embodiment of the present invention is collected from the operation log and sensor data of the preset equipment. In one embodiment, the data is collected from the operation log and sensor data of the large equipment. The data is continuously collected through a variety of sensors (such as temperature sensors, pressure sensors, vibration sensors, etc.) and transmitted to the central data storage system through an encryption protocol. When storing, the data is stored in a structured format, specifically in JSON format.

[0026] In one embodiment, the collected data attributes include: temperature R a1 , pressure R a2 , vibration R a3 , current R a4 , usage time R a5 , maintain history R a6 , energy efficiency ratio R a7 , fault history R a8 , Environmental factors R a9 , device configuration R a10 .

[0027] In this embodiment, five of the data are as follows: {70℃,150psi,0.3mm,15A,10000h,None,1.2,None,Mild,Standard}; {85℃,180psi,0.5mm,20A,15000h,mild,1.0,yes,cold,high performance}; {65℃,140psi,0.2mm,10A,8000h,None,1.5,None,Hot,Standard}; {90℃,200psi,0.6mm,25A,20000h,severe,0.8,yes,cold,heavy}; {75℃,165psi,0.4mm,18A,12000h,mild,1.1,yes,mild,high performance}.

[0028] 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 data are usually more than 10 attributes, and the number of attributes of data may reach dozens or even hundreds. In one embodiment, if there is text data in the collected data, let the text be ,in, is the i-th word in the text, nw is the length of the text word vector, and the embodiment of the present invention also converts the collected data into a vector form to adapt to the input format of the machine learning model. In one embodiment, a pre-trained word embedding model E is used. wb Realize the conversion between words and vectors, each word is mapped to a vector space of fixed dimension, namely:

[0029] In the formula, E wb is the word embedding model, is the word in the i-th text, is the word vector after the i-th conversion. Based on this, the text type T wb Convert to vector sequence { , ,…, , }.

[0030] Step S104: extract features from the sample to be tested using a pre-built feature extraction model to obtain target features corresponding to the sample to be tested.

[0031] Through feature extraction, the key features that best represent the state of the device can be extracted from high-dimensional data, and the most valuable information for evaluating the health status of the device can be retained, thereby improving the prediction accuracy and generalization ability of the model. When extracting features, traditional neural network models are easily affected by gradient vanishing, gradient explosion and local optimal solutions, and it is difficult to ensure the training stability and global search ability of the model. To solve this problem, the network parameters of the feature extraction model of the embodiment of the present invention are updated based on the assigned niche, and the niche of the network parameters is determined by simulating the competition strategy of species in the ecosystem. By defining the niche for the position and role of each neuron or parameter group in the model in the feature space, and simulating the competition of species in nature, the niche with lower fitness will be replaced by the niche with higher fitness, and multiple niches can co-evolve to jointly solve complex problems. Through continuous competition and cooperation, the network parameters of the model are gradually optimized, and finally better performance is achieved. Based on the above method, dynamically adjusting the network parameters can improve the training stability of the model and reduce the risk of gradient vanishing and explosion. It can also help the model better explore the solution space and avoid falling into the local optimal solution. In addition, the model parameters can be adaptively adjusted according to the changes in data distribution to improve the generalization ability of the model. Step S106, input the target features into a preset classifier model, perform classification decisions on the target features through the classifier model, and output the decision results.

[0032] According to the nature of the problem and the characteristics of the data, you can select a suitable classifier model and input the processed target features into the pre-trained classifier model. The classifier model will output the health status category of the device, such as "normal", "minor fault", "serious fault", etc.

[0033] In an embodiment of the present invention, the classifier model is based on a deep neural decision tree algorithm as a classifier algorithm. Traditional classifier models usually rely on fixed feature representation methods and lack flexibility. Moreover, when processing complex data structures, it is limited by its processing power and expression capabilities, making it difficult to fully capture the complex relationships in the data, and the classification accuracy is insufficient. For example, although the decision tree model can handle nonlinear relationships, it is easy to overfit when processing high-dimensional data, and the depth and number of nodes of the tree are limited.

[0034] To solve the above problems, the embodiment of the present invention improves the processing power and expression ability of the model by introducing the concept of quantum bits. Among them, the nodes of the decision tree of the deep neural decision tree algorithm are initialized as quantum bits, and the decision branches of the decision tree are determined based on the state of the quantum bits. The state of the quantum bit can represent complex data relationships through characteristics such as quantum state superposition and entanglement. Furthermore, the state of the quantum bit of the embodiment of the present invention is determined by simulating the information gain calculation in the decision tree splitting process, and the model can automatically adjust the state of the quantum bit and the structure of the decision tree to optimize the classification performance. Based on this, the classifier model of the embodiment of the present invention combines the strong expression power of the deep neural network and the interpretability of the decision tree, and can more accurately capture the complex dependencies between features through the state of the quantum bit, improve the calculation accuracy of the information gain, and improve the decision-making ability of the classifier.

[0035] Step S108: Based on the decision result, the health status of the preset device is evaluated.

[0036] The health status of the device can be interpreted based on the output of the classifier. In one embodiment, the decision label indicated by the decision result can be obtained, and the device status indicated by the decision label can be determined as the health status of the preset device. Furthermore, the results can be further analyzed in combination with business knowledge and domain experience. It is also possible to formulate corresponding maintenance plans or take preventive measures based on the health status. For example, if the model predicts that the device is about to fail, maintenance or replacement of parts can be carried out in advance.

[0037] Further, based on the above embodiment, the embodiment of the present invention also provides another model-driven device health level assessment method, which mainly describes the steps of determining the ecological niche based on the competitive strategy of species in the simulated ecosystem and updating the network parameters of the feature extraction model based on the assigned ecological niche. In the embodiment of the present invention, a 5-layer fully connected neural network is used for feature extraction. Among them, a neural network algorithm based on adaptive ecological optimization is used as a feature extraction model, which is inspired by the concept of ecological niche of the ecosystem in nature. Each species has its specific ecological niche, which specifically refers to the position occupied by a species in the ecosystem, the functions performed, and the interaction with the environment. The survival and prosperity of each species depends on whether it can effectively adapt to the changes in its ecological niche. This principle is applied to the optimization of the weights and biases of the neural network, simulating the process of organisms in their ecological niches by constantly adapting to environmental changes to improve their survival rate. In the training process of the neural network, each set of parameters (weights and biases) is regarded as a species in a high-dimensional parameter space, and each species attempts to optimize its ecological niche in the problem space.

