Method and system for predicting wear resistance of nickel-based composite coating based on machine learning

Abstract

The present application relates to the technical field of performance prediction, especially to a nickel-based composite coating wear resistance prediction method and system based on machine learning, the present application integrates multi-source data such as workpiece surface characteristics, coating microstructure, process parameters and future working conditions, and constructs an intelligent prediction model based on multilayer perception, realizing accurate prediction and early warning of the wear life of the nickel-based composite coating, the method can learn complex nonlinear relationships from historical data, identify weak links in coating performance in advance, not only improve the predictability and accuracy of quality control, but also significantly reduce the equipment maintenance cost and downtime risk caused by coating failure, providing a reliable decision basis for optimizing the spraying process and prolonging the service life of the workpiece.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of performance prediction, in particular to a nickel-based composite coating wear resistance prediction method and system based on machine learning. BACKGROUND

[0002] In the field of modern industry, key equipment such as aircraft engine blades, heavy gas turbine components, and oil drilling tools have long been used in extreme harsh conditions such as high temperature, high pressure, and strong wear. The surface performance directly determines the service life and operational reliability of the entire equipment. In order to improve the wear and corrosion resistance of these key components, thermal spraying technology, especially high-velocity oxygen fuel and atmospheric plasma spraying technology, is widely used to prepare high-performance nickel-based composite coatings. These coatings can significantly improve the wear resistance of the workpiece surface by introducing high-hardness carbide phases.

[0003] However, the wear life of nickel-based composite coatings is not a constant value, but is influenced by multiple factors in a complex system. Specifically, its life depends on the following aspects: workpiece substrate characteristics: the geometry, surface roughness, preheating temperature, and material properties of the workpiece to be sprayed directly affect the bonding strength of the coating and the substrate, and are the basis for the durability of the coating system; coating microstructure: the porosity, crack density, oxide content, unmelted particle proportion, and distribution and degradation of carbide in the coating microstructure are the fundamental reasons for determining its macroscopic mechanical properties and wear behavior; spraying process parameters: during the spraying process, small fluctuations in dozens of process parameters such as spraying distance, powder feed rate, fuel flow, and power can significantly change the flight state, flattening behavior, and solidification process of the droplets, ultimately leading to significant differences in coating microstructure and performance;

[0004] Currently, the industry has the following main technical bottlenecks in evaluating and managing the wear resistance of coatings: traditional methods mainly rely on bench testing or destructive metallographic analysis of samples after spraying. This method is costly, time-consuming, and cannot detect every actual workpiece. Life prediction relies on the personal experience of engineers, lacks quantitative and scientific basis, and has poor accuracy and consistency; due to the complex and highly nonlinear mapping relationship between process parameters, microstructure, and final performance, existing technologies cannot accurately reveal the internal laws. Process optimization often requires a large number of "trial and error" experiments, which is inefficient and difficult to ensure the stability and reproducibility of coating performance; the existing technical system cannot accurately predict the wear life of coatings under specific working conditions before they are put into service; this leads to equipment maintenance usually using fixed cycles or after obvious failure occurs, which cannot achieve predictive maintenance based on the state, and may cause unplanned downtime due to unexpected coating failure, resulting in significant economic losses and safety risks.

[0005] Therefore, there is an urgent need for an intelligent method that can systematically integrate multi-source data and achieve high-precision forward prediction. SUMMARY

[0006] In order to overcome the defects and deficiencies existing in the prior art, the present application provides a nickel-based composite coating wear resistance prediction method and system based on machine learning. The present application integrates multi-source data such as workpiece surface features, coating microstructure, process parameters and future working conditions, and constructs an intelligent prediction model based on a multi-layer perception machine, thereby realizing accurate prediction and early warning of the wear life of the nickel-based composite coating. This method can learn complex nonlinear relationships from historical data, identify weak links in coating performance in advance, not only improve the predictability and accuracy of quality control, but also significantly reduce equipment maintenance costs and downtime risks caused by coating failure, and provide a reliable decision basis for optimizing the spraying process and extending the service life of the workpiece.

[0007] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0008] In a first aspect, the present application provides a nickel-based composite coating wear resistance prediction method based on machine learning, comprising the following steps:

[0009] S100, obtaining the workpiece surface condition and the nickel-based composite coating condition, and simultaneously obtaining the environmental condition during the coating process and the future use condition of the corresponding workpiece;

[0010] S200, performing workpiece coating connection analysis by the workpiece surface flatness in the workpiece surface condition and the hard phase distribution condition in the nickel-based composite coating condition;

[0011] S300, constructing a machine learning model, and performing workpiece coating wear resistance effect prediction by the workpiece coating connection analysis result, the average microhardness condition in the nickel-based composite coating condition, and the environmental condition during the coating process;

[0012] S400, obtaining the future use condition of the corresponding workpiece and the workpiece coating wear resistance effect prediction result to perform workpiece coating wear life prediction;

[0013] S500, performing coating performance early warning by the workpiece coating wear life prediction result.

