Self-adaptive optimization gray model-based barrel ablation wear prediction method and device

Abstract

The embodiment of the invention provides a barrel ablation wear prediction method and device based on an adaptive optimization gray model, and the method comprises the steps: carrying out the first-order accumulation of an original sequence obtained through a shooting test, and building a whitening time response function; a new information priority criterion is innovatively adopted to calculate a data weight coefficient, self-adaptive optimization of initial conditions is realized, and a non-equidistant grey prediction model is constructed. The system establishes an unconstrained optimization problem based on a minimum criterion of average error sum of squares, and adopts a steepest descent method to solve an optimal time parameter, so that the model prediction precision is improved. And a shooting wear curve is established in combination with shooting number data, so that accurate prediction and evaluation of the ablation wear state of the artillery barrel are realized, and a reliable decision basis is provided for use and maintenance of the artillery.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method and device for predicting barrel ablation wear based on an adaptive optimization grey model. Background Art

[0002] The existing barrel ablation wear prediction technology has obvious shortcomings. Traditional prediction methods often use simple linear models or statistical methods, which fail to fully consider the nonlinear and time-varying characteristics of the barrel ablation wear process, resulting in low prediction accuracy. At the same time, the existing methods have limitations when processing incomplete or unequally spaced test data, making it difficult to adapt to actual application scenarios.

[0003] In addition, existing technologies also have bottlenecks in model parameter optimization and accuracy improvement. Most methods use fixed initial conditions and time parameters, lack adaptive optimization mechanisms, and cannot dynamically adjust model parameters according to actual data characteristics, which affects prediction performance. Model evaluation and verification methods are relatively simple and lack systematic error analysis and optimization methods.

[0004] The existing system has technical shortcomings in the assessment of firing wear status. The lack of reliable status judgment standards and early warning mechanisms makes it difficult to accurately assess the life of artillery and maintenance needs. Solving these problems is of great significance to improving the effectiveness of artillery use and maintenance assurance. Summary of the invention

[0005] In response to the problems in the prior art, the present application provides a barrel ablation and wear prediction method and device based on an adaptive optimization grey model, which can accurately predict and evaluate the ablation and wear state of the gun barrel, and provide a reliable decision-making basis for the use and maintenance of the gun.

[0006] In order to solve at least one of the above problems, the present application provides the following technical solutions: In a first aspect, the present application provides a method for predicting barrel ablation wear based on an adaptive optimization grey model, comprising: The barrel ablation and wear data obtained from the shooting test is used as the original sequence, the original sequence is first-order accumulated to obtain an accumulated sequence, a whitening time response function is constructed based on the accumulated sequence, a weight coefficient of each data point in the original sequence is calculated by a new information priority criterion, the initial conditions are adaptively optimized according to the weight coefficients, the optimized initial conditions are substituted into the whitening time response function, and a non-equidistant grey prediction model is constructed; An unconstrained optimization problem is established based on the minimum criterion of the average square error sum, and the error between the prediction result of the non-equidistant grey prediction model and the actual test data is used as the optimization target. The unconstrained optimization problem is solved by the steepest descent method to obtain a time parameter, and the time parameter is substituted into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model; The firing number data of the artillery to be predicted is read, the firing number data is input into the improved adaptive grey prediction model, the barrel ablation wear prediction values ​​corresponding to different firing numbers are obtained, a firing wear curve is established according to the prediction value and the firing number data, and the ablation wear state of the artillery barrel is judged based on the firing wear curve.

[0007] Furthermore, the barrel ablation and wear data obtained from the shooting test is used as an original sequence, the original sequence is first-order accumulated to obtain an accumulated sequence, a whitening time response function is constructed based on the accumulated sequence, and a weight coefficient of each data point in the original sequence is calculated by a new information priority criterion, including: Obtain barrel diameter change data during the artillery firing process, sort the barrel diameter change data according to the number of shots fired to construct a time series data sequence, calculate the barrel diameter change amount between two adjacent shots based on the time series data sequence to obtain an original ablation wear sequence, and mark the original ablation wear sequence as X(0); Each data point in the original ablation and wear sequence X(0) is added to all the previous data points to obtain the cumulative sequence X(1), the means of the adjacent data of the cumulative sequence X(1) are calculated to generate the adjacent mean sequence Z(1), the adjacent mean sequence Z(1) is substituted into the whitening differential equation to construct the whitening time response function, and the influence weight of each data point in the original ablation and wear sequence X(0) on the prediction result is calculated based on the data recency principle.

[0008] Furthermore, the initial conditions are adaptively optimized according to the weight coefficients, the optimized initial conditions are substituted into the whitening time response function, and a non-equidistant grey prediction model is constructed, including: Based on the weight coefficient, an initial condition expression for adaptive optimization is constructed, the data in the cumulative sequence X(1) are multiplied by the corresponding weight coefficients in sequence and the sum is calculated to obtain the optimized initial condition value, a non-equidistant model equation group is established using the adjacent mean sequence Z(1), and the least square method is used to solve the equation group to obtain the development coefficient and the gray action; The optimized initial condition value, the development coefficient and the gray action are substituted into the whitening time response function to construct a prediction model expression considering non-equidistant characteristics. The prediction model expression is restored to obtain a non-equidistant gray prediction model, which outputs the prediction value of barrel ablation and wear under different numbers of firing rounds.

[0009] Furthermore, the unconstrained optimization problem is established based on the minimum criterion of the mean square error sum, the error between the prediction result of the non-equidistant grey prediction model and the actual test data is used as the optimization target, and the steepest descent method is used to solve the unconstrained optimization problem to obtain the time parameter, including: Comparing the predicted value of the non-equidistant grey prediction model with the actual test data, calculating the relative error under each shooting number, performing square sum operation on all relative errors to obtain an overall error function, establishing an unconstrained optimization model with the overall error function as the objective function, and substituting the time parameter as the optimization variable into the unconstrained optimization model; The objective function of the unconstrained optimization model is differentiated to obtain a gradient vector, and the value of the time parameter is iteratively updated based on the steepest descent method. In each iteration, the algorithm moves along the negative gradient direction and determines the optimal step size through a one-dimensional search. When the norm of the gradient vector is less than the preset accuracy requirement or the maximum number of iterations is reached, the optimal time parameter value is output.

[0010] Furthermore, the step of substituting the time parameter into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model includes: Construct a prediction model expression with time parameters, substitute the optimal time parameters, development coefficient and grey action into the prediction model expression, modify the prediction model expression according to the relationship between the prediction values ​​of the previous and next time points, and generate an adaptive grey prediction model considering time correlation; The parameters in the adaptive grey prediction model are normalized, the optimized initial condition values ​​are substituted into the adaptive grey prediction model to calculate the predicted values, and the parameters of the adaptive grey prediction model are fine-tuned based on the predicted values ​​to obtain an improved grey prediction model with adaptive optimization characteristics.

[0011] Further, the reading of the firing number data of the artillery to be predicted, inputting the firing number data into the improved adaptive grey prediction model, and obtaining the barrel ablation wear prediction values ​​corresponding to different firing numbers, includes: Acquire historical firing data of the artillery to be predicted from the artillery data acquisition system, divide the historical firing data into time series to obtain firing number data of different time periods, arrange the firing number data in time sequence to form a sequence to be predicted, and perform data standardization processing on the sequence to be predicted; The standardized sequence to be predicted is input into the improved adaptive grey prediction model, the barrel ablation wear amount corresponding to each time point is calculated based on the prediction function of the model, and the predicted ablation wear amount is de-standardized to obtain the actual barrel ablation wear prediction value.

[0012] Furthermore, establishing a firing wear curve according to the predicted value and the firing number data, and judging the ablation wear state of the gun barrel based on the firing wear curve, comprises: A coordinate system is constructed by using the firing number data as the horizontal coordinate and the barrel ablation wear prediction value as the vertical coordinate, and a cubic spline interpolation algorithm is used to continuously smooth discrete prediction data points to generate a continuous curve reflecting the trend of firing wear changes, and the slope and curvature of the continuous curve in different intervals are calculated; The shooting wear curve is subjected to piecewise fitting processing, the tangent slope and curvature change rate of the fitting curve at each shooting number are calculated, the tangent slope is compared with a preset wear rate threshold, the acceleration characteristics of the barrel ablation wear are analyzed based on the curvature change rate, and the ablation wear state of the gun barrel is determined according to the wear change characteristics.

[0013] In a second aspect, the present application provides a barrel ablation wear prediction device based on an adaptive optimization grey model, comprising: A grey model building module is used to use barrel ablation and wear data obtained from a shooting test as an original sequence, perform first-order accumulation on the original sequence to obtain an accumulated sequence, build a whitening time response function based on the accumulated sequence, calculate a weight coefficient of each data point in the original sequence by a new information priority criterion, adaptively optimize initial conditions according to the weight coefficients, substitute the optimized initial conditions into the whitening time response function, and build a non-equidistant grey prediction model; A grey model tuning module is used to establish an unconstrained optimization problem based on the minimum criterion of the average square error sum, take the error between the prediction result of the non-equidistant grey prediction model and the actual test data as the optimization target, use the steepest descent method to solve the unconstrained optimization problem to obtain a time parameter, substitute the time parameter into the non-equidistant grey prediction model, and obtain an improved adaptive grey prediction model; The barrel state prediction module is used to read the firing number data of the artillery to be predicted, input the firing number data into the improved adaptive grey prediction model, obtain the barrel ablation wear prediction values ​​corresponding to different firing numbers, establish a firing wear curve according to the prediction value and the firing number data, and judge the ablation wear state of the gun barrel based on the firing wear curve.