[0038] Figure 2FIG. 1 is a flow chart showing another model-driven device health level assessment method provided by an embodiment of the present invention. Figure 2 , the method comprises the following steps: Step S202: training a preset feature extraction model using a preset training sample set.

[0039] The network parameters of the feature extraction model are assigned with initial ecological niches. First, an initial set of neural network parameters is generated, each parameter set is regarded as a species in the ecosystem, and an initial ecological niche is randomly assigned, corresponding to the initialization of the neural network parameters. In one embodiment, the initialization method is expressed as:

[0040]

[0041] In the formula, and They represent the initial weight and initial bias of the i-th species, that is, the initial values ​​of the weight and bias of the corresponding neural network; W p,i and b p,i Respectively represent the weight and bias of the i-th species; is the initialization standard deviation of the neural network, which is used to control the initial distribution range of the parameters; is a normal distribution with a mean of 0 and a variance of 1; represents normal distribution; Indicates that it obeys a specific distribution.

[0042] Furthermore, each set of parameters is randomly assigned to its ecological niche in the ecosystem, expressed as: L p,i ~rand(0,1). L p,i represents the ecological niche of the i-th species, that is, the ecological niche corresponding to the neural network parameters; rand(0,1) represents a random value between 0 and 1.

[0043] Step S204, calculating the loss function value corresponding to the network parameter of the feature extraction model, taking the loss function value as the adaptability score corresponding to the network parameter, and calculating the energy adjustment factor corresponding to the network parameter.

[0044] Each species is evaluated and its loss function value under the current neural network architecture is calculated as its fitness score. The fitness evaluation method is expressed as:

[0045] In the formula, S p,i is the fitness score of the ith species; is the fitness score hyperparameter, which is used to adjust the loss function to the fitness score S p,iThe influence of Loss(W p,i ,b p,i ) represents the usage weight W p,i and bias b p,i Calculated network loss; Loss represents the loss function of the neural network. Preferably, Set to 5. In one embodiment, considering the model complexity and prediction error, the calculation method of the loss function of the neural network is expressed as:

[0046] In the formula, is the weight coefficient for adjusting prediction error and model complexity; MSE is the mean square error of the neural network; Com is the model complexity function. Preferably, Set to 0.5.

[0047] Furthermore, the calculation method of the mean square error of the neural network is expressed as:

[0048] In the formula, is the feature vector output by the neural network for the jth sample, and the model prediction value obtained by the preset Softmax function; y j is the true label of the jth sample; n MSE is the number of samples input to the neural network in the current batch.

[0049] Furthermore, the calculation method of the model complexity function is expressed as:

[0050] Where W p,ik and represents the kth parameter in the weight parameter combination of the i-th species.

[0051] In one embodiment, the embodiment of the present invention adopts an adaptive energy adjustment mechanism to dynamically adjust the variation energy of each species (parameter set) according to its historical performance. By adjusting the energy input at each variation, species with lower adaptability can get greater adjustment opportunities, while species with higher adaptability can maintain a relatively stable state, thereby increasing the global search capability of the algorithm and avoiding local optimality. Specifically, the calculation method of defining the energy adjustment factor is expressed as:

[0052] In the formula, is the energy regulation factor of the i-th species; H p,i represents the historical fitness of the i-th species, that is, the accumulated fitness score before this iteration; is the sensitivity parameter of energy regulation, is the average of the accumulated fitness scores before this iteration. Preferably, Set to 0.1.

[0053] Step S206, calculating the dynamic mutation rate corresponding to the network parameter based on the energy adjustment factor.

[0054] Step S208, migrating the initial ecological niche based on the dynamic mutation rate, thereby updating the network parameters.

[0055] Specifically, the embodiment of the present invention calculates the variation energy corresponding to the network parameters according to the energy adjustment factor and the preset maximum energy input, calculates the adaptability score based on the variation energy and the preset migration and variation adjustment coefficients, and determines the dynamic variation rate corresponding to the network parameters.

[0056] The calculation of the variation energy is expressed as:

[0057] In the formula, E p,max is the maximum energy input, is a parameter orthogonal term. Preferably, E p,max Set to 0.4.

[0058] Furthermore, the parameters are orthogonalized by using the environment-aware parameter orthogonalization adjustment mechanism to ensure that appropriate diversity is maintained between different parameter sets (species), thereby avoiding premature convergence to local optimality and enhancing global search capabilities. Specifically, by using the orthogonalization adjustment of parameters in the process of adjusting the ecological niche, it is ensured that the parameter update process not only responds to the current adaptability, but also maintains the independence between parameters, thereby promoting the algorithm to explore unknown and potentially better solution space. The calculation method of orthogonalization energy is expressed as:

[0059] In the formula, Ortho p,i represents the orthogonalized energy of the i-th species; represents the vector dot product, represents the L2 norm; W p,j represents the weight and bias of the j-th species.

[0060] Furthermore, the orthogonal term of the parameter is calculated based on the orthogonalized energy, and the calculation method is expressed as:

[0061] In the formula, is an orthogonally adjusted learning rate. Preferably, Set to 0.01.

[0062] The simulated species migrate and mutate parameters according to their adaptability. Species with high adaptability have a smaller variation rate and remain stable; species with low adaptability try larger variations to seek improvement of ecological niche. Specifically, the migration and variation of parameters depends on the adaptability score of the species, which is calculated as follows:

[0063]

[0064] In the formula, Represents the weight of the neural network after migration and mutation, which is used as the weight of the neural network for the next iteration; It is a dynamic mutation rate based on fitness, which reflects the inverse of the fitness of the species. The lower the fitness, the higher the mutation rate. represents migration and mutation influencing factors; is the migration and variation adjustment coefficient; is the variation energy. Preferably, Set to 0.1, Set to 0.3.