[0014] In an implementation manner of the present application, the S100 comprises the following specific contents:

[0015] Step 110, obtaining the three-dimensional image of the workpiece surface and the nickel-based composite coating condition by the corresponding sensor, wherein the nickel-based composite coating condition comprises:

[0016] Hard phase distribution: the wear resistance of the nickel-based composite coating mainly comes from the uniformly dispersed hard phase therein; the hard phase plays a role in bearing load and resisting abrasive wear; its uniformity of distribution is a core index of coating quality; average microhardness: this is a direct measure of the coating's resistance to plastic deformation and indentation capacity;

[0017] Step 120, acquiring temperature, humidity and particle flow rate during the thermal spraying process through a sensor, wherein the temperature: affects the flattening behavior of the particles and the heat exchange with the substrate; the particle flow rate: determines the kinetic energy of the particles when they impact the substrate, affecting the density and bonding strength of the coating;

[0018] Step 130, obtaining the service friction load, service frequency and friction speed of the corresponding workpiece through the service log of the workpiece.

[0019] In an implementation manner of the present application, the workpiece coating connection analysis in S200 includes the following specific steps:

[0020] Step 210, acquiring a three-dimensional image of the workpiece surface, and then acquiring the pit distribution of the workpiece surface and the roughness of the corresponding workpiece surface, and acquiring the hard phase distribution in the coating, wherein the roughness of the workpiece surface is evaluated by: the average absolute value of each point on the workpiece surface relative to the reference surface height, the roughness is obtained by the ratio of the obtained average absolute value to the safety height, the roughness anomaly is obtained by the standard deviation of the roughness and the safety roughness range, the average absolute value of each point relative to the reference surface height is equivalent to the arithmetic average height in the roughness parameter, which is a direct, stable three-dimensional roughness measurement that can fully reflect the overall fluctuation of the surface, rather than a single profile; the normalization by the ratio to the safety height makes the evaluation standard adaptable to different process requirements or different types of workpieces, enhancing the universality and comparability of the method; instead of obtaining a roughness value, the standard deviation is calculated by comparing it with the safety roughness range to quantify the abnormality;

[0021] Step 220, acquiring the pit size of the workpiece surface and the size deviation of the hard phase particles to obtain the size matching anomaly, and acquiring the hard phase distribution uniformity anomaly from the hard phase distribution in the coating, wherein the hard phase distribution uniformity anomaly is obtained by: acquiring the hard phase distribution density of each position in the coating, acquiring the standard deviation of the hard phase distribution density of each position, and obtaining the hard phase distribution uniformity anomaly; if the hard phase particles are much larger than the surface pits, they will not be embedded; if they are much smaller, the embedding effect will be poor; the ideal condition is that the particle size and the pit size have an optimal matching range;

[0022] Step 230, the pit distribution of the workpiece surface is acquired, the standard deviation of the pit size is acquired as the pit size anomaly, and the spacing between the pits is acquired, and the standard deviation of the spacing is acquired as the pit distribution anomaly;

[0023] Step 240, the roughness anomaly, the size matching anomaly, the hard phase distribution uniformity anomaly, the pit size anomaly and the pit distribution anomaly are sequentially constructed into a data sequence, and the constructed sequence is set as the workpiece coating connection analysis sequence; a plurality of different dimensions of indexes describing the surface and the coating state are integrated into a unified and structured data sequence.

[0024] In an implementation manner of the present application, the workpiece coating wear-resistant effect prediction in the S300 specifically includes the following specific contents:

[0025] Step 310, the temperature, humidity and particle flow rate in the historical thermal spraying process are acquired; the historical workpiece coating connection analysis sequence, the average microhardness in the nickel-based composite coating condition and the wear-resistant life of the historical workpiece coating under the action of the standard friction frequency, the standard friction speed and the standard friction force are acquired, and a multilayer perception machine model with the input of the temperature, humidity, particle flow rate in the historical thermal spraying process, the average microhardness in the nickel-based composite coating condition and the workpiece coating connection analysis sequence and the output of the wear-resistant life of the workpiece coating under the action of the standard friction frequency, the standard friction speed and the standard friction force is constructed, wherein the standard friction frequency, the standard friction speed and the standard friction force are the output standard characteristics of the test equipment, the thermal spraying process is a complex physical and chemical process involving droplet spreading, solidification, phase change and the like, and it is difficult to accurately describe it with a pure physical model; the multilayer perception machine model is very suitable for learning such a complex nonlinear mapping relationship from a large amount of historical data;