[0014] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for predicting barrel ablation and wear based on the adaptive optimization grey model are implemented.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting barrel ablation and wear based on an adaptive optimization grey model.

[0016] In a fifth aspect, the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for predicting barrel ablation and wear based on an adaptive optimization grey model.

[0017] It can be seen from the above technical solution that the present application provides a method and device for predicting barrel ablation and wear based on an adaptive optimization gray model. By performing first-order accumulation on the original sequence obtained from the shooting test, a whitening time response function is established. The new information priority criterion is innovatively used to calculate the data weight coefficient, realize the adaptive optimization of the initial conditions, and construct a non-equidistant gray prediction model. The system establishes an unconstrained optimization problem based on the minimum criterion of the mean sum of squared errors, and uses the steepest descent method to solve the optimal time parameters to improve the prediction accuracy of the model. The shooting wear curve is established in combination with the shooting number data, which realizes the accurate prediction and evaluation of the ablation and wear state of the gun barrel, and provides a reliable decision-making basis for the use and maintenance of the gun. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0019] Figure 1 This is one of the flow charts of the barrel ablation wear prediction method based on the adaptive optimization grey model in the embodiment of the present application; Figure 2 This is a second flow chart of a method for predicting barrel ablation and wear based on an adaptive optimization grey model in an embodiment of the present application; Figure 3 The third flowchart of the barrel ablation wear prediction method based on the adaptive optimization grey model in the embodiment of the present application; Figure 4This is a fourth flow chart of a method for predicting barrel ablation and wear based on an adaptive optimization grey model in an embodiment of the present application; Figure 5 FIG5 is a flow chart of a method for predicting barrel ablation and wear based on an adaptive optimization grey model in an embodiment of the present application; Figure 6 This is a sixth flow chart of a method for predicting barrel ablation and wear based on an adaptive optimization grey model in an embodiment of the present application; Figure 7 FIG7 is a flow chart of a method for predicting barrel ablation and wear based on an adaptive optimization grey model in an embodiment of the present application; Figure 8 : is a structural diagram of a barrel ablation wear prediction device based on an adaptive optimization grey model in an embodiment of the present application; Fig. 9 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0020] Reference numerals: Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0023] Taking into account the problems existing in the prior art, the present application provides a method and device for predicting barrel ablation and wear based on an adaptive optimization gray model. By performing first-order accumulation on the original sequence obtained from the shooting test, a whitening time response function is established. The new information priority criterion is innovatively adopted to calculate the data weight coefficient, realize the adaptive optimization of the initial conditions, and construct a non-equidistant gray prediction model. The system establishes an unconstrained optimization problem based on the minimum criterion of the mean sum of squared errors, and uses the steepest descent method to solve the optimal time parameters to improve the prediction accuracy of the model. The shooting wear curve is established in combination with the shooting number data, which realizes the accurate prediction and evaluation of the ablation and wear state of the gun barrel, and provides a reliable decision-making basis for the use and maintenance of the gun.

[0024] In order to accurately predict and evaluate the ablation and wear state of the gun barrel and provide a reliable decision-making basis for the use and maintenance of the gun, the present application provides an embodiment of a method for predicting ablation and wear of the gun barrel based on an adaptive optimization grey model. Figure 1 The barrel ablation wear prediction method based on the adaptive optimization grey model specifically includes the following contents: Step S101: taking barrel ablation and wear data obtained from a shooting test as an original sequence, performing first-order accumulation on the original sequence to obtain an accumulated sequence, constructing a whitening time response function based on the accumulated sequence, calculating a weight coefficient of each data point in the original sequence by a new information priority criterion, adaptively optimizing initial conditions according to the weight coefficients, substituting the optimized initial conditions into the whitening time response function, and constructing a non-equidistant grey prediction model; Optionally, this embodiment adopts a high-precision measurement strategy in the data acquisition stage. First, multiple measurement points are set at key parts of the gun barrel, including the strong bore section, transition section and free section, and 12 circumferentially evenly distributed measurement positions are selected for each part. The measuring equipment uses a digital caliper with an accuracy of 0.001mm, and measurements are performed before and after each shot. In order to eliminate random errors, each measuring point is measured three times, and the arithmetic mean is taken after eliminating outliers. This multi-point measurement scheme effectively improves the reliability of the original data.

[0025] This embodiment adopts a multi-level filtering mechanism in the data preprocessing stage. The original measurement data is firstly subjected to median filtering to remove mutation points, and then the wavelet decomposition method is used to suppress high-frequency noise. For missing data, local polynomial interpolation is used to supplement, and the interpolation process considers the spatial correlation of adjacent measurement points. The data smoothing adopts the Savitzky-Golay filter with adaptive window, and the window size is dynamically adjusted according to the data change rate to ensure that the random disturbance is effectively reduced while retaining the wear trend.

[0026] This embodiment implements an innovative first-order accumulation generation mechanism. The accumulation process takes into account the non-uniformity of the shooting interval and adopts a time-weighted accumulation method. For data points with large shooting intervals, intermediate transition data are generated by piecewise cubic Hermite interpolation to ensure the smoothness of the accumulated sequence. The accumulation operation uses a dynamic programming algorithm to optimize the calculation efficiency and avoid repeated calculations through recursive relationships. In order to prevent error propagation during the accumulation process, an error compensation mechanism is designed to dynamically adjust the compensation coefficient according to the measurement accuracy of the original data.

[0027] This embodiment constructs an accurate whitening time response function. The function form is derived based on the physical model of barrel ablation, taking into account the temperature field distribution, stress-strain relationship and material properties. The response function contains linear terms and nonlinear terms. The linear terms describe the basic wear trend and the nonlinear terms characterize the accelerated wear effect. The function parameters are determined by iterative least squares method, and the Levenberg-Marquardt algorithm is used to optimize the parameter values ​​in each iteration until the convergence condition is met.

[0028] This embodiment designs an adaptive weight calculation system. The new information priority criterion is implemented through a combined weighting method, and the weight calculation considers the three dimensions of data timeliness, reliability, and representativeness. The timeliness weight adopts an exponential decay function, and the decay coefficient is adaptively adjusted according to the shooting conditions. The reliability weight is evaluated based on the discrete degree of the measured data and calculated by the coefficient of variation. The representative weight is determined by the information entropy method, and the entropy value calculation considers the distribution characteristics of the data.

[0029] This embodiment implements an innovative initial condition optimization strategy. The optimization objective function includes a prediction error term, a smoothness constraint term, and a physical constraint term. The error term is in the form of a weighted mean square error, and the weight coefficient is dynamically adjusted over time. The smoothness constraint is implemented through a differential operator to suppress the oscillation of the solution. The physical constraint ensures that the optimization result meets the basic law of ablation and wear. The solution process adopts an improved gradient descent method, the step size is adaptively determined by the Armijo criterion, and the convergence criterion is comprehensively determined by combining the gradient norm and the rate of change of the function value.

[0030] This embodiment constructs a complete non-equidistant grey prediction model. The model structure is based on differential equation theory, and non-equidistant features are processed by time scale transformation. The prediction function adopts a piecewise continuous form to ensure the smoothness of the prediction when the shooting interval changes. The model parameter optimization adopts a hybrid strategy. The outer layer uses grid search to determine the initial value of the parameter, and the inner layer is fine-tuned by the quasi-Newton method. In order to improve the generalization ability of the model, a regularization term is introduced to prevent overfitting, and the regularization coefficient is determined by cross-validation.

[0031] This embodiment establishes a reliable ablation wear prediction framework through systematic data processing and accurate model construction. This solution overcomes the limitations of the traditional equidistant model and improves the prediction accuracy through multi-level optimization. The overall design fully considers the characteristics of the artillery firing process and realizes the optimization of the entire process from data collection to model prediction. The implementation of this solution significantly improves the accuracy and reliability of the prediction, providing a scientific basis for artillery life assessment and maintenance decisions.

[0032] Step S102: establishing an unconstrained optimization problem based on the minimum criterion of the mean square error sum, taking the error between the prediction result of the non-equidistant grey prediction model and the actual test data as the optimization target, solving the unconstrained optimization problem by the steepest descent method to obtain a time parameter, substituting the time parameter into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model; Optionally, this embodiment adopts a multi-level structure when constructing an error evaluation system. First, for each shooting point, the relative error between the predicted value and the measured value is calculated, and the error calculation takes into account the influence of measurement accuracy. In order to improve the reliability of the evaluation, the error data is robustly processed, and the influence of outliers is reduced by the Huber loss function. The error statistics adopt a weighted average method, and the weight coefficient is dynamically adjusted with the reliability of the measurement time.

[0033] This embodiment implements an innovative optimization target construction mechanism. The objective function includes three parts: prediction error term, timing correlation constraint term and physical constraint term. The prediction error term is in the form of weighted mean square error, and the weight design takes into account the timeliness and reliability of the data. The timing correlation constraint is characterized by the autocorrelation function to ensure that the prediction result maintains the continuous characteristics of the ablation process. The physical constraint term is established based on the barrel material properties and mechanical model to limit the prediction result within a reasonable range.