[0065] Simulate the niche competition strategy, that is, the competition between species. Species with low loss function (high fitness) will occupy a better niche, while species with low fitness may be forced to migrate to a poorer niche or be eliminated, thereby achieving niche migration, which corresponds to the change of neural network weights and bias parameters, expressed as:

[0066]

[0067] In the formula, and are the updated weights and biases of the neural network, ΔW p,i and Δb p,i Update increments for neural network weights and biases.

[0068] In one embodiment, the calculation method of the update increment of the neural network weights and biases is expressed as:

[0069]

[0070] In the formula, is the learning rate of the neural network; is the gradient of the fitness score of the ith species with respect to the weight parameter; is the gradient of the fitness score of the i-th species with respect to the bias parameter. Preferably, Set to 0.01.

[0071] Further, the above steps are repeated until the preset stop iteration condition is met, indicating that the model training is completed. In one embodiment, the preset stop iteration condition is reaching a preset maximum number of iterations, preferably, the preset maximum number of iterations is set to 1000 times.

[0072] In summary, the embodiments of the present invention dynamically adjust the weight and bias parameters through an adaptive ecological optimization algorithm, and compare the parameter optimization process of the neural network to the niche adaptation process of species in the ecosystem, thereby improving the stability and global search capability of the network, reducing the occurrence of gradient vanishing, gradient explosion and local optimal problems, and improving the accuracy and efficiency of feature extraction.

[0073] Furthermore, the embodiment of the present invention also performs dimensionality reduction processing on the target features through a pre-built dimensionality reduction model. After dimensionality reduction, the number of features is reduced, and the computing resources required for the training and reasoning process of the model are significantly reduced, which accelerates the training and prediction speed of the model. The low-dimensional data after dimensionality reduction is easier to visualize, helping researchers and engineers to understand the distribution and structure of the data more intuitively. The features after dimensionality reduction can better reflect the essential structure of the data, help the model to better generalize to unseen data, and improve the robustness and accuracy of the model. Moreover, high-dimensional data can easily lead to overfitting of the model. Dimensionality reduction can reduce the number of features and reduce the complexity of the model, thereby reducing the risk of overfitting.

[0074] Traditional feature dimensionality reduction methods mainly rely on statistical methods or linear transformations, and it is difficult to effectively distinguish which features are important and which are redundant. This may lead to inaccurate feature representation after dimensionality reduction, loss of key information or introduction of irrelevant information. In addition, traditional dimensionality reduction methods are usually fixed and cannot be adjusted according to the characteristics of a specific data set, lacking flexibility and adaptability. The embodiment of the present invention adopts an autoencoding neural network based on feature refinement as a dimensionality reduction model, and can dynamically adjust the importance of each feature by recursively optimizing the feature weight vector. This method can flexibly determine the weight of each feature based on the actual distribution of the data and the relationship between the features. The autoencoding neural network can more accurately identify which features are important and which are redundant.

[0075] Among them, the autoencoding neural network based on feature refinement of the embodiment of the present invention is composed of three parts: an encoder, a decoder, and a feature adjustment module. The encoder is responsible for mapping high-dimensional input data to a low-dimensional feature space, and the decoder is used to reconstruct the reduced-dimensional features back to the original space to ensure the reversibility of the dimensionality reduction process. The feature adjustment module dynamically adjusts the reduced-dimensional feature space through recursive feature adaptive optimization, so that important features can be strengthened and secondary features are gradually weakened, so that the reduced-dimensional feature representation has better simplicity while retaining data information. Correspondingly, Figure 3 A schematic diagram of the training process of the dimensionality reduction model is shown.

[0076] Specifically, the method of constructing the dimensionality reduction model is as follows: 1) Input the preset training sample set into the preset autoencoder neural network, and map the training sample set to the initial low-dimensional feature space through the encoder of the autoencoder neural network to obtain the initial low-dimensional features.

[0077] First, let the data input to the autoencoder neural network be X r , the encoder uses a multi-layer nonlinear mapping structure to map high-dimensional data to the initial low-dimensional feature space, which can be expressed as:

[0078] In the formula, Z r represents the initial low-dimensional features, W r is the weight matrix of the encoder, b r is the bias vector of the encoder, Sig enc is the multi-layer Sigmoid activation function of the encoder.

[0079] 2) Based on the loss function value corresponding to the initial low-dimensional feature, determine the feature importance of the initial low-dimensional feature.

[0080] 3) Based on the feature importance, the feature weight vector of the autoencoder neural network is recursively optimized.

[0081] After the low-dimensional features are generated, the feature adjustment module automatically generates feature weights according to the feature importance in the current feature space. When the module is initialized, the same initial weight is given to all features so that they can be gradually adjusted according to the feature contribution in subsequent steps. The calculation method of the initial weight matrix is ​​expressed as:

[0082] In the formula, A r is the initial weight matrix; is the feature weight vector, Each element in Initialized to the same value, indicating that all features have the same importance in the initial stage; diag is a function for extracting the diagonal elements of a matrix.

[0083] Furthermore, the adjusted features can be expressed as:

[0084] In the formula, is the feature representation after feature weight adjustment.

[0085] The feature adjustment module recursively optimizes the initially generated low-dimensional features. In each iteration, the module adjusts the weights of each feature according to the performance of the previous round of features, gradually enhances those features with important influences, and gradually weakens redundant or noise features. In the round iteration, the weight update rule is as follows:

[0086] In the formula, represents the feature weight of the t+1th iteration, represents the feature weight of the tth iteration, is the learning rate of the autoencoder neural network, L r is the loss function of the autoencoder neural network, Y r is the label data, represents the gradient of the loss function with respect to the feature weight. Preferably, the loss function of the autoencoder neural network is the reconstruction error loss function. Preferably, Set to 0.01.

