Artillery barrel wear prediction method and device based on escape optimization

Abstract

The embodiment of the invention provides an artillery barrel wear prediction method and device based on escape optimization, and the method comprises the steps: converting the temperature, pressure and charge features into a comprehensive load feature vector, and constructing a corresponding relation sample set with the ablation wear amount; and global optimization of prediction model parameters is realized by designing a search space of an escape optimization algorithm in combination with a cross validation method and a fitness sorting mechanism. The system adopts Bayesian correction to improve prediction precision, extracts wear change rate and acceleration characteristics based on a degradation trend curve, and establishes a reliable ablation wear state early warning mechanism. According to the method, the defects of a traditional prediction technology in parameter optimization and state early warning are effectively overcome, and a scientific basis is provided for use and maintenance of the artillery barrel.

Description

Technical Field

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

[0002] The existing gun barrel wear prediction technology has obvious shortcomings. Traditional prediction methods often use a single model or a simple optimization algorithm, which fails to fully utilize the characteristic information of multi-source shooting data, resulting in insufficient prediction accuracy and robustness. At the same time, the existing methods are prone to fall into local optimality when dealing with high-dimensional nonlinear prediction problems, making it difficult to obtain a global optimal solution.

[0003] In addition, existing technologies also have bottlenecks in parameter optimization and prediction result correction. Most methods lack an effective parameter search mechanism and cannot dynamically adjust the optimization strategy according to sample characteristics, which affects model performance. The uncertainty quantification and correction methods of prediction results are relatively simple, and lack a systematic error analysis and correction mechanism.

[0004] The existing system has technical shortcomings in comprehensive load assessment and status warning. The lack of unified modeling of multi-dimensional characteristics such as temperature, pressure, and charge makes it difficult to accurately describe the wear evolution law under actual use conditions, affecting the accuracy and timeliness of warning. Solving these problems is of great significance to improving the safety of artillery use and maintenance efficiency. Summary of the invention

[0005] In response to the problems in the prior art, the present application provides a method and device for predicting gun barrel wear based on escape optimization, which can establish a reliable ablation wear status early warning mechanism, effectively solve the shortcomings of traditional prediction technology in parameter optimization and status early warning, and provide a scientific basis for the use and maintenance of gun barrels.

[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 gun barrel wear prediction method based on escape optimization, comprising: Collect historical ablation and wear data of multiple guns at different firing stages, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, convert the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, construct a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and use a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; A search space of an escape optimization algorithm is established based on the training sample subset, an initial position distribution of a search particle swarm is set, position information of the search particle swarm is mapped to prediction model parameters, a prediction error corresponding to each particle is calculated using a cross-validation method, the prediction error is used as a particle fitness function, the particle swarm is sorted according to the fitness function, the particle with the best fitness is selected as the global optimal solution, the speed and position of the particle are updated based on the escape mechanism in each round of iteration, the search is stopped when the number of iterations reaches a preset value, and the global optimal solution that finally converges is used as the optimal prediction model parameter; The real-time firing data of the artillery to be predicted is read, and temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters are extracted from the real-time firing data. The temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters are converted into a comprehensive load characteristic vector according to the same conversion rule. The comprehensive load characteristic vector is input into a prediction model with the optimal prediction model parameters, and the ablation wear prediction values ​​at different shooting moments are calculated. The ablation wear prediction values ​​are subjected to Bayesian correction to obtain the corrected prediction results. Based on the corrected prediction results, an ablation wear degradation trend curve of the artillery barrel is constructed, and a wear change rate characteristic and a wear acceleration characteristic are extracted from the degradation trend curve. According to the wear change rate characteristic and the wear acceleration characteristic, an early warning is given to the ablation wear state of the artillery barrel.