[0034] This embodiment designs an accurate steepest descent solution strategy. The solution process first calculates the objective function gradient by numerical differentiation, and uses the central difference format to improve accuracy. The step size is selected using the Armijo criterion, and the optimal step size is determined by backtracking line search. In order to avoid local optimality, a multi-starting point search strategy is designed, and the starting point selection is determined based on physical model analysis. An adaptive step size adjustment mechanism is used in the iterative process to dynamically adjust the step size factor according to the change in the objective function value.

[0035] This embodiment constructs a reliable convergence criterion system. The convergence judgment comprehensively considers three indicators: gradient norm, relative change rate of function value and parameter update amount. The gradient norm adopts the weighted Euclidean norm, and the weight design reflects the importance of each direction. The function value change rate is calculated by sliding window, and the window size is adaptively adjusted according to the optimization process. The parameter update amount judgment adopts the relative change threshold to improve the stability of the convergence judgment.

[0036] This embodiment implements an innovative time parameter optimization mechanism. The parameter optimization process takes into account the dynamic characteristics of the shooting conditions and describes the evolution of the parameters over time through a time-varying weight function. The weight function design is based on the analysis of the ablation mechanism and reflects the differences in wear characteristics at different stages. Regularization terms are introduced in the optimization process to prevent excessive oscillation of parameter values.

[0037] This embodiment designs a complete model improvement strategy. The improvement process first verifies the physical validity of the optimized time parameters and makes a rationality judgment through the expert knowledge base. The parameter mapping uses piecewise continuous functions to ensure the stability of the model at different time scales. The model structure improvement takes into account the temporal characteristics of the prediction error and improves the prediction accuracy through adaptive weight adjustment.

[0038] This embodiment constructs a systematic model evaluation framework. The evaluation indicators include three dimensions: fitting accuracy, prediction stability, and physical rationality. The fitting accuracy is comprehensively evaluated through a variety of statistical quantities, including root mean square error, mean absolute percentage error, etc. The prediction stability is evaluated through perturbation analysis to examine the sensitivity of the model to input changes. The physical rationality evaluation is based on expert experience and theoretical analysis to ensure that the prediction results conform to the laws of ablation and wear.

[0039] This embodiment implements a reliable parameter update mechanism. The parameter update adopts a sliding window strategy, and the window size is dynamically adjusted according to the data characteristics. The forgetting factor mechanism is used in the update process to reduce the influence weight of historical data. In order to improve the stability of the update, a parameter smoothing method is designed to eliminate short-term fluctuations through weighted moving average.

[0040] This embodiment establishes a reliable adaptive prediction framework through precise optimization solution and systematic model improvement. This solution improves prediction accuracy through multi-level optimization while ensuring computational efficiency. The overall design fully considers the characteristics of the artillery ablation process and realizes precise optimization and dynamic adjustment of model parameters.

[0041] The prediction framework of this embodiment has high accuracy and high adaptability, providing a reliable tool for artillery life assessment. The prediction results are verified by complete error analysis and physical verification, ensuring the scientificity and reliability of the assessment. The implementation of this solution significantly improves the accuracy and stability of the prediction, provides an important basis for artillery maintenance decision-making and use management, and realizes the accurate prediction and dynamic monitoring of the barrel ablation and wear process.

[0042] Step S103: Read the firing number data of the artillery to be predicted, input the firing number data into the improved adaptive grey prediction model, obtain the barrel ablation wear prediction values ​​corresponding to different firing numbers, establish a firing wear curve according to the prediction value and the firing number data, and judge the ablation wear state of the artillery barrel based on the firing wear curve.

[0043] Optionally, this embodiment adopts a multi-source fusion strategy in the data collection stage. The gun control system records key parameters such as propellant temperature, chamber pressure, bullet speed, etc. of each shot in real time, and collects external condition data such as ambient temperature and atmospheric pressure. The data collection equipment uses a high-precision sensor network to achieve millisecond-level synchronous sampling. To ensure data integrity, a data backup and anomaly detection mechanism is designed to mark and process abnormal data in real time during the collection process.

[0044] This embodiment implements an innovative shooting data preprocessing mechanism. The preprocessing first performs a time series analysis on the shooting number data to identify continuous shooting sequences and intermittent shooting sequences. The data is divided into multiple time windows according to the statistical characteristics of the shooting intervals through an adaptive segmentation algorithm. For high-intensity continuous shooting segments, fine-grained data sampling is used; for intermittent shooting segments, feature extraction methods are used to reduce data redundancy.

[0045] This embodiment designs an accurate prediction value calculation strategy. The prediction process adopts a progressive calculation framework. First, a short-term prediction model is established based on historical data, and then the prediction range is gradually extended through a sliding window. The model parameter update adopts an online learning method to adjust the model coefficients in real time according to the newly added data. In order to improve the prediction stability, confidence interval estimation is introduced, and the uncertainty of the prediction results is evaluated through Monte Carlo simulation.

[0046] This embodiment constructs a reliable wear curve generation mechanism. The curve fitting adopts a piecewise smoothing strategy to maintain continuity and differentiability at key turning points. The fitting function uses an improved B-spline basis function, and local adjustment of the curve is achieved through control point optimization. Node selection is based on data distribution characteristics, and node density is increased in areas with drastic wear changes to improve fitting accuracy.

[0047] This embodiment implements an innovative state assessment framework. The assessment process first extracts the characteristic parameters of the wear curve, including the local slope, curvature change rate, and acceleration factor. The characteristic parameter calculation adopts the numerical differentiation method, and the noise effect is reduced by window filtering. The state judgment is based on multi-dimensional feature space analysis, and the quantitative assessment of the wear state is achieved through the fuzzy inference system.

[0048] This embodiment designs a complete early warning mechanism. The early warning threshold is set based on the barrel life model, taking into account the influence of material properties and usage conditions. The threshold update adopts an adaptive strategy to dynamically adjust the early warning parameters according to the actual wear data. The early warning level division considers the two dimensions of wear rate and accumulation, and achieves accurate early warning through combined criteria.

[0049] This embodiment implements the system's result visualization solution. The visualization interface adopts a multi-level display method, including real-time monitoring view, trend analysis view and warning information view. Data display supports multi-dimensional switching, and the display content can be flexibly adjusted according to user needs. Graphic interaction adopts responsive design, supporting data drilling and detail zooming.

[0050] This embodiment builds a reliable decision support system. The decision suggestion generation is based on the status assessment results and combines the expert knowledge base to provide maintenance suggestions. The suggestions include maintenance schedule, spare parts preparation, and usage adjustment. The decision process considers the balance between task requirements and maintenance costs, and gives the optimal maintenance plan through the optimization algorithm.

[0051] This embodiment establishes a reliable barrel life prediction system through precise data analysis and system status evaluation. While ensuring the accuracy of the evaluation, the solution provides comprehensive decision support through multi-level analysis. The overall design fully considers the characteristics of artillery use and realizes a complete closed loop from data collection to maintenance decision-making.

[0052] The evaluation system of this embodiment is highly accurate and practical, and provides a powerful tool for artillery equipment management. The evaluation results are verified through complete analysis, ensuring the scientificity and reliability of the judgment. The implementation of this solution significantly improves the efficiency and scientificity of equipment management, provides an important basis for equipment maintenance, and realizes accurate evaluation and predictive maintenance of the artillery barrel status.

[0053] From the above description, it can be seen that the barrel ablation and wear prediction method based on the adaptive optimization gray model provided in the embodiment of the present application can establish a whitening time response function by performing first-order accumulation on the original sequence obtained from the shooting test. The new information priority criterion is innovatively used to calculate the data weight coefficient, realize the adaptive optimization of the initial conditions, and construct a non-equidistant gray prediction model. The system establishes an unconstrained optimization problem based on the minimum criterion of the mean sum of squared errors, and uses the steepest descent method to solve the optimal time parameters to improve the prediction accuracy of the model. The shooting wear curve is established in combination with the shooting number data, which realizes the accurate prediction and evaluation of the ablation and wear state of the gun barrel, and provides a reliable decision-making basis for the use and maintenance of the gun.

[0054] In one embodiment of the barrel ablation wear prediction method based on the adaptive optimization grey model of the present application, see Figure 2 , and can also include the following: Step S201: Obtain barrel diameter change data during the artillery firing process, sort the barrel diameter change data according to the number of shots fired to construct a time series data sequence, calculate the barrel diameter change amount between two adjacent shots based on the time series data sequence to obtain an original ablation wear sequence, and mark the original ablation wear sequence as X(0); Step S202: Perform a cumulative operation on each data point in the original ablation and wear sequence X(0) and all previous data points to obtain a cumulative sequence X(1), calculate the mean of adjacent data of the cumulative sequence X(1) to generate a neighboring mean sequence Z(1), substitute the neighboring mean sequence Z(1) into the whitening differential equation to construct a whitening time response function, and calculate the influence weight of each data point in the original ablation and wear sequence X(0) on the prediction result based on the data recency principle.

[0055] Optionally, this embodiment adopts a multi-dimensional integrated measurement strategy in the barrel diameter measurement stage. The measurement system consists of a high-precision laser interferometer, a digital holographic imaging device and a micro-displacement sensor to form a complementary verification measurement network. A measurement reference ring is set every 50 mm along the axial direction of the inner wall of the barrel, and 12 measurement points are evenly distributed on each reference ring. The measuring probe adopts an automatic centering mechanism, and the air floating support ensures that the probe and the barrel wall remain precisely perpendicular. The measurement platform adopts a six-degree-of-freedom active vibration reduction system, combined with a real-time compensation mechanism for the temperature field, to effectively eliminate the influence of environmental disturbances.