[0026] Step 320, the temperature, humidity, particle flow rate in the real-time thermal spraying process, the average microhardness in the nickel-based composite coating condition and the workpiece coating connection analysis sequence are introduced into the constructed multilayer perception machine model, and the wear-resistant life of the workpiece coating under the action of the standard friction frequency, the standard friction speed and the standard friction force is outputted and predicted.

[0027] In an implementation manner of the present application, the workpiece coating wear-resistant life prediction in the S400 includes the following specific contents:

[0028] In step 410, the future usage friction force load, usage frequency, and friction speed of the workpiece are obtained, and the obtained usage friction force load, usage frequency, and friction speed are divided by the corresponding standard value to obtain a usage abnormality prediction result, wherein the corresponding standard value is the output standard feature of the test equipment, and the usage abnormality prediction result is multiplied by a wear resistance influence coefficient to obtain a wear resistance abnormality prediction result.

[0029] In step 420, the wear resistance life of the workpiece coating under the standard friction frequency, standard friction speed, and standard friction force is obtained by dividing the wear resistance abnormality prediction result by the wear resistance life of the workpiece coating predicted by the multi-layer perception model.

[0030] In an implementation manner of the present application, the coating performance early warning in S500 includes the following specific contents.

[0031] The obtained wear resistance life prediction result of the workpiece coating is compared with the corresponding wear resistance life requirement value of the workpiece coating, if the wear resistance life prediction result of the workpiece coating is greater than or equal to the corresponding wear resistance life requirement value of the workpiece coating, it means that the coating performance meets the workpiece use requirement, if the wear resistance life prediction result of the workpiece coating is less than the corresponding wear resistance life requirement value of the workpiece coating, it means that the coating performance does not meet the workpiece use requirement, and coating early warning is needed to remind the maintenance personnel to replace.

[0032] In a second aspect, the present application also provides a nickel-based composite coating wear resistance performance prediction system based on machine learning, which comprises:

[0033] A data acquisition module acquires the workpiece surface condition and the nickel-based composite coating condition, and simultaneously acquires the environment condition in the coating process and the future use condition of the corresponding workpiece.

[0034] A coating connection analysis module performs workpiece coating connection analysis by the workpiece surface flatness in the workpiece surface condition and the hard phase distribution condition in the nickel-based composite coating condition.

[0035] A machine learning analysis module constructs a machine learning model, and performs workpiece coating wear resistance effect prediction by the workpiece coating connection analysis result, the average microhardness condition in the nickel-based composite coating condition, and the environment condition in the coating process.

[0036] A life prediction module performs workpiece coating wear resistance life prediction by the future use condition of the corresponding workpiece and the workpiece coating wear resistance effect prediction result.

[0037] A performance early warning module performs coating performance early warning by the workpiece coating wear resistance life prediction result.

[0038] In a third aspect, the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program that can be invoked by the processor, and the processor executes the method for predicting the wear resistance of a nickel-based composite coating based on machine learning by invoking the computer program stored in the memory.

[0039] In a fourth aspect, the present application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to execute the method for predicting the wear resistance of a nickel-based composite coating based on machine learning.

[0040] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0041] By integrating multi-source data such as workpiece surface features, coating microstructure, process parameters, and future working conditions, and constructing an intelligent prediction model based on a multi-layer perception machine, the present application realizes accurate prediction and early warning of the wear life of a nickel-based composite coating. This method can learn complex nonlinear relationships from historical data, identify weak links in coating performance in advance, not only improve the predictability and accuracy of quality control, but also significantly reduce the cost of equipment maintenance and the risk of downtime caused by coating failure, and provides a reliable basis for decision-making for optimizing the spraying process and extending the service life of the workpiece. BRIEF DESCRIPTION OF DRAWINGS

[0042] Other features, objects, and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the accompanying drawings:

[0043] Figure 1 The figure is a schematic diagram of the overall flow structure of the method embodiment of the present application;

[0044] Figure 2 The figure is a schematic diagram of the flow structure of step S200 of the method embodiment of the present application;

[0045] Figure 3 The figure is a schematic diagram of the construction of the machine learning model of the method embodiment of the present application;

[0046] Figure 4 The figure is a schematic diagram of the module composition structure of the system embodiment of the present application. DETAILED DESCRIPTION

[0047] The technical solutions of the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solutions of the present application, and not limitations of the technical solutions of the present application. In the case of no conflict, the technical features in the embodiments of the present application and the embodiments can be combined with each other.