[0087] 4) Based on the feature weight vector, new low-dimensional features are determined, and the new low-dimensional features are reconstructed through the decoder.

[0088] 5) Until the autoencoder neural network meets the preset iteration conditions, a dimensionality reduction model is constructed based on the autoencoder neural network.

[0089] To ensure the effectiveness of the dimensionality reduction process, the decoder remaps the low-dimensional features back to the high-dimensional space to ensure that the dimensionality reduction process does not lose important information. The reconstruction process of the decoder is expressed as:

[0090] In the formula, is the reconstructed high-dimensional data, is the weight matrix of the decoder, is the bias vector of the decoder, Sig dec is the multi-layer Sigmoid activation function of the decoder.

[0091] Repeat the above steps until the preset stop iteration condition is met, which means that the model training is completed. In one embodiment, the preset stop iteration condition is reaching a preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times.

[0092] In summary, the embodiment of the present invention adopts an autoencoding neural network based on feature refinement to map high-dimensional data to a low-dimensional feature space, and gradually strengthens important features and weakens redundant features by recursively optimizing feature weights, thereby ensuring effective dimensionality reduction while retaining key information, effectively reducing irrelevant and redundant information, and making the feature representation after dimensionality reduction more concise while retaining the key information of the data.

[0093] Furthermore, the embodiment of the present invention also describes a method for constructing a classifier model. In combination with the above embodiments, the prior art is limited by its processing power and expression ability when processing complex data structures, and it is difficult to fully capture the complex relationships in the data, and the classification accuracy is insufficient. The embodiment of the present invention adopts a deep neural decision tree algorithm based on quantum entangled state as a classifier model. Deep Neural Decision Tree (DNDT) is a model that combines the advantages of traditional decision trees and deep neural networks. It embeds the structure of the decision tree into the neural network, so that the powerful representation ability and optimization method of the neural network can be used to train the model. The middle layer of the deep neural decision tree consists of multiple neurons, each of which corresponds to a node of the decision tree. These neurons are nonlinearly transformed by activation functions (such as ReLU). At each internal node (decision node), the data is determined according to the eigenvalue and the weight of the node (similar to the threshold in the decision tree). The final data will reach a leaf node, and the output of the leaf node is the final prediction result. The embodiment of the present invention combines the entangled state of quantum bits to simulate the multi-dimensional representation of information, thereby enhancing the ability of the classifier to process complex data structures. Specifically, the classifier is constructed as follows: 1) Simulate the information gain calculation during the decision tree splitting process to determine the state of the qubit.

[0094] a-Initialize the nodes of the decision tree of the preset deep neural decision tree algorithm as qubits.

[0095] Initialize the qubits of the model. Each qubit represents a node in the decision tree. The initial state is the maximum superposition state to ensure that each feature dimension can be considered fairly in the early stage of training. Specifically, set the initial state of the qubit to , the calculation method is expressed as:

[0096] Where n u represents the number of qubits, is the base state of the data.

[0097] b- Train through the deep neural decision tree algorithm of the preset training sample set, and determine the rotation parameters of the custom revolving door corresponding to the training sample set through exponential operation in the complex field.

[0098] c- Adjust the entanglement and phase between qubits based on the rotation parameters and preset quantum logic gate operations.

[0099] During the training process, quantum logic gates are used to adjust the entanglement and phase between qubits according to the training data to simulate the information gain calculation in the decision tree splitting process. Specifically, the quantum logic gate is set to , including the Pauli-X gate, CNOT gate, and custom rotation gate to simulate the calculation of information gain, expressed as:

[0100] In the formula, For custom revolving doors; It is a rotation parameter that is dynamically adjusted according to the data, representing the rotation angle on the Bloch sphere, which is used to adjust the entanglement and phase relationship of the quantum bit and enhance the nonlinear decision-making ability of the model, and, is the first rotation parameter, is the second rotation parameter, is the third rotation parameter; CNOT represents the CNOT gate; Pauli-X represents the Pauli-X gate.

[0101] In one embodiment, the custom rotation gate realizes the rotation of quantum bits through exponential operations in the complex field, allowing the information gain process of the decision tree to be simulated in a high-dimensional feature space. The calculation method is expressed as:

[0102] Where, X us and Z us The Pauli matrix represents the rotation operation on the X and Z axes; Is an imaginary unit.

[0103] d- Based on the entanglement and phase between the adjusted qubits, quantum measurements are performed on the qubits to determine the probability distribution corresponding to the qubits to evaluate the state of the qubits.

[0104] After each iteration, the qubits are measured to obtain the specific state of each node. This state information is used to determine the splitting conditions in the decision tree, such as the optimization of information gain or Gini impurity, expressed as:

[0105] Where M uIt is a quantum measurement operation, which extracts the projection probability of a quantum state on a specific ground state to evaluate whether the current node meets a certain splitting condition; represents the probability distribution obtained after the measurement of the tth iteration, which is used to evaluate the state of the current qubit and thus determine the splitting path that maximizes the information gain; is the qubit of the tth iteration; is the conjugate of the qubit at the tth iteration.

[0106] In one embodiment, the quantum measurement operation is a measurement operation based on a specific decision rule. According to the logical rules of the decision tree nodes and the state superposition and entanglement characteristics in quantum computing, the splitting condition can be evaluated more accurately. The matrix representation is as follows:

[0107] In the formula, and It is a base state projection operation. The purpose of the base state projection operation is to determine whether the node state meets the logical conditions of the decision tree node through a specific base state. For example, a node needs to check whether the data meets a certain feature threshold, which is expressed by the weight of the projection operation in the direction of a specific base vector, and is the first basis state projection operation, is the second basis state projection operation; I is the identity matrix, Z is the Pauli-Z matrix; is matrix multiplication.

[0108] 2) Generate decision branches of the decision tree based on the state of the qubit.

[0109] Among them, according to the measurement results and training data, node splitting is performed to generate new decision branches. In the specific implementation, the following steps are included: a-Calculate the decision conditions corresponding to the training sample set based on the state of the qubit.