[0056] This embodiment achieves multiple error suppression during the data acquisition process. First, the original measurement signal is obtained through high-frequency sampling, and the sampling frequency is set to 1kHz to ensure that transient change characteristics are captured. The signal preprocessing uses a wavelet noise reduction algorithm, and the high-frequency noise is removed by the threshold decomposition method. For each measurement point, data is collected 100 times continuously, and outliers are eliminated by a robust statistical method, and the weighted median is finally taken as the valid measurement value. The temperature compensation uses a distributed thermocouple array to monitor the temperature field distribution of the pipe wall in real time, and dynamic correction is performed through the thermal expansion model.

[0057] This embodiment implements an innovative data sequence construction mechanism. The sequence construction first establishes a multidimensional index structure, including four dimensions: shooting order, timestamp, spatial coordinates, and measurement parameters. The data organization adopts a hierarchical storage strategy to support fast retrieval and association analysis according to different dimensions. For missing data in the measurement process, an adaptive interpolation algorithm is designed. The algorithm selects the optimal interpolation basis function based on local curvature features, and determines the interpolation parameters through cross-validation to ensure the physical rationality of the supplementary data.

[0058] This embodiment designs an accurate diameter change calculation framework. The calculation process first performs measurement benchmark correction, and determines the spatial correspondence of measurement data at different times through the least squares registration method. The change calculation adopts a multi-scale analysis method, and obtains the accurate deformation field distribution in the local area through surface fitting. The weight coefficient is determined by measurement uncertainty evaluation, taking into account the comprehensive influence of measurement repeatability, instrument accuracy and environmental factors.

[0059] This embodiment constructs a reliable cumulative sequence generation mechanism. The accumulation process adopts an improved dynamic programming algorithm to optimize the calculation efficiency through recursive relations. In order to prevent error accumulation, a multi-level error compensation mechanism is designed. First, random fluctuations are eliminated by the sliding mean method, then the system error is corrected based on local trend analysis, and finally the dynamic error tracking is achieved through Kalman filtering. The sequence stability is evaluated by the Lyapunov index to ensure the convergence of the accumulation process.

[0060] This embodiment implements an innovative mean sequence generation strategy. The mean calculation adopts an adaptive window mechanism, and the window size is determined by optimizing the information entropy criterion. In each window, the weighted mean is calculated by the M-estimation method, and the weight function uses the double-weight Huber function to effectively suppress the influence of outliers. In order to improve the calculation accuracy, local polynomial fitting is introduced, and the smoothness of the result is ensured by regularization constraints.

[0061] This embodiment designs a complete method for constructing a whitening response function. The function form is derived based on the constitutive relationship of the barrel material, taking into account the coupling of elastic deformation, plastic deformation and creep effect. The model parameters are determined by iterative least squares method, and the Trust-Region algorithm is used in the solution process to ensure global convergence. Physical constraints are introduced through the Lagrange multiplier method to ensure the physical rationality of the solution.

[0062] This embodiment implements the weight calculation system of the system. The weight design is based on the multi-criteria decision-making theory, which comprehensively considers the timeliness, reliability and representativeness of the data. Timeliness is described by an exponential decay function, and the decay coefficient is adaptively adjusted according to the shooting conditions. The reliability assessment adopts a fuzzy comprehensive evaluation method, and the weight of each factor is determined through hierarchical analysis. The representativeness of the data is calculated through information gain, which reflects the contribution of the data to the prediction accuracy.

[0063] This embodiment builds a complete quality control system. Data quality assessment uses a multi-dimensional indicator system, including measurement system analysis (MSA), process capability index (Cpk) and time series stability analysis. The assessment results are comprehensively judged by the fuzzy reasoning system to generate data reliability levels as an important basis for weight calculation.

[0064] This embodiment establishes a reliable ablation and wear analysis framework through precise data processing and systematic sequence construction. This solution optimizes the entire process from measurement data acquisition to sequence construction, significantly improving data quality and analysis accuracy. The overall design fully considers the characteristics of artillery shooting and provides a high-quality data foundation for subsequent prediction models.

[0065] The analysis framework of this embodiment achieves accurate characterization of the barrel ablation and wear process through multi-level data processing and strict quality control, providing reliable data support for equipment maintenance decisions. The implementation of this solution significantly improves the accuracy and reliability of data processing, laying a solid data foundation for artillery life assessment.

[0066] In one embodiment of the barrel ablation wear prediction method based on the adaptive optimization grey model of the present application, see Figure 3 , and can also include the following: Step S301: constructing an adaptively optimized initial condition expression based on the weight coefficient, multiplying the data in the cumulative sequence X(1) by the corresponding weight coefficient in turn and summing them to obtain the optimized initial condition value, using the adjacent mean sequence Z(1) to establish a non-equidistant model equation group, and using the least squares method to solve the equation group to obtain the development coefficient and the gray action; Step S302: Substitute the optimized initial condition value, the development coefficient, and the gray action into the whitening time response function, construct a prediction model expression that takes into account non-equidistant characteristics, perform a reduction operation on the prediction model expression to obtain a non-equidistant gray prediction model, and the non-equidistant gray prediction model outputs the predicted value of barrel ablation and wear under different numbers of shots.

[0067] Optionally, this embodiment adopts a multi-level optimization strategy in the initial condition construction stage. The initial expression design is based on the generalized exponential function family, and adaptive adjustment is achieved through weight coefficients. The weight allocation adopts a multi-criteria evaluation method, including three dimensions: time decay factor, data reliability index, and prediction contribution. The time decay factor is described by an improved Logistic function, and the decay parameter is determined by cross-validation. The reliability index is calculated based on measurement uncertainty analysis, and the confidence level of the data is evaluated by Monte Carlo simulation. The prediction contribution is determined by sensitivity analysis, which reflects the degree of influence of the data point on the prediction accuracy.

[0068] This embodiment implements an innovative weight optimization mechanism. The optimization process first establishes an objective function, which includes a prediction error term, a smoothness constraint term, and a physical constraint term. The error term is in the form of a weighted bi-norm, and the weight design reflects the importance of data in different time periods. The smoothness constraint is implemented by a second-order difference operator to suppress drastic changes in the weight coefficient. The physical constraint is established based on the analysis of the ablation mechanism to ensure that the weight distribution conforms to the actual wear law. The optimization solution adopts the augmented Lagrangian method, and fast convergence is achieved through the alternating direction multiplier method.

[0069] This embodiment designs an accurate sequence weighted accumulation strategy. The accumulation process adopts a segmented processing method to divide the data sequence into multiple sub-intervals according to the characteristics of the shooting conditions. In each interval, the trend component is estimated by local weighted regression, and the residual analysis is described by the ARIMA model. The weight update adopts an adaptive mechanism to dynamically adjust the weight coefficient through residual analysis. In order to improve the calculation stability, a regularization term is introduced to constrain the weight change range.

[0070] This embodiment constructs a reliable non-equidistant model equation set. The equation set is constructed based on the generalized difference equation theory, and the non-equidistant characteristics are processed by time scale transformation. The difference format uses improved central difference to improve the accuracy of discrete approximation. The coefficient matrix is ​​analyzed by condition number to ensure the stability of the solution. In order to deal with the ill-conditioned problem of the equation set, the Tikhonov regularization method is used, and the regularization parameter is determined by the L-curve method.

[0071] This embodiment implements an innovative parameter solution strategy. The solution process adopts the weighted iterative least squares method, and the weight matrix reflects the reliability of the equation. The trust region strategy is adopted in the iterative process, and convergence is ensured by adaptive step size control. In order to avoid local optimality, a multi-starting point search mechanism is designed, and the starting point selection is based on physical model analysis. The uncertainty of parameter estimation is evaluated by the Fisher information matrix.

[0072] This embodiment designs a complete prediction model construction method. The model structure is based on differential equation theory, and the nonlinear characteristics are described by time-varying coefficients. The physical meanings of the development coefficient and the ash action amount correspond to the wear rate and the cumulative effect, respectively. The model is verified by multiple cross-validation, and the prediction performance is evaluated by the holdout method. The parameter sensitivity analysis uses the Sobol index method to identify key influencing factors.

[0073] This embodiment implements the restoration calculation framework of the system. The restoration process adopts a piecewise continuous strategy and achieves smooth transition through spline interpolation. In order to reduce the cumulative error, a multi-level error compensation mechanism is designed. The compensation term includes two parts: system error correction and random error filtering. The reliability of the prediction results is evaluated through confidence interval analysis.

[0074] This embodiment builds a reliable prediction value generation mechanism. The prediction process adopts a progressive calculation framework and realizes dynamic update through sliding windows. The window size is adaptively adjusted based on data characteristics to ensure the sensitivity of the prediction model to new information. The reliability of the prediction results is confirmed by residual analysis and extrapolation performance evaluation.

[0075] This embodiment establishes a reliable ablation wear prediction framework through precise model construction and systematic parameter optimization. While ensuring computational efficiency, this solution improves prediction accuracy through multi-level optimization. The overall design fully considers the characteristics of artillery shooting and realizes full process optimization from data processing to prediction output.