[0048] Please refer to Figures 1 to 3The embodiment of the application provides a nickel-based composite coating wear resistance prediction method based on machine learning, and specifically comprises the following steps:

[0049] S100, acquiring a workpiece surface condition and a nickel-based composite coating condition, simultaneously acquiring an environment condition in a coating process and a future use condition of the corresponding workpiece;

[0050] In the embodiment, the S100 comprises the following specific contents:

[0051] Step 110, acquiring a three-dimensional image of the workpiece surface and the nickel-based composite coating condition through the corresponding sensor, wherein the nickel-based composite coating condition comprises:

[0052] Hard phase distribution: the wear resistance of the nickel-based composite coating is mainly derived from the uniformly dispersed hard phase (such as tungsten carbide, chromium carbide, etc.) in the nickel-based composite coating; the hard phase plays a role in bearing load and resisting abrasive wear; the uniformity of the distribution is a core index of the quality of the coating, and uneven distribution will cause local premature wear; average microhardness: this is a direct measure of the resistance of the coating to plastic deformation and indentation capacity, and higher average hardness is usually related to better wear resistance, but it must be combined with toughness, and a coating with pure high hardness and brittleness is easy to peel off;

[0053] Step 120, acquiring the temperature, humidity and particle flow rate condition in the thermal spraying process through the sensor, wherein the temperature: affects the flattening behavior of the particles and the heat exchange with the matrix; the particle flow rate: determines the kinetic energy of the particles when they impact the matrix, and affects the compactness and bonding strength of the coating; these parameters jointly determine the microstructure of the coating, such as porosity, oxide content and residual stress, thereby directly affecting the wear resistance;

[0054] Step 130, acquiring the use friction force load, use frequency condition and friction speed condition of the corresponding workpiece through the use log of the workpiece, and then analyzing the wear load of the workpiece;

[0055] S200, performing workpiece coating connection analysis through the workpiece surface flatness in the workpiece surface condition and the hard phase distribution condition in the nickel-based composite coating condition.

[0056] In the embodiment, the workpiece coating connection analysis in the S200 comprises the following specific steps:

[0057] Step 210, obtain the three-dimensional image of the workpiece surface, and then obtain the pit distribution of the workpiece surface and the roughness of the workpiece surface, and obtain the hard phase distribution in the coating, wherein the evaluation method of the roughness of the workpiece surface is: obtaining the average absolute value of the height of each point on the workpiece surface relative to the reference surface, obtaining the roughness by the ratio of the obtained average absolute value to the safety height, obtaining the roughness anomaly by the standard deviation of the roughness and the safety roughness range, using the average absolute value of the height of each point relative to the reference surface, which is equivalent to the arithmetic average height in the roughness parameter, is a direct and stable three-dimensional roughness measurement, which can fully reflect the overall fluctuation of the surface, rather than a single profile; the normalization by the ratio of the safety height makes the evaluation standard adaptable to different process requirements or different types of workpieces, enhancing the universality and comparability of the method; instead of obtaining a roughness value, the standard deviation is calculated by the safety roughness range to quantify the abnormality; the basis of this step is surface metrology and industrial quality control standards. In thermal spraying, the roughness of the substrate surface directly affects the bonding strength of the coating. Moderate roughness can provide mechanical embedding effect, but excessive or insufficient roughness will lead to decreased bonding strength or coating defects.

[0058] Step 220, obtain the pit size of the workpiece surface and the size deviation of the hard phase particles to obtain the size matching anomaly, and obtain the hard phase distribution uniformity anomaly from the hard phase distribution in the coating, wherein the hard phase distribution uniformity anomaly is obtained by: obtaining the hard phase distribution density of each position in the coating, obtaining the standard deviation of the hard phase distribution density of each position, and obtaining the hard phase distribution uniformity anomaly; if the hard phase particles are much larger than the surface pits, they will not be embedded; if they are much smaller, the embedding effect will be poor; the ideal condition is that the particle size and the pit size have an optimal matching range; the two independent measurements (pit size and particle size) are related to form a quality index (matching degree) with clear physical meaning, which can predict whether the coating is easy to fall off due to poor mechanical embedding; the performance of the coating (such as hardness and wear resistance) requires uniform distribution of its components. Too little local hard phase will cause soft spots and easy wear; too much may cause stress concentration or peeling.