[0007] Furthermore, the historical ablation and wear data of multiple guns at different firing stages are collected, temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters in the historical ablation and wear data are extracted, the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters are converted into comprehensive load characteristic vectors, and a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation and wear amount is constructed, including: During the firing process of multiple guns, barrel temperature data, bore pressure data and charging propellant data are collected, the barrel temperature data are normalized according to the overload factor to obtain a temperature characteristic parameter, the bore pressure data are normalized according to the peak pressure to obtain a pressure characteristic parameter, the charging propellant data are normalized according to the quality standard to obtain a charging characteristic parameter, and the temperature characteristic parameter, the pressure characteristic parameter and the charging characteristic parameter are reduced in dimension by using a principal component analysis method to obtain a characteristic vector reflecting the comprehensive load of the firing process; The feature vector is sampled in time series and the accumulated ablation wear amount corresponding to each shooting moment is calculated, a mapping relationship between the feature vector sequence and the ablation wear amount is established, a stratified random sampling method is used to divide the mapping relationship into a training sample set and a verification sample set, the data in the training sample set and the verification sample set are normalized and standardized, and a corresponding relationship sample set between the standardized feature vector and the ablation wear amount is constructed.

[0008] Furthermore, the adopting of a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset includes: Based on the distribution characteristics of the ablation wear amount in the corresponding relationship sample set, the sample data is divided into multiple wear level layers, the proportion of sample data in each wear level layer is calculated, the sample data in the wear level layer is randomly shuffled, and a corresponding number of sample data are extracted from each wear level layer according to a preset ratio of training set to verification set, and the extracted sample data are respectively used to construct a training sample subset and a verification sample subset; A data balance check is performed on the training sample subset and the verification sample subset, and the distribution difference of each wear level layer sample in the training set and the verification set is calculated. When the distribution difference exceeds a preset threshold, stratified sampling is performed again until the data distribution of the training sample subset and the verification sample subset meets the balance requirement.

[0009] Furthermore, the search space of the escape optimization algorithm is established based on the training sample subset, the initial position distribution of the search particle swarm is set, the position information of the search particle swarm is mapped to the prediction model parameters, the prediction error corresponding to each particle is calculated by a cross-validation method, the prediction error is used as the particle fitness function, the particle swarm is sorted according to the fitness function, and the particle with the best fitness is selected as the global optimal solution, including: Determining the boundary range of the search space according to the characteristic dimension of the training sample subset, randomly generating an initial particle swarm in the search space, mapping each dimensional component of the initial particle swarm to a weight parameter and a bias parameter of a prediction model, constructing a parameter matrix of the prediction model based on the weight parameter and the bias parameter, and dividing the training sample subset into a plurality of training sets and a test set using a K-fold cross validation method; The training set is input into the prediction model for training, the prediction error of the model is calculated using the test set, the cross-validation process is repeated for each particle to obtain the average prediction error, the average prediction error is used as the fitness value of the particle, the particle group is sorted in ascending order based on the fitness value, and the particle position with the smallest fitness value is selected as the global optimal solution of the current iteration.

[0010] Furthermore, in each round of iteration, the particle speed and position are updated based on the escape mechanism, and the search is stopped when the number of iterations reaches a preset value, and the final converged global optimal solution is used as the optimal prediction model parameter, including: Calculate the Euclidean distance between each particle and the global optimal solution, trigger the escape mechanism when the Euclidean distance is less than a preset escape threshold, add an escape disturbance term in the direction of the particle's current velocity, update the particle's velocity vector according to the escape disturbance term and the inertia weight, calculate the particle's new position based on the velocity vector, map the position components that exceed the search space boundary, and convert the mapped position information into new prediction model parameters; The fitness value of the particle swarm after each round of iteration is recalculated and the global optimal solution is updated. The historical change trajectory of the global optimal solution is recorded. When the change of the global optimal solution in multiple consecutive rounds of iterations is less than the convergence threshold or the number of iterations reaches the preset upper limit, the search is stopped and the global optimal solution at the final convergence is used as the optimized prediction model parameter.