[0110] In one embodiment, the decision condition calculated based on the state of the qubit and the training data is based on a nonlinear mapping to adapt to the complex data characteristics, and the calculation method is expressed as:

[0111] Where D u is the decision condition calculated based on the state of the quantum bit and the training data, used for the node splitting decision. u is the number of qubits, w uj is the quantum state weight coefficient of the jth node, Sig is the Sigmoid activation function; x uj is the split data attribute of the jth node; is the first rotation parameter of the jth node.

[0112] b- Based on the decision conditions, the node splitting operation of the decision tree is controlled by a preset decision function to generate the decision branches of the decision tree.

[0113] The superposition and entanglement capabilities of quantum states enable the algorithm to partition data more efficiently in high-dimensional space. Specifically, the node splitting operation is controlled by a decision function, which is executed according to the probability threshold adjusted by the revolving door, expressed as:

[0114] In the formula, is the threshold value determined after information gain or Gini impurity optimization; Split is the splitting behavior. Preferably, Set to 0.04.

[0115] 3) Conduct model evaluation.

[0116] Among them, after the node splitting is completed, the classification effect of the current model is evaluated, and the quantum gate parameters are adjusted according to the evaluation results to optimize the entire model, which is expressed as:

[0117]

[0118]

[0119] In the formula, is the learning rate of the deep neural decision tree, Represents the loss function Gradients with respect to parameters; is the first rotation parameter of the tth iteration, is the second rotation parameter of the tth iteration, is the third rotation parameter of the tth iteration; is the first rotation parameter of the t+1th iteration, is the second rotation parameter of the t+1th iteration, is the third rotation parameter of the t+1th iteration. Preferably, Set to 0.2.

[0120] Repeat the above steps until the preset stop iteration condition is met, which means that the model training is completed. In one embodiment, the preset stop iteration condition is reaching a preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 500 times.

[0121] In summary, the embodiment of the present invention uses a deep neural decision tree algorithm based on quantum entanglement to make classification decisions. By using the superposition and entanglement characteristics of quantum bits to simulate the multi-dimensional representation of complex data, it can efficiently process multi-dimensional complex data. And by optimizing the node splitting conditions of the decision tree through quantum logic gate operations, the node splitting process is optimized, which can improve the classification ability of the classifier under complex data structures.

[0122] Furthermore, based on the above embodiment, the embodiment of the present invention also provides another model-driven device health level assessment method, which mainly describes the method for constructing a training sample set. Figure 4 A flow chart of a method for constructing a training sample set provided by an embodiment of the present invention is shown. Figure 4 As shown, the method comprises the following steps: Step S10, obtaining pre-collected equipment operation monitoring samples.

[0123] Step S11, based on the equipment health status represented by the operation monitoring samples, the operation monitoring samples are labeled, and the text data in the operation monitoring samples are converted into a vector format to construct an initial sample set.

[0124] The equipment operation monitoring sample of the embodiment of the present invention includes the operation log and sensor data of the preset equipment, and the contents of the above embodiment can be referred to. Further, the collected data is annotated for training the model. In one embodiment, the annotation method can be manual annotation, and the annotated categories include: health, warning, and fault, a total of 3 categories.

[0125] Step S12, using a pre-built data expansion model to perform data expansion on the initial sample set to generate expanded samples of the initial sample set.

[0126] It is understandable that in the task of the present invention, the collection, acquisition, labeling and preprocessing of training data are time-consuming and labor-intensive, and insufficient training samples can easily lead to poor generalization ability of the model, while affecting the accuracy of the model. However, traditional generative adversarial networks often find it difficult to generate accurate samples of relatively scarce categories when faced with skewed data distribution, resulting in data imbalance and affecting the generalization performance of the model. The data augmentation model of an embodiment of the present invention includes a generative adversarial network algorithm, and the generator parameters and discriminator parameters of the generative adversarial network algorithm are adjusted based on the dynamic strategy of game theory, which can achieve a trade-off between the generator and the discriminator during the confrontation process, so that each generated data better simulates the real data distribution. Specifically, the following steps are included: 1) The feature weights of the preset training sample set are adjusted through the adaptive feature transformation strategy.

[0127] In one embodiment, the quality of generated samples is improved and the model's learning ability for different data is enhanced through an adaptive feature transformation strategy. Based on the dynamic weighting and adjustment of the input features of the generator, feature weight learning is achieved. The calculation method of the feature transformation weight is expressed as:

[0128] In the formula, Sig is the Sigmoid activation function; V c It is the parameter matrix of weight learning, specifically the weight parameter matrix of the generator.

[0129] 2) Through the generator of the generative adversarial network algorithm, the training sample set after feature weight adjustment is expanded to generate initial expanded samples.

[0130] Let the generator be G c , the discriminator is D c , set the weight of the generator to W Gc , the weight of the discriminator is W Dc In one embodiment, the parameters of the generator and the discriminator are initialized as follows:

[0131]

[0132] In the formula, is a normal distribution with mean 0 and standard deviation as the identity matrix I; To obey a specific distribution.

[0133] In the generation phase, the generator generates a batch of new data samples based on the current network parameters. These samples initially try to imitate the distribution characteristics of real data, which can be expressed as:

[0134] In the formula, z c is the latent vector sampled from the prior distribution; G c is the generator function, is the generator weight at the tth iteration; is the sample generated by the generator in the tth iteration; W Tc Represents the weight of feature transformation; Represents element-wise multiplication.

[0135] 3) Evaluate the initial expanded samples through the discriminator of the generative adversarial network algorithm to obtain the evaluation results.

[0136] The discriminator evaluates the generated data samples and the real data samples, and outputs the probability that each sample is a real sample, expressed as:

[0137] In the formula, Indicates that the generated samples and real samples are merged into one batch input; D c is the discriminator function, is the discriminator weight at the tth iteration; is the discrimination result of the discriminator in the tth iteration.