[0076] The prediction framework of this embodiment has high accuracy and high adaptability, providing a reliable tool for artillery life assessment. The prediction results are verified by complete error analysis and physical verification, ensuring the scientificity and reliability of the assessment. The implementation of this solution significantly improves the accuracy and stability of the prediction, providing an important basis for artillery maintenance decision-making and use management.

[0077] In one embodiment of the barrel ablation wear prediction method based on the adaptive optimization grey model of the present application, see Figure 4 , and can also include the following: Step S401: comparing the predicted value of the non-equidistant grey prediction model with the actual test data, calculating the relative error under each shooting number, performing square sum operation on all relative errors to obtain an overall error function, establishing an unconstrained optimization model with the overall error function as the objective function, and substituting the time parameter as the optimization variable into the unconstrained optimization model; Step S402: Derivative the objective function of the unconstrained optimization model to obtain a gradient vector, iteratively update the value of the time parameter based on the steepest descent method, move along the negative gradient direction in each iteration and determine the optimal step size through one-dimensional search, and output the optimal time parameter value when the norm of the gradient vector is less than the preset accuracy requirement or the maximum number of iterations is reached.

[0078] Optionally, this embodiment adopts a multi-dimensional analysis strategy in the error evaluation stage. The relative error calculation comprehensively considers the measurement accuracy and data reliability, and is quantified by an improved percentage error formula. An adaptive weight mechanism is introduced in the error calculation process to assign different weights to errors in different shooting stages. The weight allocation is based on the characteristics of the shooting conditions, and different evaluation criteria are used for high-intensity continuous shooting segments and intermittent shooting segments. In order to improve the reliability of the evaluation, an outlier detection mechanism is designed to identify and process abnormal errors through statistical test methods.

[0079] This embodiment implements an innovative overall error function construction method. The error function design adopts an improved weighted square sum form, and the weight coefficient is determined by multi-objective optimization. The optimization objectives include three dimensions: prediction accuracy, model stability, and physical rationality. The prediction accuracy is evaluated by residual analysis, the model stability is determined by parameter sensitivity analysis, and the physical rationality is verified by an expert knowledge base. A smoothing term is introduced in the error function construction process, and the influence of noise is suppressed by regularization methods.

[0080] This embodiment designs an accurate unconstrained optimization model. The model is constructed based on the variational principle, and a mathematical description of the optimization problem is established through functional analysis. The objective function design considers the physical constraints of the time parameter, and the constraints are converted into an unconstrained form through the penalty function method. In order to improve the optimization efficiency, the quadratic approximation technology is used to accelerate the convergence process through the local quadratic model. The trust region method is used to solve the model, and the search range is dynamically adjusted.

[0081] This embodiment builds a reliable gradient calculation framework. The gradient vector calculation adopts an adaptive difference format, and the step size selection is based on truncation error analysis. In order to improve numerical stability, a multi-grid strategy is designed to improve accuracy by alternating coarse and fine grid calculations. The parameter scale difference is considered in the gradient calculation process, and the scale normalization is achieved through the preprocessing matrix. In order to prevent the gradient from disappearing, an improved chain derivation rule is used.

[0082] This embodiment implements an innovative iterative optimization strategy. The implementation of the steepest descent method adopts an adaptive step size mechanism, and the step size factor is dynamically adjusted through the Armijo criterion. The search direction selection combines the conjugate gradient idea and accelerates convergence through the momentum mechanism. To avoid oscillation, a momentum decay strategy is designed to adjust the momentum factor according to the change of the objective function value. A multi-starting point strategy is adopted in the iterative process, and the starting point selection is based on physical model analysis.

[0083] This embodiment designs a complete one-dimensional search method. The search process adopts the improved golden section method and improves the search efficiency through interpolation technology. The interval update adopts an adaptive strategy to dynamically adjust the search range according to the change of the function value. In order to improve the search reliability, the Wolfe condition is introduced to ensure the optimality of the step size selection. The quadratic interpolation is used to estimate the optimal point in the search process to accelerate the convergence speed.

[0084] This embodiment implements the convergence criterion system of the system. The convergence judgment comprehensively considers three indicators: gradient norm, function value change and parameter update amount. The gradient norm adopts the weighted Euclidean norm, and the weight design reflects the importance of each direction. The function value change is evaluated by the relative change rate, and the parameter update amount is measured by the maximum norm. The criterion is combined with a fuzzy logic system to achieve multi-criteria comprehensive decision-making.

[0085] This embodiment constructs a reliable parameter output mechanism. After the optimal parameters are determined, their generalization performance is evaluated by cross-validation. The validation process uses the k-fold method, and the evaluation indicators include mean square error, maximum relative error, and prediction stability. Parameter reliability is evaluated by perturbation analysis to examine the sensitivity of parameters to input changes.

[0086] This embodiment establishes a reliable model improvement framework through precise error analysis and systematic parameter optimization. While ensuring optimization efficiency, this solution improves the accuracy of parameter estimation through multi-level strategies. The overall design fully considers the characteristics of artillery ablation and realizes full process optimization from error evaluation to parameter optimization.

[0087] The optimization framework of this embodiment is highly efficient and reliable, providing a powerful tool for model improvement. The optimization results are fully verified and analyzed to ensure the scientificity and reliability of parameter estimation. The implementation of this solution significantly improves the accuracy and stability of model prediction, providing reliable technical support for artillery life assessment.

[0088] In one embodiment of the barrel ablation wear prediction method based on the adaptive optimization grey model of the present application, see Figure 5 , and can also include the following: Step S501: construct a prediction model expression with time parameters, substitute the optimal time parameters, development coefficient and grey action into the prediction model expression, modify the prediction model expression according to the relationship between the prediction values ​​of the previous and next time points, and generate an adaptive grey prediction model considering time correlation; Step S502: normalize the parameters in the adaptive grey prediction model, substitute the optimized initial condition values ​​into the adaptive grey prediction model to calculate the predicted values, and fine-tune the parameters of the adaptive grey prediction model based on the predicted values ​​to obtain an improved grey prediction model with adaptive optimization features.

[0089] Optionally, this embodiment adopts a multi-dimensional fusion strategy in the prediction model construction stage. The model expression design is based on the generalized time response theory, and the dynamic characteristics are described by introducing the time parameter. The coupling relationship between the time parameter and the development coefficient and the gray action is expressed by a nonlinear mapping function, and the form of the mapping function is determined by the physical mechanism analysis. In order to improve the adaptability of the model, a parameter adaptive mechanism is designed to achieve dynamic adjustment of parameters through feedback control.

[0090] This embodiment implements an innovative temporal correlation analysis method. The correlation analysis adopts a multi-scale method and realizes multi-resolution decomposition of time series through wavelet transform. Local prediction models are established at different scales, and the final prediction results are obtained through weighted combination. The weight coefficient is determined by cross-validation to reflect the importance of features at different scales. In order to capture long-term dependencies, memory items are introduced, and the effective use of historical information is realized through the LSTM structure.

[0091] This embodiment designs an accurate model correction strategy. The correction process first establishes an error feedback mechanism and identifies model deviations through residual analysis. Feedback control uses a PID structure, and control parameters are optimized through a genetic algorithm. To improve the correction effect, an adaptive filter is designed, and state estimation is achieved through Kalman filtering. The weight of the correction term is dynamically adjusted through the information entropy criterion to ensure the rationality of the correction.

[0092] This embodiment constructs a reliable parameter normalization framework. The normalization process uses an improved min-max method to achieve parameter scale unification through piecewise linear transformation. The selection of the transformation interval is based on the physical meaning of the parameter to ensure that the normalized parameter maintains its original physical interpretability. In order to deal with the influence of outliers, a robust normalization strategy is designed to reduce the influence of extreme values ​​through median normalization.

[0093] This embodiment implements an innovative prediction value calculation mechanism. The calculation process adopts a hierarchical structure, first establishing a trend prediction model at the macro scale, and then superimposing local fluctuation terms at the micro scale. Trend prediction uses an improved exponential smoothing method, and the smoothing parameters are optimized through grid search. The fluctuation term prediction uses a conditional autoregressive model, and the model parameters are determined by maximum likelihood estimation.

[0094] This embodiment designs a complete parameter fine-tuning scheme. The fine-tuning process is based on sensitivity analysis, and the key parameters are identified through the local perturbation method. The perturbation amplitude is controlled by an adaptive mechanism to ensure the stability of the adjustment. The parameter update adopts the gradient projection method, and the parameter constraints are satisfied through the projection matrix. To avoid over-adjustment, regularization constraints are introduced, and sparsity control is achieved through the L1 / L2 mixed norm.

[0095] This embodiment implements the model optimization mechanism of the system. The optimization process adopts a multi-objective framework, taking into account prediction accuracy, model complexity and physical interpretability. The objective function is solved by Pareto optimization, and the NSGA-II algorithm is used to achieve multi-objective optimization. In order to improve the optimization efficiency, a parallel computing strategy is designed, and parallel search is achieved through task decomposition.

[0096] This embodiment constructs a reliable adaptive feature extraction method. Feature extraction uses a deep learning framework and realizes automatic extraction of time series features through convolutional neural networks. The network structure design takes into account the time scale characteristics and captures different frequency features through multi-scale convolution. Model training adopts a transfer learning strategy to improve the generalization ability of the model through pre-training.