[0059] Step 230, obtain the pit distribution of the workpiece surface, obtain the standard deviation of the pit size as the pit size anomaly, and obtain the pit spacing condition, and obtain the pit distribution anomaly by obtaining the standard deviation of the pit spacing condition;

[0060] Step 240, constructing a data sequence in order according to the acquired roughness anomaly, size matching anomaly, hard phase distribution uniformity anomaly, pit size anomaly and pit distribution anomaly, setting the constructed sequence as a workpiece coating connection analysis sequence; integrating a plurality of different dimensions of indexes describing the surface and coating state into a unified and structured data sequence (or called feature vector); this sequence highly concentrates the key pre-process features affecting the coating bonding strength and final performance, providing high-quality input data for the subsequent prediction model;

[0061] S300, constructing a machine learning model to predict the workpiece coating wear resistance effect through the workpiece coating connection analysis result, the average microhardness condition in the nickel-based composite coating condition and the environment condition in the coating process;

[0062] In the embodiment, the workpiece coating wear resistance effect prediction in S300 specifically includes the following specific contents:

[0063] Step 310, acquiring the temperature, humidity and particle flow rate conditions in the historical thermal spraying process; at the same time, acquiring the historical workpiece coating connection analysis sequence, the average microhardness condition in the nickel-based composite coating condition and the wear life condition of the historical workpiece coating under the action of standard friction frequency, standard friction speed and standard friction force, constructing a multi-layer perception machine model with the input of the temperature, humidity, particle flow rate conditions in the historical thermal spraying process, the average microhardness condition in the nickel-based composite coating condition and the workpiece coating connection analysis sequence, and the output of the wear life condition of the workpiece coating under the action of standard friction frequency, standard friction speed and standard friction force, wherein the standard friction frequency, standard friction speed and standard friction force are the output standard features of the test equipment, the thermal spraying process is a complex physical and chemical process involving droplet spreading, solidification, phase change, etc., and it is difficult to accurately describe it with a pure physical model; the multi-layer perception machine model is very suitable for learning such a complex nonlinear mapping relationship from a large amount of historical data;

[0064] The specific steps of the multi-layer perception machine model are: input layer construction: directly corresponding to 9 input features, this layer does not do any calculation, only receives and transmits data to the first hidden layer, therefore, this layer has 9 neurons;

[0065] The hidden layers are preferably set to three layers. The first hidden layer, as the information receiving and preliminary processing layer, should have a relatively large number of neurons to capture enough initial feature combinations; it can be set to a large number, such as 128 neurons. The second hidden layer further refines and compresses the output of the first layer; the number of neurons is halved, set to 64 neurons. The third hidden layer performs a final high-precision feature extraction before the final decision; the number of neurons is halved again, set to 32 neurons. Through multi-layer nonlinear transformation, the input features are abstracted and combined layer by layer to learn the extremely complex mapping relationship between "process parameters" and "coating characteristics" and "wear resistance life". All three hidden layers use the ReLU (Modified Linear Unit) activation function.

[0066] The output layer has only one neuron, corresponding to the predicted wear life; activation function: since it is a regression prediction task (outputting continuous values), a linear activation function (i.e., without non-linear transformation) is usually used, or to ensure that the output is positive, functions such as Softplus can be used;

[0067] Loss Function: MSE is the standard loss function for regression tasks. It amplifies the penalty for large prediction errors, making the model training process focus on reducing predictions that deviate significantly from the true value. This is crucial for industrial quality prediction. Optimizer: Adam combines the advantages of other optimizers, automatically adjusting the learning rate. In most cases, it converges quickly and stably, making it the preferred optimizer in deep learning. Key Hyperparameter Settings: Learning Rate: This is the most important tuning knob. The initial value can be set to 0.001. If the loss decreases too slowly during training, it can be increased appropriately. If the loss oscillates without decreasing, it can be decreased appropriately. Batch Size: Choose a suitable value based on the amount of data and computing resources, such as 32 or 64. Number of Iteration Cycles: Set a large value (such as 1000) and use it in conjunction with early stopping techniques. That is, when the model's performance on the validation set no longer improves for several consecutive cycles, training is automatically stopped to prevent overfitting.