[0011] Furthermore, the temperature characteristic parameter, the pressure characteristic parameter and the charge characteristic parameter are converted into a comprehensive load characteristic vector according to the same conversion rule, the comprehensive load characteristic vector is input into a prediction model with the optimal prediction model parameters, ablation wear prediction values ​​at different shooting moments are calculated, and ablation wear prediction values ​​are subjected to Bayesian correction to obtain a corrected prediction result, including: Substituting the temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters collected in real time into the principal component analysis model, performing dimension reduction transformation on the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters according to the characteristic conversion matrix obtained in the training phase, generating a characteristic vector reflecting the comprehensive load of the current shooting process, inputting the characteristic vector into the prediction model, and calculating the ablation wear prediction value at the current moment based on the optimal prediction model parameters; A Bayesian prior distribution model is constructed, and historical prediction data is used as prior information to calculate the posterior probability distribution of the ablation wear prediction value. The prediction result is corrected based on the mean and variance of the posterior probability distribution, and the corrected prediction results at different shooting moments are calculated in sequence using a sliding time window method.

[0012] Further, the constructing of the gun barrel ablation wear degradation trend curve based on the corrected prediction result, extracting the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and giving an early warning of the ablation wear state of the gun barrel according to the wear change rate characteristics and the wear acceleration characteristics, comprises: The corrected prediction result sequence is subjected to time series fitting processing, and a degradation trend curve reflecting the ablation wear evolution process is constructed by using a polynomial regression method. The first-order derivative of the degradation trend curve at different times is calculated to obtain the wear change rate characteristics, and the second-order derivative of the degradation trend curve at different times is calculated to obtain the wear acceleration characteristics. Based on the wear change rate characteristics and the wear acceleration characteristics, a quantitative evaluation index of the ablation wear state of the gun barrel is established. The wear change rate threshold and the wear acceleration threshold are determined based on statistical analysis of historical data. When the wear change rate characteristic exceeds the wear change rate threshold or the wear acceleration characteristic exceeds the wear acceleration threshold, the early warning mechanism is triggered, and the early warning level of the ablation wear state of the gun barrel is determined based on the preset early warning level classification standard.

[0013] In a second aspect, the present application provides a gun barrel wear prediction device based on escape optimization, comprising: A data set construction module is used to collect historical ablation and wear data of multiple guns at different firing stages, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, convert the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, construct a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and use a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; An optimal parameter determination module is used to establish a search space for the escape optimization algorithm based on the training sample subset, set the initial position distribution of the search particle group, map the position information of the search particle group to the prediction model parameters, use the cross-validation method to calculate the prediction error corresponding to each particle, use the prediction error as the particle fitness function, sort the fitness of the particle group according to the fitness function, select the particle with the best fitness as the global optimal solution, update the speed and position of the particle based on the escape mechanism in each round of iteration, stop the search when the number of iterations reaches a preset value, and use the finally converged global optimal solution as the optimal prediction model parameter; The wear state prediction module is used to read the real-time shooting data of the artillery to be predicted, extract the temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the real-time shooting data, generate a comprehensive load characteristic vector according to the same conversion rule for the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters, input the comprehensive load characteristic vector into the prediction model with the optimal prediction model parameters, calculate the ablation wear prediction values ​​at different shooting moments, perform Bayesian correction on the ablation wear prediction values ​​to obtain the corrected prediction results, construct the ablation wear degradation trend curve of the artillery barrel based on the corrected prediction results, extract the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and issue an early warning for the ablation wear state of the artillery barrel according to the wear change rate characteristics and the wear acceleration characteristics.

[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 gun barrel wear prediction method based on escape optimization 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 gun barrel wear prediction method based on escape optimization.

[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 gun barrel wear prediction method based on escape optimization.