[0097] This embodiment establishes a reliable prediction framework through precise model construction and systematic parameter optimization. While ensuring prediction accuracy, this solution improves model adaptability through multi-level optimization. The overall design fully considers the characteristics of artillery ablation and realizes full process optimization from model construction to parameter optimization.

[0098] The prediction framework of this embodiment has high adaptability and reliability, providing an accurate tool for artillery life assessment. The optimization results are fully verified and analyzed to ensure the scientificity and reliability of the prediction. The implementation of this solution significantly improves the accuracy and adaptability of the prediction, provides reliable technical support for artillery maintenance decisions, and realizes accurate prediction of the ablation and wear process.

[0099] The prediction results of this embodiment show excellent time series correlation and physical consistency, and the model can quickly adapt to different working conditions through adaptive optimization. This solution has shown strong prediction capabilities in practical applications, providing a scientific basis for artillery life assessment and maintenance decisions, and achieving accurate assessment and prediction of equipment status.

[0100] In one embodiment of the barrel ablation wear prediction method based on the adaptive optimization grey model of the present application, see Figure 6 , and can also include the following: Step S601: acquiring historical firing data of the artillery to be predicted from the artillery data acquisition system, dividing the historical firing data into time series to obtain firing number data of different time periods, arranging the firing number data in time sequence to form a sequence to be predicted, and performing data standardization processing on the sequence to be predicted; Step S602: inputting the normalized sequence to be predicted into the improved adaptive grey prediction model, calculating the barrel ablation wear amount corresponding to each time point based on the prediction function of the model, and performing denormalization processing on the predicted ablation wear amount to obtain an actual barrel ablation wear prediction value.

[0101] Optionally, this embodiment adopts a multi-source integration strategy in the data acquisition stage. The data acquisition system includes a shooting parameter acquisition unit, an environmental monitoring unit, and a barrel state detection unit. Shooting parameter acquisition is achieved through a high-speed data acquisition card, and the sampling frequency can reach 100kHz to ensure the capture of transient change characteristics. Environmental monitoring adopts a distributed sensor network to record parameters such as temperature, humidity, and atmospheric pressure in real time. State detection adopts fiber optic sensing technology, and continuous monitoring of tube wall strain and temperature field is achieved through a grating array.

[0102] This embodiment implements an innovative time series partitioning method. The partitioning strategy is based on the multi-scale analysis theory, and the multi-resolution decomposition of the time series is achieved through wavelet transform. Data features are identified at different scales, and the optimal partitioning point is determined by the information entropy criterion. In order to improve the reliability of the partitioning, an adaptive segmentation algorithm is designed, and the global optimal partitioning is achieved through dynamic programming. The rationality of the partitioning result is verified by consistency test.

[0103] This embodiment designs an accurate shooting data sorting mechanism. The sorting process first establishes a multidimensional index structure, including three dimensions: timestamp, shooting parameters, and environmental conditions. The data organization adopts a B+ tree structure to support efficient time series retrieval and range query. In order to handle timestamp conflicts, a priority mechanism is designed to determine the time series relationship of data through shooting characteristics.

[0104] This embodiment builds a reliable data standardization framework. The standardization process uses an improved Z-score method to determine the center position and scale parameters through robust statistical estimation. The parameter estimation uses the M estimator to effectively suppress the influence of outliers. In order to maintain the physical interpretability of the data, a segmented standardization strategy is designed, and different standardization parameters are used in different shooting stages.

[0105] This embodiment implements an innovative prediction sequence generation strategy. The sequence construction adopts a sliding window mechanism, and the window size is determined by autocorrelation analysis. An overlapping strategy is adopted during the window sliding process, and the boundary effect is eliminated through weighted averaging. In order to improve the quality of the sequence, a data cleaning process is designed to identify and process abnormal data through multiple verifications.

[0106] This embodiment designs a complete prediction implementation scheme. The prediction process adopts a hierarchical structure, first establishing a trend prediction model at the macro level, and then superimposing local fluctuation terms at the micro level. Trend prediction uses an improved time response function to achieve dynamic characteristic description through parameter adaptation. Fluctuation prediction uses a conditional random field model to capture local correlation characteristics.

[0107] This embodiment implements the system's denormalization processing mechanism. The denormalization process adopts an inverse transformation strategy to ensure the physical rationality of the predicted value by maintaining monotonicity. The transformation parameters are determined by the distribution characteristics of the original data to ensure that the restored results meet the actual working conditions. In order to improve the restoration accuracy, an error compensation mechanism is designed to correct the system deviation through residual analysis.

[0108] This embodiment builds a reliable prediction result verification system. The verification process adopts a multi-dimensional evaluation strategy, including point prediction accuracy, interval prediction reliability and trend prediction accuracy. The accuracy evaluation adopts an improved mean square error index, which highlights the importance of key points through a weighted method. Reliability evaluation is achieved through confidence interval analysis, and trend evaluation adopts a consistency test method.

[0109] This embodiment establishes a reliable ablation wear assessment framework through precise data processing and systematic prediction. While ensuring prediction accuracy, this solution improves prediction reliability through multi-level optimization. The overall design fully considers the characteristics of artillery shooting and realizes full process optimization from data processing to prediction output.

[0110] The prediction framework of this embodiment has high accuracy and high reliability, providing a powerful tool for artillery life assessment. The prediction results are fully verified and analyzed to ensure the scientificity and reliability of the assessment. The implementation of this solution significantly improves the accuracy and stability of the prediction, providing reliable technical support for artillery maintenance decisions.

[0111] The prediction results of this embodiment show excellent time correlation and physical consistency, and achieve accurate characterization of the ablation and wear process through multi-level optimization. This solution has shown strong prediction capabilities in practical applications, providing a scientific basis for equipment life assessment and maintenance decisions, and achieving accurate assessment and prediction of equipment status.

[0112] In one embodiment of the barrel ablation wear prediction method based on the adaptive optimization grey model of the present application, see Figure 7 , and can also include the following: Step S701: constructing a coordinate system by using the firing number data as the horizontal coordinate and the barrel ablation wear prediction value as the vertical coordinate, using a cubic spline interpolation algorithm to continuously smooth discrete prediction data points, generating a continuous curve reflecting the trend of firing wear, and calculating the slope and curvature of the continuous curve in different intervals; Step S702: performing piecewise fitting processing on the shooting wear curve, calculating the tangent slope and curvature change rate of the fitting curve at each shooting number, comparing the tangent slope with a preset wear rate threshold, analyzing the acceleration characteristics of the barrel ablation wear based on the curvature change rate, and judging the ablation wear state of the gun barrel according to the wear change characteristics.

[0113] Optionally, this embodiment adopts an adaptive mapping strategy in the coordinate system construction stage. The coordinate axis design takes into account the data distribution characteristics, and the effective display of a wide range of data is achieved through logarithmic transformation. The coordinate system resolution is determined by data density analysis, and local magnification processing is used in key areas. In order to improve the visualization effect, a dynamic scaling mechanism is designed to achieve multi-scale observation through interactive operation.

[0114] This embodiment implements an innovative cubic spline interpolation method. The interpolation algorithm uses piecewise cubic Hermite polynomials and controls the curve shape through tension parameters. Node selection is based on curvature analysis, and node density is increased in areas of drastic changes. In order to maintain the physical rationality of the curve, a monotonicity constraint is designed, and the monotonicity of the curve is ensured by inequality constraints. Boundary conditions are set by natural boundaries to avoid excessive oscillation at the endpoints.

[0115] This embodiment designs an accurate curve smoothing strategy. Local weighted regression is used for smoothing, and the weight function uses the improved dual-weight Huber function. The window size is determined by cross-validation to achieve adaptive adjustment of the smoothing degree. In order to maintain data features, conformal filtering is introduced, and the key features are retained by curvature preservation. The rationality of the smoothing result is verified by residual analysis.

[0116] This embodiment builds a reliable slope calculation framework. The slope calculation adopts the central difference format and improves the calculation accuracy through Richardson extrapolation. To deal with the endpoint problem, a boundary processing strategy is designed to maintain the calculation stability through one-sided difference. The uncertainty of the slope estimation is evaluated through error propagation analysis to ensure the reliability of the calculation results.

[0117] This embodiment implements an innovative curvature analysis method. The curvature calculation adopts the differential geometry method and achieves accurate calculation through curve parameterization. In order to improve the calculation stability, a regularization scheme is designed to suppress the influence of noise by adding smoothing terms. The curvature features are extracted through multi-scale analysis to identify the change features at different scales.

[0118] This embodiment designs a complete segmented fitting strategy. The segmentation point selection is based on curvature extreme value analysis, and global optimal segmentation is achieved through dynamic programming. The fitting function form is determined according to the physical model, including linear terms, exponential terms, and power function terms. Parameter estimation uses weighted least squares method, and the weight design reflects data reliability.

[0119] This embodiment implements the tangent analysis method of the system. The tangent calculation uses analytical differentiation and symbolic calculation to improve accuracy. The slope threshold is determined by statistical analysis of historical data, taking into account the characteristics of different shooting conditions. The threshold update adopts an adaptive mechanism and realizes dynamic adjustment through online learning.

[0120] This embodiment builds a reliable acceleration feature extraction mechanism. Acceleration analysis is based on the curvature change rate and is achieved through second-order derivative estimation. Feature extraction uses wavelet analysis to identify change features of different scales through multi-resolution decomposition. Feature classification uses a fuzzy inference system to achieve state judgment through expert rules.