[0068] The training process: all the weights of the hidden layers and the output layer are initialized as small random numbers, and the bias is initialized as 0 or a small constant; take a batch (for example, 32) of historical workpiece data; the data flows from the input layer (6 neurons) to the first hidden layer (128 ReLU neurons) for calculation, and then flows into the second hidden layer (64 ReLU neurons), followed by the third hidden layer (32 ReLU neurons), and finally reaches the output layer (1 linear neuron) to obtain the predicted value of the wear life of the batch of workpieces; compare the predicted value of the model with the real and known wear life of the 32 workpieces, calculate a total error value by the MSE formula; use the chain rule of calculus to calculate the gradient of the loss function with respect to each weight and each bias in the network from the output layer; the Adam optimizer intelligently updates all weights and biases in the network according to the calculated gradient and the preset learning rate (0.001), while considering historical gradient information; repeat the training steps for the next batch of data, and stop training when the model performs best on the validation dataset; obtain the final multilayer perceptron model;

[0069] Step 320, the acquired temperature, humidity, particle flow rate, average microhardness of the nickel-based composite coating, and workpiece coating connection analysis sequence in the real-time thermal spraying process are introduced into the constructed multilayer perceptron model, and the predicted wear life of the workpiece coating under the action of standard friction frequency, standard friction speed and standard friction force is output. In this way, it can be analyzed whether the coating is qualified, that is, even if it cannot be used by this workpiece after S400-S500, it can be used by other workpieces after being judged to be qualified;

[0070] S400, acquiring the future use of the corresponding workpiece and the wear effect prediction result of the workpiece coating to predict the wear life of the workpiece coating;

[0071] In this embodiment, the workpiece coating wear life prediction in S400 includes the following specific contents:

[0072] Step 410, acquiring the use friction load, use frequency and friction speed of the corresponding workpiece in the future, weighting and summing the obtained use friction load, use frequency and friction speed after dividing by the corresponding standard value to obtain a use abnormality prediction result, wherein the corresponding standard value is the output standard feature of the test equipment, and multiplying the use abnormality prediction result by the wear influence coefficient to obtain a wear abnormality prediction result;

[0073] Step 420, the wear-resistant life of the workpiece coating under the standard friction frequency, the standard friction speed and the standard friction force is obtained by dividing the wear-resistant abnormal prediction result by the wear-resistant life prediction result of the workpiece coating obtained by passing through the multilayer perception machine model, and the weighting weight and the wear-resistant influence coefficient are obtained by fitting historical data;

[0074] S500, the coating performance warning is performed through the wear-resistant life prediction result of the workpiece coating;

[0075] In the embodiment, the coating performance warning in the S500 includes the following specific contents:

[0076] The obtained wear-resistant life prediction result of the workpiece coating is compared with the corresponding wear-resistant life required value of the workpiece coating, if the wear-resistant life prediction result of the workpiece coating is greater than or equal to the corresponding wear-resistant life required value of the workpiece coating, it indicates that the coating performance meets the workpiece use requirement, if the wear-resistant life prediction result of the workpiece coating is less than the corresponding wear-resistant life required value of the workpiece coating, it indicates that the coating performance does not meet the workpiece use requirement, and the coating warning is needed to remind the maintenance personnel to replace.

[0077] The above embodiment has the following advantages: by integrating the workpiece surface features, the coating microstructure, the process parameters and the future working conditions and the like multi-source data, and constructing an intelligent prediction model based on a multilayer perception machine, the accurate prediction and warning of the wear-resistant life of the nickel-based composite coating are realized, the method can learn the complex nonlinear relationship from the historical data, the weak link of the coating performance is identified in advance, not only the predictability and accuracy of quality control are improved, but also the equipment maintenance cost and downtime risk caused by coating failure are significantly reduced, and a reliable decision basis is provided for optimizing the spraying process and prolonging the service life of the workpiece.

[0078] Please refer to Figure 4 , Figure 4 is a structure schematic diagram of a nickel-based composite coating wear-resistant performance prediction system based on machine learning provided by the embodiment of the application, comprising:

[0079] A data acquisition module acquires the workpiece surface condition and the nickel-based composite coating condition, and simultaneously acquires the environment condition in the coating process and the future use condition of the corresponding workpiece;

[0080] A coating connection analysis module performs workpiece coating connection analysis through the workpiece surface flatness in the workpiece surface condition and the hard phase distribution condition in the nickel-based composite coating condition;

[0081] A machine learning analysis module constructs a machine learning model, and performs workpiece coating wear-resistant effect prediction through the workpiece coating connection analysis result, the average microhardness condition in the nickel-based composite coating condition and the environment condition in the coating process;

[0082] a life prediction module, which predicts the wear-resistant life of the workpiece coating based on the future use of the workpiece and the wear-resistant effect prediction result of the workpiece coating;

[0083] a performance early warning module, which performs coating performance early warning based on the wear-resistant life prediction result of the workpiece coating.