[0138] 4) Calculate the loss function of the generator based on the evaluation results and the entropy of the initial augmented samples; and calculate the loss function of the discriminator based on the preset regularization term.

[0139] The loss function of the generator in the embodiment of the present invention is comprehensively calculated in combination with the entropy of the generated data to more comprehensively evaluate the quality of the generated data and increase the diversity of the generated samples, which is expressed as:

[0140] In the formula, H c represents the entropy function, is the adjustment coefficient that controls the influence of the entropy term.

[0141] Furthermore, the loss function of the discriminator uses a regularization term to prevent overfitting, which is expressed as:

[0142] In the formula, y c Represents the real sample label (1 is the real sample, 0 is the generated sample); is the regularization coefficient, which is used for weight decay to reduce model complexity and overfitting risk; Represents the square of the L2 norm of the discriminator weights.

[0143] 5) Based on the loss function of the generator and the loss function of the discriminator, and the preset dynamic game strategy, the generator parameters of the generator and the discriminator parameters of the discriminator are adjusted respectively.

[0144] The learning rate and update strategy of the generator and discriminator are adjusted using a dynamic game strategy. The loss function is dynamically adjusted according to the results of each round of confrontation, giving priority to improving the ability to generate scarce category data, which can be expressed as:

[0145]

[0146] In the formula, is the generator weight at the t+1th iteration; is the discriminator weight at the t+1th iteration; is the learning rate of the generator; is the learning rate of the discriminator; is the loss function of the generator; is the loss function of the discriminator; X rc Samples generated by the generator; represents the gradient with respect to the generator weights; Represents the weights about the discriminator parameters. The goal of the generator is to maximize the loss function, so gradient ascent is used to update the parameters. The goal of the discriminator is to maximize its loss function and gradient descent is used to update the parameters. The generator continuously adjusts its parameters to generate more realistic samples. In each iteration, the loss function of the generator changes with the quality of the samples generated by the generator. The goal of the generator is to make the loss function as large as possible, that is, to make the discriminator think that the generated samples are real. The loss function of the discriminator changes with the quality of the samples generated by the generator and the real samples. The goal of the discriminator is to maximize the loss function to correctly identify real samples and generated samples.

[0147] During the adversarial training of the generator and the discriminator, the dynamic game strategy dynamically adjusts the adjustment coefficient and regularization coefficient that control the influence of the entropy term so that each generated data better simulates the real data distribution, which is expressed as:

[0148]

[0149] In the formula, is the learning rate adjustment factor, which is used to control the dynamic adjustment of the learning rate; is the symbol of partial derivative; is the adjustment coefficient of the control entropy term at the tth iteration, is the regularization coefficient of the tth iteration; is the adjustment coefficient of the control entropy term at the tth iteration, is the regularization coefficient for the tth iteration. Preferably, Set to 0.01.

[0150] Repeat the above steps until the preset stop iteration condition is met, indicating 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. After the data expansion model training is completed, the number of samples is increased using the trained data expansion model. In one embodiment, assuming that the original collected samples are 800, and the data expansion model expands and generates 200 samples, the expanded data set contains 1000 samples.

[0151] Step S13: construct a training sample set based on the expanded samples and the initial sample set.

[0152] In summary, the embodiment of the present invention adopts a generative adversarial network algorithm based on dynamic game. By adjusting the dynamic strategies of the generator and the discriminator and combining the entropy function to evaluate the quality and diversity of the generated data, it can give priority to generating data of scarce categories, ensure the balance of data expansion, improve the generalization ability and training accuracy of the model, and effectively solve the problem of insufficient training data.

[0153] Furthermore, based on the above embodiment, the embodiment of the present invention also provides a device health level assessment device based on model driving. Figure 5 A schematic diagram of a device health level assessment apparatus based on model drive provided by an embodiment of the present invention is shown. Figure 5 The device includes: a data acquisition module 100, which is used to acquire the operation monitoring data of the preset device, convert the text data in the operation monitoring data into a vector format, and generate a sample to be tested of the preset device; the operation monitoring data is collected from the operation log and sensor data of the preset device; a data processing module 200, which is used to extract features of the sample to be tested through a pre-built feature extraction model to obtain the target features corresponding to the sample to be tested; wherein the network parameters of the feature extraction model are updated based on the assigned ecological niche, and the ecological niche of the network parameters is determined by simulating the competition strategy of species in the ecosystem; an execution module 300, which is used to input the target features into a preset classifier model, classify and decide the target features through the classifier model, and output the decision results; wherein the classifier model is based on a deep neural decision tree algorithm as a classifier algorithm, the nodes of the decision tree of the deep neural decision tree algorithm are initialized as quantum bits, and the decision branches of the decision tree are determined based on the state of the quantum bits, and the state of the quantum bits is determined by simulating the information gain calculation in the decision tree splitting process; an output module 400, which is used to evaluate the health status of the preset device based on the decision results.

[0154] A model-driven equipment health level assessment device provided in an embodiment of the present invention has the same technical features as a model-driven equipment health level assessment method provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0155] Furthermore, the embodiment of the present invention also provides another device health level assessment apparatus based on model drive, Figure 6 FIG. 4 shows a schematic diagram of another device health level assessment apparatus based on model drive provided by an embodiment of the present invention. Figure 6 As shown, the output module 400 is further used to obtain a decision tag indicated by a decision result; and determine the device state indicated by the decision tag as a health state of a preset device.

[0156] The above-mentioned execution module 300 is also used to initialize the nodes of the decision tree of the preset deep neural decision tree algorithm as quantum bits; train the deep neural decision tree algorithm through the preset training sample set, and determine the rotation parameters of the custom rotation gate corresponding to the training sample set through exponential operations in the complex field; adjust the entanglement and phase between the quantum bits based on the rotation parameters and the preset quantum logic gate operation; based on the adjusted entanglement and phase between the quantum bits, perform quantum measurement on the quantum bits, determine the probability distribution corresponding to the quantum bits, so as to evaluate the state of the quantum bits.