[0121] This embodiment establishes a reliable state assessment framework through precise curve analysis and systematic feature extraction. While ensuring the assessment accuracy, this solution improves the reliability of judgment through multi-level analysis. The overall design fully considers the characteristics of artillery ablation and realizes the optimization of the entire process from curve analysis to state judgment.

[0122] The evaluation framework of this embodiment has high accuracy and high reliability, providing a scientific basis for artillery maintenance decisions. The evaluation results are fully verified and analyzed to ensure the scientificity and reliability of the judgment. The implementation of this solution significantly improves the accuracy and stability of the status assessment, providing reliable technical support for equipment maintenance decisions.

[0123] The analysis results of this embodiment show excellent physical consistency and judgment reliability, and achieve accurate assessment of the ablation and wear status through multi-level analysis. This solution has shown strong analytical capabilities in practical applications, providing a scientific basis for equipment maintenance decisions and achieving accurate assessment and early warning of equipment status.

[0124] In order to accurately predict and evaluate the ablation and wear state of the gun barrel and provide a reliable decision-making basis for the use and maintenance of the gun, the present application provides an embodiment of a barrel ablation and wear prediction device based on an adaptive optimization grey model for realizing all or part of the contents of the barrel ablation and wear prediction method based on an adaptive optimization grey model, see Figure 8 The barrel ablation wear prediction device based on the adaptive optimization grey model specifically includes the following contents: A grey model building module 10 is used to use barrel ablation and wear data obtained from a shooting test as an original sequence, perform first-order accumulation on the original sequence to obtain an accumulated sequence, build a whitening time response function based on the accumulated sequence, calculate a weight coefficient of each data point in the original sequence by a new information priority criterion, adaptively optimize initial conditions according to the weight coefficients, substitute the optimized initial conditions into the whitening time response function, and build a non-equidistant grey prediction model; A grey model tuning module 20 is used to establish an unconstrained optimization problem based on a minimum mean square error sum criterion, take the error between the prediction result of the non-equidistant grey prediction model and the actual test data as an optimization target, use the steepest descent method to solve the unconstrained optimization problem to obtain a time parameter, substitute the time parameter into the non-equidistant grey prediction model, and obtain an improved adaptive grey prediction model; The barrel state prediction module 30 is used to read the firing number data of the artillery to be predicted, input the firing number data into the improved adaptive grey prediction model, obtain the barrel ablation and wear prediction values ​​corresponding to different firing numbers, establish a firing wear curve according to the prediction value and the firing number data, and judge the ablation and wear state of the artillery barrel based on the firing wear curve.

[0125] From the above description, it can be seen that the barrel ablation and wear prediction device based on the adaptive optimization gray model provided in the embodiment of the present application can establish a whitening time response function by performing first-order accumulation on the original sequence obtained from the shooting test. The new information priority criterion is innovatively used to calculate the data weight coefficient, realize the adaptive optimization of the initial conditions, and construct a non-equidistant gray prediction model. The system establishes an unconstrained optimization problem based on the minimum criterion of the mean sum of squared errors, and uses the steepest descent method to solve the optimal time parameters to improve the prediction accuracy of the model. The shooting wear curve is established in combination with the shooting number data, which realizes the accurate prediction and evaluation of the ablation and wear state of the gun barrel, and provides a reliable decision-making basis for the use and maintenance of the gun.

[0126] From the hardware level, in order to accurately predict and evaluate the ablation and wear state of the gun barrel and provide a reliable decision-making basis for the use and maintenance of the gun, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the barrel ablation and wear prediction method based on the adaptive optimization grey model, and the electronic device specifically includes the following contents: Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the barrel ablation wear prediction device based on the adaptive optimization grey model and related devices such as the core business system, user terminal and related database; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., and the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the barrel ablation wear prediction method based on the adaptive optimization grey model and the embodiment of the barrel ablation wear prediction device based on the adaptive optimization grey model in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0127] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0128] In practical applications, part of the barrel ablation wear prediction method based on the adaptive optimization grey model can be executed on the electronic device side as described above, or all operations can be completed in the client device. The selection can be made based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0129] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0130] Fig. 9 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig. 9 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig. 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0131] In one embodiment, the barrel ablation wear prediction method function based on the adaptive optimization grey model can be integrated into the central processor 9100. The central processor 9100 can be configured to perform the following control: Step S101: taking barrel ablation and wear data obtained from a shooting test as an original sequence, performing first-order accumulation on the original sequence to obtain an accumulated sequence, constructing a whitening time response function based on the accumulated sequence, calculating a weight coefficient of each data point in the original sequence by a new information priority criterion, adaptively optimizing initial conditions according to the weight coefficients, substituting the optimized initial conditions into the whitening time response function, and constructing a non-equidistant grey prediction model; Step S102: establishing an unconstrained optimization problem based on the minimum criterion of the mean square error sum, taking the error between the prediction result of the non-equidistant grey prediction model and the actual test data as the optimization target, solving the unconstrained optimization problem by the steepest descent method to obtain a time parameter, substituting the time parameter into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model; Step S103: Read the firing number data of the artillery to be predicted, input the firing number data into the improved adaptive grey prediction model, obtain the barrel ablation wear prediction values ​​corresponding to different firing numbers, establish a firing wear curve according to the prediction value and the firing number data, and judge the ablation wear state of the artillery barrel based on the firing wear curve.

[0132] From the above description, it can be seen that the electronic device provided in the embodiment of the present application establishes a whitening time response function by performing first-order accumulation on the original sequence obtained from the shooting test. The new information priority criterion is innovatively adopted to calculate the data weight coefficient, realize the adaptive optimization of the initial conditions, and construct a non-equidistant gray prediction model. The system establishes an unconstrained optimization problem based on the minimum criterion of the mean sum of squared errors, and uses the steepest descent method to solve the optimal time parameters to improve the prediction accuracy of the model. The shooting wear curve is established in combination with the shooting number data, which realizes the accurate prediction and evaluation of the ablation and wear state of the artillery barrel, and provides a reliable decision-making basis for the use and maintenance of artillery.

[0133] In another embodiment, the barrel ablation and wear prediction device based on the adaptive optimization gray model can be configured separately from the central processing unit 9100. For example, the barrel ablation and wear prediction device based on the adaptive optimization gray model can be configured as a chip connected to the central processing unit 9100, and the function of the barrel ablation and wear prediction method based on the adaptive optimization gray model can be realized through the control of the central processing unit.

[0134] like Fig. 9 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig. 9 In addition, the electronic device 9600 may also include Fig. 9 For components not shown, reference may be made to the prior art.

[0135] like Fig. 9 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0136] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0137] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0138] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0139] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0140] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0141] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0142] The embodiment of the present application also provides a computer-readable storage medium capable of implementing all the steps of the barrel ablation wear prediction method based on the adaptive optimization grey model in the above embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, all the steps of the barrel ablation wear prediction method based on the adaptive optimization grey model in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented: Step S101: taking barrel ablation and wear data obtained from a shooting test as an original sequence, performing first-order accumulation on the original sequence to obtain an accumulated sequence, constructing a whitening time response function based on the accumulated sequence, calculating a weight coefficient of each data point in the original sequence by a new information priority criterion, adaptively optimizing initial conditions according to the weight coefficients, substituting the optimized initial conditions into the whitening time response function, and constructing a non-equidistant grey prediction model; Step S102: establishing an unconstrained optimization problem based on the minimum criterion of the mean square error sum, taking the error between the prediction result of the non-equidistant grey prediction model and the actual test data as the optimization target, solving the unconstrained optimization problem by the steepest descent method to obtain a time parameter, substituting the time parameter into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model; Step S103: Read the firing number data of the artillery to be predicted, input the firing number data into the improved adaptive grey prediction model, obtain the barrel ablation wear prediction values ​​corresponding to different firing numbers, establish a firing wear curve according to the prediction value and the firing number data, and judge the ablation wear state of the artillery barrel based on the firing wear curve.

[0143] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application establishes a whitening time response function by performing first-order accumulation on the original sequence obtained from the shooting test. The new information priority criterion is innovatively adopted to calculate the data weight coefficient, realize the adaptive optimization of the initial conditions, and construct a non-equidistant gray prediction model. The system establishes an unconstrained optimization problem based on the minimum criterion of the mean sum of squared errors, and uses the steepest descent method to solve the optimal time parameters to improve the prediction accuracy of the model. The shooting wear curve is established in combination with the shooting number data, which realizes the accurate prediction and evaluation of the ablation and wear state of the artillery barrel, and provides a reliable decision-making basis for the use and maintenance of artillery.

[0144] The embodiment of the present application also provides a computer program product capable of implementing all the steps of the barrel ablation wear prediction method based on the adaptive optimization grey model in the above embodiment, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the barrel ablation wear prediction method based on the adaptive optimization grey model are implemented. For example, the computer program / instruction implements the following steps: Step S101: taking barrel ablation and wear data obtained from a shooting test as an original sequence, performing first-order accumulation on the original sequence to obtain an accumulated sequence, constructing a whitening time response function based on the accumulated sequence, calculating a weight coefficient of each data point in the original sequence by a new information priority criterion, adaptively optimizing initial conditions according to the weight coefficients, substituting the optimized initial conditions into the whitening time response function, and constructing a non-equidistant grey prediction model; Step S102: establishing an unconstrained optimization problem based on the minimum criterion of the mean square error sum, taking the error between the prediction result of the non-equidistant grey prediction model and the actual test data as the optimization target, solving the unconstrained optimization problem by the steepest descent method to obtain a time parameter, substituting the time parameter into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model; Step S103: Read the firing number data of the artillery to be predicted, input the firing number data into the improved adaptive grey prediction model, obtain the barrel ablation wear prediction values ​​corresponding to different firing numbers, establish a firing wear curve according to the prediction value and the firing number data, and judge the ablation wear state of the artillery barrel based on the firing wear curve.