[0084] The steps of implementing the respective functions of the parameters and the unit modules in the nickel-based composite coating wear-resistant performance prediction system based on machine learning of the present application can refer to the parameters and steps in the embodiments of the nickel-based composite coating wear-resistant performance prediction method based on machine learning, which will not be repeated here.

[0085] The embodiments of the present application also provide an electronic device, which includes a memory, a processor and a communication bus; the memory and the processor are connected through the communication bus. The memory stores the nickel-based composite coating wear-resistant performance prediction method based on machine learning provided by the above embodiments, which can be loaded and executed by the processor 320.

[0086] The memory can be used to store instructions, programs, codes, code sets or instruction sets. The memory 310 can include a storage program area and a storage data area, wherein the storage program area can store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the nickel-based composite coating wear-resistant performance prediction method based on machine learning provided by the above embodiments, etc.; the storage data area can store data involved in the nickel-based composite coating wear-resistant performance prediction method based on machine learning provided by the above embodiments, etc.

[0087] The processor can include one or more processing cores. The processor calls the data stored in the memory by running or executing the instructions, program code sets or instruction sets stored in the memory, and performs various functions and processes data of the present application. The processor can be at least one of an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a Central Processing Unit (CPU), a controller, a microcontroller and a microprocessor. It can be understood that, for different devices, the electronic devices used to implement the functions of the processor 320 described above can also be other devices, and the embodiments of the present application do not make specific limitations.

[0088] The communication bus can include a path to transmit information between the above components. The communication bus 330 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0089] The embodiment of the present application provides a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor to perform the method for predicting wear resistance of a nickel-based composite coating based on machine learning provided by the above embodiment.

[0090] In the embodiment of the present application, the computer readable storage medium can be a tangible device that maintains and stores instructions for use by an instruction execution device. The computer readable storage medium can be, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination of the above. Specifically, the computer readable storage medium can be a portable computer disk, a hard disk, a U disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a standing random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical coding device, and any combination of the above.

[0091] The term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or other elements inherent to such a process, method, article or device.

[0092] The above description is only the preferred embodiment of the present application and the explanation of the applied technical principles. Those skilled in the art should understand that the application range involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above application concept. For example, the technical solutions formed by the mutual replacement of the above features and the technical features applied in the present application (but not limited to) having similar functions.

Claims

1. A method for predicting the wear resistance of a nickel-based composite coating based on machine learning, characterized by, Comprise the following steps: S100, acquire the workpiece surface condition and the nickel-based composite coating condition, and simultaneously acquire the environmental condition in the coating process and the future use condition of the corresponding workpiece; S200, perform workpiece coating connection analysis by the workpiece surface flatness in the workpiece surface condition and the hard phase distribution in the nickel-based composite coating condition; The workpiece coating connection analysis in the S200 comprises the following specific steps: Step 210, acquire the three-dimensional image of the workpiece surface, and further acquire the pit distribution condition of the workpiece surface and the roughness condition of the corresponding workpiece surface, and simultaneously acquire the hard phase distribution in the coating, acquire the roughness abnormality by the standard deviation of the roughness and the safe roughness range; Step 220, acquire the pit size condition of the workpiece surface and the size deviation of the hard phase particles to obtain the size matching abnormality, and acquire the hard phase distribution uniformity abnormality by the hard phase distribution in the coating, wherein the hard phase distribution uniformity abnormality is acquired in the following manner: acquire the hard phase distribution density of each position in the coating, acquire the standard deviation of the hard phase distribution density of each position, and obtain the hard phase distribution uniformity abnormality; Step 230, acquire the pit distribution condition of the workpiece surface, acquire the standard deviation of the pit size as the pit size abnormality, and simultaneously acquire the spacing condition between the pits, acquire the standard deviation of the spacing condition to obtain the pit distribution abnormality; Step 240, construct a data sequence in order according to the acquired roughness abnormality, size matching abnormality, hard phase distribution uniformity abnormality, pit size abnormality and pit distribution abnormality, and set the constructed sequence as the workpiece coating connection analysis sequence; S300, construct a machine learning model, and perform workpiece coating wear-resistant effect prediction by the workpiece coating connection analysis result, the average microhardness condition in the nickel-based composite coating condition and the environmental condition in the coating process; S400, acquire the future use condition of the corresponding workpiece and the workpiece coating wear-resistant effect prediction result to perform workpiece coating wear-resistant life prediction; S500, perform coating performance early warning by the workpiece coating wear-resistant life prediction result.