[0157] The above-mentioned execution module 300 is also used for the step of generating a decision branch of a decision tree based on the state of the quantum bit, including: calculating the decision condition corresponding to the training sample set according to the state of the quantum bit; based on the decision condition, controlling the node splitting operation of the decision tree through a preset decision function to generate a decision branch of the decision tree.

[0158] The above-mentioned data processing module 200 is also used to train a preset feature extraction model through a preset training sample set; wherein, the network parameters of the feature extraction model are assigned an initial ecological niche; the loss function value corresponding to the network parameters of the feature extraction model is calculated, and the loss function value is used as the fitness score corresponding to the network parameters, and the energy adjustment factor corresponding to the network parameters is calculated; based on the energy adjustment factor, the dynamic mutation rate corresponding to the network parameters is calculated; based on the dynamic mutation rate, the initial ecological niche is migrated, thereby updating the network parameters.

[0159] The above-mentioned data processing module 200 is also used to calculate the variation energy corresponding to the network parameters according to the energy adjustment factor and the preset maximum energy input; based on the variation energy and the preset migration and variation adjustment coefficients, the adaptability score is calculated to determine the dynamic variation rate corresponding to the network parameters.

[0160] The above-mentioned data processing module 200 is also used to reduce the dimension of the target feature through a pre-constructed dimension reduction model; wherein, the method for constructing the dimension reduction model includes: inputting a preset training sample set into a preset autoencoding neural network, mapping the training sample set to an initial low-dimensional feature space through the encoder of the autoencoding neural network to obtain the initial low-dimensional feature; determining the feature importance of the initial low-dimensional feature based on the loss function value corresponding to the initial low-dimensional feature; recursively optimizing the feature weight vector of the autoencoding neural network based on the feature importance; determining new low-dimensional features based on the feature weight vector, and reconstructing the new low-dimensional features through the decoder; until the autoencoding neural network meets the preset iteration conditions, constructing a dimension reduction model based on the autoencoding neural network.

[0161] The device also includes a construction module 500, which is used to obtain pre-collected equipment operation monitoring samples; the equipment operation monitoring samples include operation logs and sensor data of preset equipment; based on the equipment health status represented by the operation monitoring samples, the operation monitoring samples are labeled, and the text data in the operation monitoring samples are converted into a vector format to construct an initial sample set; the initial sample set is expanded using a pre-built data expansion model to generate expanded samples of the initial sample set; wherein the data expansion model includes a generative adversarial network algorithm, and the generator parameters and discriminator parameters of the generative adversarial network algorithm are adjusted based on the dynamic strategy of game theory; a training sample set is constructed based on the expanded samples and the initial sample set.

[0162] The above-mentioned construction module 500 is also used to adjust the feature weights of a preset training sample set through an adaptive feature transformation strategy, and to perform data expansion on the training sample set after the feature weights are adjusted through the generator of the generative adversarial network algorithm to generate initial expanded samples; to evaluate the initial expanded samples through the discriminator of the generative adversarial network algorithm to determine the initial loss value of the generator; to calculate the loss function of the generator according to the initial loss value and the entropy of the initial expanded samples; and to calculate the loss function of the discriminator based on a preset regularization term; and to adjust the generator parameters of the generator and the discriminator parameters of the discriminator based on the loss function of the generator and the loss function of the discriminator, as well as the preset dynamic game strategy.

[0163] An embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above Figures 1 to 4 The embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to perform the above Figures 1 to 4 The embodiment of the present invention also provides a structural diagram of an electronic device, such as Figure 7 FIG. 1 is a schematic diagram of the structure of the electronic device, wherein the electronic device includes a processor 71 and a memory 70, the memory 70 stores computer executable instructions that can be executed by the processor 71, and the processor 71 executes the computer executable instructions to implement the above Figures 1 to 4 Any of the methods shown. Figure 7 In the illustrated embodiment, the electronic device further includes a bus 72 and a communication interface 73 , wherein the processor 71 , the communication interface 73 and the memory 70 are connected via the bus 72 .

[0164] Among them, the memory 70 may include a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 73 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 72 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, on-chip bus standard) bus, wherein AMBA defines three types of buses, including APB (Advanced Peripheral Bus) bus, AHB (Advanced High-performance Bus) bus and AXI (Advanced eXtensible Interface) bus. The bus 72 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7Only one bidirectional arrow is used in the diagram, but it does not mean that there is only one bus or one type of bus. The processor 71 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 71. The above-mentioned processor 71 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor, or a combination of hardware and software modules in the decoding processor. The software module can be located in a random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register or other mature storage media in the art. The storage medium is located in the memory, and the processor 71 reads the information in the memory and completes the above-mentioned Figures 1 to 4 Any of the methods shown.

[0165] A computer program product of a model-driven equipment health level assessment method and apparatus provided in an embodiment of the present invention includes a computer-readable storage medium storing a program code, and the instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can refer to the method embodiment, which will not be repeated here. A person skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system described above can refer to the corresponding process in the above method embodiment, which will not be repeated here. In addition, in the description of the embodiment of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood in specific circumstances. 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 understanding, the technical solution of the present invention, in essence, 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, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk, and other media that can store program codes. In the description of the present invention, it should be noted that the orientation or position relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside", etc. is based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.Finally, it should be noted that the above embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions recorded in the above embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A model-driven equipment health level assessment method, characterized in that: The method comprises: Acquire operation monitoring data of a preset device, convert text data in the operation monitoring data into a vector format, and generate a sample to be tested of the preset device; the operation monitoring data is collected from the operation log and sensor data of the preset device; Extracting features of the sample to be tested by a pre-built feature extraction model to obtain target features corresponding to the sample to be tested; wherein the network parameters of the feature extraction model are updated based on the assigned ecological niche, and the ecological niche of the network parameters is determined by simulating the competition strategies of species in the ecosystem; The target feature is input into a preset classifier model, and a classification decision is made on the target feature through the classifier model, and a decision result is output; wherein the classifier model is based on a deep neural decision tree algorithm as a classifier algorithm, the nodes of the decision tree of the deep neural decision tree algorithm are initialized as qubits, and the decision branches of the decision tree are determined based on the state of the qubit, and the state of the qubit is determined by calculating the information gain in the process of simulating the splitting of the decision tree; Based on the decision result, the health status of the preset device is evaluated.