[0145] From the above description, it can be seen that the computer program product provided in the embodiment of the present application establishes a whitening time response function by performing first-order accumulation on the original sequence obtained from the shooting test. The new information priority criterion is innovatively used to calculate the data weight coefficient, realize the adaptive optimization of the initial conditions, and construct a non-equidistant gray prediction model. The system establishes an unconstrained optimization problem based on the minimum criterion of the mean sum of squared errors, and uses the steepest descent method to solve the optimal time parameters to improve the prediction accuracy of the model. The shooting wear curve is established in combination with the shooting number data, which realizes the accurate prediction and evaluation of the ablation and wear state of the artillery barrel, and provides a reliable decision-making basis for the use and maintenance of artillery.

[0146] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0150] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A barrel ablation wear prediction method based on adaptive optimization grey model, characterized in that: The method comprises: The barrel ablation and wear data obtained from the shooting test is used as the original sequence, the original sequence is first-order accumulated to obtain an accumulated sequence, a whitening time response function is constructed based on the accumulated sequence, a weight coefficient of each data point in the original sequence is calculated by a new information priority criterion, the initial conditions are adaptively optimized according to the weight coefficients, the optimized initial conditions are substituted into the whitening time response function, and a non-equidistant grey prediction model is constructed; An unconstrained optimization problem is established based on the minimum criterion of the average square error sum, and the error between the prediction result of the non-equidistant grey prediction model and the actual test data is used as the optimization target. The unconstrained optimization problem is solved by the steepest descent method to obtain a time parameter, and the time parameter is substituted into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model; The firing number data of the artillery to be predicted is read, the firing number data is input into the improved adaptive grey prediction model, the barrel ablation wear prediction values ​​corresponding to different firing numbers are obtained, a firing wear curve is established according to the prediction value and the firing number data, and the ablation wear state of the artillery barrel is judged based on the firing wear curve.

2. The barrel ablation wear prediction method based on the adaptive optimization grey model according to claim 1 is characterized in that: The barrel ablation and wear data obtained from the shooting test is used as an original sequence, the original sequence is first-order accumulated to obtain an accumulated sequence, a whitening time response function is constructed based on the accumulated sequence, and a weight coefficient of each data point in the original sequence is calculated by a new information priority criterion, including: Obtain barrel diameter change data during the artillery firing process, sort the barrel diameter change data according to the number of shots fired to construct a time series data sequence, calculate the barrel diameter change amount between two adjacent shots based on the time series data sequence to obtain an original ablation wear sequence, and mark the original ablation wear sequence as X(0); Each data point in the original ablation and wear sequence X(0) is added to all the previous data points to obtain the cumulative sequence X(1), the means of the adjacent data of the cumulative sequence X(1) are calculated to generate the adjacent mean sequence Z(1), the adjacent mean sequence Z(1) is substituted into the whitening differential equation to construct the whitening time response function, and the influence weight of each data point in the original ablation and wear sequence X(0) on the prediction result is calculated based on the data recency principle.

3. The barrel ablation wear prediction method based on the adaptive optimization grey model according to claim 1 is characterized in that: The adaptive optimization of the initial conditions according to the weight coefficients, substituting the optimized initial conditions into the whitening time response function, and constructing a non-equidistant grey prediction model includes: Based on the weight coefficient, an initial condition expression for adaptive optimization is constructed, the data in the cumulative sequence X(1) are multiplied by the corresponding weight coefficients in sequence and the sum is calculated to obtain the optimized initial condition value, a non-equidistant model equation group is established using the adjacent mean sequence Z(1), and the least square method is used to solve the equation group to obtain the development coefficient and the gray action; The optimized initial condition value, the development coefficient and the gray action are substituted into the whitening time response function to construct a prediction model expression considering non-equidistant characteristics. The prediction model expression is restored to obtain a non-equidistant gray prediction model, which outputs the prediction value of barrel ablation and wear under different numbers of firing rounds.

4. The barrel ablation wear prediction method based on the adaptive optimization grey model according to claim 1 is characterized in that: The unconstrained optimization problem is established based on the minimum criterion of the mean square error sum, the error between the prediction result of the non-equidistant grey prediction model and the actual test data is used as the optimization target, and the steepest descent method is used to solve the unconstrained optimization problem to obtain the time parameter, including: Comparing the predicted value of the non-equidistant grey prediction model with the actual test data, calculating the relative error under each shooting number, performing square sum operation on all relative errors to obtain an overall error function, establishing an unconstrained optimization model with the overall error function as the objective function, and substituting the time parameter as the optimization variable into the unconstrained optimization model; The objective function of the unconstrained optimization model is differentiated to obtain a gradient vector, and the value of the time parameter is iteratively updated based on the steepest descent method. In each iteration, the algorithm moves along the negative gradient direction and determines the optimal step size through a one-dimensional search. When the norm of the gradient vector is less than the preset accuracy requirement or the maximum number of iterations is reached, the optimal time parameter value is output.

5. The barrel ablation wear prediction method based on the adaptive optimization grey model according to claim 1 is characterized in that: Substituting the time parameter into the non-equidistant grey prediction model to obtain an improved adaptive grey prediction model comprises: Construct a prediction model expression with time parameters, substitute the optimal time parameters, development coefficient and grey action into the prediction model expression, modify the prediction model expression according to the relationship between the prediction values ​​of the previous and next time points, and generate an adaptive grey prediction model considering time correlation; The parameters in the adaptive grey prediction model are normalized, the optimized initial condition values ​​are substituted into the adaptive grey prediction model to calculate the predicted values, and the parameters of the adaptive grey prediction model are fine-tuned based on the predicted values ​​to obtain an improved grey prediction model with adaptive optimization characteristics.

6. The barrel ablation wear prediction method based on the adaptive optimization grey model according to claim 1 is characterized in that: The method of reading the firing number data of the artillery to be predicted, inputting the firing number data into the improved adaptive grey prediction model, and obtaining the barrel ablation wear prediction values ​​corresponding to different firing numbers includes: Acquire historical firing data of the artillery to be predicted from the artillery data acquisition system, divide the historical firing data into time series to obtain firing number data of different time periods, arrange the firing number data in time sequence to form a sequence to be predicted, and perform data standardization processing on the sequence to be predicted; The standardized sequence to be predicted is input into the improved adaptive grey prediction model, the barrel ablation wear amount corresponding to each time point is calculated based on the prediction function of the model, and the predicted ablation wear amount is de-standardized to obtain the actual barrel ablation wear prediction value.

7. The barrel ablation wear prediction method based on the adaptive optimization grey model according to claim 1 is characterized in that: The step of establishing a firing wear curve according to the predicted value and the firing number data, and judging the ablation wear state of the gun barrel based on the firing wear curve, comprises: A coordinate system is constructed by using the firing number data as the horizontal coordinate and the barrel ablation wear prediction value as the vertical coordinate, and a cubic spline interpolation algorithm is used to continuously smooth discrete prediction data points to generate a continuous curve reflecting the trend of firing wear changes, and the slope and curvature of the continuous curve in different intervals are calculated; The shooting wear curve is subjected to piecewise fitting processing, the tangent slope and curvature change rate of the fitting curve at each shooting number are calculated, the tangent slope is compared with a preset wear rate threshold, the acceleration characteristics of the barrel ablation wear are analyzed based on the curvature change rate, and the ablation wear state of the gun barrel is determined according to the wear change characteristics.

8. A barrel ablation wear prediction device based on adaptive optimization grey model, characterized in that: The device comprises: A grey model building module is used to use barrel ablation and wear data obtained from a shooting test as an original sequence, perform first-order accumulation on the original sequence to obtain an accumulated sequence, build a whitening time response function based on the accumulated sequence, calculate a weight coefficient of each data point in the original sequence by a new information priority criterion, adaptively optimize initial conditions according to the weight coefficients, substitute the optimized initial conditions into the whitening time response function, and build a non-equidistant grey prediction model; A grey model tuning module is used to establish an unconstrained optimization problem based on the minimum criterion of the average square error sum, take the error between the prediction result of the non-equidistant grey prediction model and the actual test data as the optimization target, use the steepest descent method to solve the unconstrained optimization problem to obtain a time parameter, substitute the time parameter into the non-equidistant grey prediction model, and obtain an improved adaptive grey prediction model; The barrel state prediction module is used to read the firing number data of the artillery to be predicted, input the firing number data into the improved adaptive grey prediction model, obtain the barrel ablation wear prediction values ​​corresponding to different firing numbers, establish a firing wear curve according to the prediction value and the firing number data, and judge the ablation wear state of the gun barrel based on the firing wear curve.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the barrel ablation and wear prediction method based on the adaptive optimization grey model as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the barrel ablation wear prediction method based on the adaptive optimization grey model as described in any one of claims 1 to 7 are implemented.

Images