2. The machine learning based nickel-based composite coating wear resistance performance prediction method according to claim 1, characterized in that, The S100 comprises the following specific contents: Step 110, acquire the three-dimensional image of the workpiece surface and the nickel-based composite coating condition by the corresponding sensor, wherein the nickel-based composite coating condition comprises the hard phase distribution and the average microhardness; Step 120, acquire the temperature, humidity and particle flow rate condition in the thermal spraying process by the sensor; Step 130, acquire the use friction force load, use frequency condition and friction speed condition of the corresponding workpiece by the use log of the workpiece.

3. The machine learning based nickel-based composite coating wear performance prediction method of claim 1, wherein, The workpiece coating wear-resistant effect prediction in the S300 specifically comprises the following specific contents: Step 310, acquire the temperature, humidity and particle flow rate in the historical thermal spraying process; at the same time, acquire the average microhardness in the historical workpiece coating connection analysis sequence and the nickel-based composite coating condition, and the wear life of the historical workpiece coating under the action of the standard friction frequency, the standard friction speed and the standard friction force, and construct a multilayer perception machine model with the input of the temperature, humidity, particle flow rate in the historical thermal spraying process, the average microhardness in the nickel-based composite coating condition and the workpiece coating connection analysis sequence, and the output of the wear life of the workpiece coating under the action of the standard friction frequency, the standard friction speed and the standard friction force; Step 320, import the acquired real-time temperature, humidity, particle flow rate in the thermal spraying process, the average microhardness in the nickel-based composite coating condition and the workpiece coating connection analysis sequence into the constructed multilayer perception machine model, and output the predicted wear life of the workpiece coating under the action of the standard friction frequency, the standard friction speed and the standard friction force.

4. The machine learning based nickel-based composite coating wear performance prediction method of claim 1, wherein, The workpiece coating wear life prediction in the S400 includes the following specific contents: Step 410, acquire the use friction force load, use frequency condition and friction speed condition of the corresponding workpiece in the future, and perform weighted summation after dividing the obtained use friction force load, use frequency condition and friction speed condition by the corresponding standard value to obtain a use abnormality prediction result, wherein the corresponding standard value is the output standard feature of the test equipment, and the use abnormality prediction result is multiplied by a wear influence coefficient to obtain a wear abnormality prediction result; Step 420, divide the wear life of the workpiece coating under the action of the standard friction frequency, the standard friction speed and the standard friction force predicted by the multilayer perception machine model by the wear abnormality prediction result to obtain the workpiece coating wear life prediction result.

5. The machine learning based nickel-based composite coating wear performance prediction method of claim 2, wherein, The coating performance early warning in the S500 includes the following specific contents: The obtained workpiece coating wear life prediction result is compared with the corresponding workpiece coating wear life required value, if the workpiece coating wear life prediction result is greater than or equal to the corresponding workpiece coating wear life required value, it means that the coating performance meets the workpiece use demand, if the workpiece coating wear life prediction result is less than the corresponding workpiece coating wear life required value; it means that the coating performance does not meet the workpiece use demand, and coating early warning is needed to remind the maintenance personnel to replace.

6. The machine learning based nickel-based composite coating wear resistance performance prediction method according to claim 2, wherein, The roughness evaluation method of the workpiece surface is to obtain the average absolute value of the height of each point on the workpiece surface relative to the reference surface, and to obtain the roughness by the ratio of the obtained average absolute value to the safety height.

7. A machine learning based nickel-based composite coating wear resistance performance prediction system for implementing the machine learning based nickel-based composite coating wear resistance performance prediction method of any one of claims 1-6, characterized in that, Specifically includes: A data acquisition module acquires the workpiece surface condition and the nickel-based composite coating condition, and simultaneously acquires the environment condition in the coating process and the future use condition of the corresponding workpiece; A coating connection analysis module performs workpiece coating connection analysis through the workpiece surface flatness in the workpiece surface condition and the hard phase distribution in the nickel-based composite coating condition; The machine learning analysis module constructs a machine learning model to predict the wear resistance of the workpiece coating by analyzing the connection between the workpiece coating, the average microhardness in the nickel-based composite coating, and the environmental conditions during the coating process. The life prediction module predicts the wear resistance life of the workpiece coating by obtaining the future use of the corresponding workpiece and the wear resistance prediction result of the workpiece coating. The performance warning module warns the coating performance by the wear resistance life prediction result of the workpiece coating.

8. An electronic device comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor executes the machine learning-based nickel-based composite coating wear resistance performance prediction method according to any one of claims 1-6 by calling the computer program stored in the memory.

Images