2. The method according to claim 1, characterized in that Based on the decision result, the step of evaluating the health status of the preset device includes: Obtaining a decision label indicated by the decision result; The device state indicated by the decision tag is determined as the health state of the preset device.

3. The method according to claim 1, characterized in that: The information gain calculation in the simulation decision tree splitting process determines the steps of the state of the quantum bit, including: Initialize the nodes of the decision tree of the preset deep neural decision tree algorithm as qubits; Training is performed using a preset training sample set deep neural decision tree algorithm, and determining the rotation parameters of the custom revolving door corresponding to the training sample set through exponential operations in the complex domain; Adjusting the entanglement and phase between the qubits based on the rotation parameters and a preset quantum logic gate operation; Based on the adjusted entanglement and phase between the qubits, quantum measurements are performed on the qubits to determine the probability distribution corresponding to the qubits to evaluate the state of the qubits.

4. The method according to claim 3, characterized in that: The step of generating a decision branch of the decision tree based on the state of the qubit comprises: Calculating a decision condition corresponding to the training sample set according to the state of the qubit; Based on the decision condition, the node splitting operation of the decision tree is controlled by a preset decision function to generate a decision branch of the decision tree.

5. The method according to claim 1, characterized in that The steps of determining ecological niches based on the competition strategies of species in the simulated ecosystem and updating the network parameters of the feature extraction model based on the assigned ecological niches include: The preset feature extraction model is trained by using a preset training sample set; wherein the network parameters of the feature extraction model are assigned with initial ecological niches; Calculating the loss function value corresponding to the network parameter of the feature extraction model, using the loss function value as the adaptability score corresponding to the network parameter, and calculating the energy adjustment factor corresponding to the network parameter; Based on the energy adjustment factor, calculating the dynamic mutation rate corresponding to the network parameter; The initial ecological niche is migrated based on the dynamic mutation rate, thereby updating the network parameters.

6. The method according to claim 5, characterized in that The step of calculating the dynamic mutation rate corresponding to the network parameter based on the energy adjustment factor includes: Calculate the variation energy corresponding to the network parameter according to the energy adjustment factor and the preset maximum energy input; Based on the mutation energy and preset migration and mutation adjustment coefficients, the adaptability score is calculated to determine the dynamic mutation rate corresponding to the network parameter.

7. The method according to claim 1, characterized in that The method further comprises: Performing dimensionality reduction processing on the target feature through a pre-built dimensionality reduction model; The method for constructing the dimensionality reduction model comprises: Inputting a preset training sample set into a preset autoencoding neural network, mapping the training sample set to an initial low-dimensional feature space through an encoder of the autoencoding neural network to obtain an initial low-dimensional feature; Determining the feature importance of the initial low-dimensional feature based on the loss function value corresponding to the initial low-dimensional feature; recursively optimizing the feature weight vector of the autoencoder neural network based on the feature importance; Based on the feature weight vector, determine a new low-dimensional feature, and perform feature reconstruction on the new low-dimensional feature through a decoder; Until the autoencoding neural network meets the preset iteration condition, a dimensionality reduction model is constructed based on the autoencoding neural network.

8. The method according to claim 1, characterized in that: The method further comprises: Acquire pre-collected equipment operation monitoring samples; the equipment operation monitoring samples include operation logs and sensor data of preset equipment; Based on the equipment health status represented by the operation monitoring sample, the operation monitoring sample is annotated, and the text data in the operation monitoring sample is converted into a vector format to construct an initial sample set; Performing data expansion on the initial sample set using a pre-built data expansion model to generate expanded samples of the initial sample set; wherein the data expansion model includes a generative adversarial network algorithm, and the generator parameters and the discriminator parameters of the generative adversarial network algorithm are adjusted based on a dynamic strategy of game theory; A training sample set is constructed based on the expanded samples and the initial sample set.

9. The method according to claim 8, characterized in that The steps of adjusting the generator parameters and discriminator parameters of the generative adversarial network algorithm based on the dynamic strategy of game theory include: The feature weights of the preset training sample set are adjusted through the adaptive feature transformation strategy; Using the generator of the generative adversarial network algorithm, data expansion is performed on the training sample set after feature weight adjustment to generate initial expanded samples; Evaluate the initial expanded sample by using the discriminator of the generative adversarial network algorithm to obtain an evaluation result; Calculating the loss function of the generator according to the evaluation result and the entropy of the initial expanded sample; and calculating the loss function of the discriminator based on a preset regularization term; Based on the loss function of the generator and the loss function of the discriminator, and a preset dynamic game strategy, the generator parameters of the generator and the discriminator parameters of the discriminator are adjusted respectively.

10. A device for evaluating equipment health level based on model drive, characterized in that: The device comprises: A data acquisition module, used to acquire operation monitoring data of a preset device, convert text data in the operation monitoring data into a vector format, and generate a sample to be tested of the preset device; the operation monitoring data is collected from the operation log and sensor data of the preset device; A data processing module, used to extract features from the sample to be tested by using a pre-built feature extraction model to obtain target features corresponding to the sample to be tested; wherein the network parameters of the feature extraction model are updated based on the assigned ecological niche, and the ecological niche of the network parameters is determined by simulating the competition strategies of species in the ecosystem; An execution module, used for inputting the target feature into a preset classifier model, performing classification decision on the target feature through the classifier model, and outputting a decision result; wherein the classifier model is based on a deep neural decision tree algorithm as a classifier algorithm, the nodes of the decision tree of the deep neural decision tree algorithm are initialized as qubits, and the decision branches of the decision tree are determined based on the state of the qubit, and the state of the qubit is determined by calculating the information gain in the process of simulating the splitting of the decision tree; An output module is used to evaluate the health status of the preset device based on the decision result.

Images