[0017] It can be seen from the above technical solution that the present application provides a method and device for predicting the wear of a gun barrel based on escape optimization, which converts the temperature, pressure and charge characteristics into a comprehensive load characteristic vector to construct a sample set corresponding to the amount of ablation wear. By designing the search space of the escape optimization algorithm, combined with the cross-validation method and the fitness ranking mechanism, the global optimization of the prediction model parameters is achieved. The system uses Bayesian correction to improve the prediction accuracy, extracts the wear change rate and acceleration characteristics based on the degradation trend curve, and establishes a reliable ablation wear status warning mechanism. This method effectively solves the shortcomings of traditional prediction technology in parameter optimization and status warning, and provides a scientific basis for the use and maintenance of gun barrels. 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 gun barrel wear prediction method based on escape optimization in the embodiment of the present application; Figure 2 This is a second flow chart of a gun barrel wear prediction method based on escape optimization in an embodiment of the present application; Figure 3 The third flowchart of the gun barrel wear prediction method based on escape optimization in the embodiment of the present application; Figure 4 This is a fourth flow chart of a gun barrel wear prediction method based on escape optimization in an embodiment of the present application; Figure 5 FIG5 is a flow chart of a gun barrel wear prediction method based on escape optimization in an embodiment of the present application; Figure 6 This is a sixth flow chart of the gun barrel wear prediction method based on escape optimization in an embodiment of the present application; Figure 7 FIG7 is a flow chart of a gun barrel wear prediction method based on escape optimization in an embodiment of the present application; Figure 8 It is a structural diagram of a gun barrel wear prediction device based on escape optimization 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 the wear of a gun barrel based on escape optimization, which converts the temperature, pressure and charge characteristics into a comprehensive load characteristic vector to construct a sample set corresponding to the amount of ablation wear. By designing the search space of the escape optimization algorithm, combined with the cross-validation method and the fitness ranking mechanism, the global optimization of the prediction model parameters is achieved. The system uses Bayesian correction to improve the prediction accuracy, extracts the wear change rate and acceleration characteristics based on the degradation trend curve, and establishes a reliable ablation wear status warning mechanism. This method effectively solves the shortcomings of traditional prediction technology in parameter optimization and status warning, and provides a scientific basis for the use and maintenance of gun barrels.

[0024] In order to establish a reliable ablation wear status warning mechanism, effectively solve the deficiencies of traditional prediction technology in parameter optimization and status warning, and provide a scientific basis for the use and maintenance of artillery barrels, this application provides an embodiment of a gun barrel wear prediction method based on escape optimization, see Figure 1The gun barrel wear prediction method based on escape optimization specifically includes the following contents: Step S101: collecting historical ablation and wear data of multiple guns at different firing stages, extracting temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, converting the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, constructing a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and using a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; Optionally, this embodiment uses a distributed data acquisition system to monitor the artillery firing process in all directions. The barrel temperature is collected using a K-type thermocouple array, with 16 measuring points arranged along the axial direction and 4 layers of measuring points set radially. The thermocouple has a measurement range of -50°C to 1100°C, and the sampling frequency is set to 100Hz to ensure accurate capture of transient temperature changes. The installation position of each measuring point is optimized through thermal field analysis to achieve accurate monitoring of the temperature field of key parts.

[0025] This embodiment uses a piezoelectric sensor in the bore pressure measurement, and the installation positions include two key positions: the bottom and the middle of the bore. The sensor range covers 0-600MPa, and the sampling frequency is set to 200kHz to achieve high-precision acquisition of pressure fluctuations. The signal conditioning circuit adopts a charge amplifier configuration, which has excellent linearity and temperature stability.

[0026] This embodiment uses a high-precision weighing system to collect the data of the loaded propellant, and the weighing accuracy reaches 0.1g. The influence of ambient temperature and humidity is considered during the weighing process, and the measurement reliability is improved through a compensation algorithm. The charge data collection also includes charge temperature measurement, which is used to analyze the influence of temperature on charge performance.

[0027] This embodiment designs an innovative temperature feature extraction method. Temperature normalization uses an overload factor F_t = (T-T_ref) / (T_max-T_ref), where T is the measured temperature, T_ref is the ambient reference temperature, and T_max is the maximum temperature allowed by the material. This normalization method takes into account the material properties and makes the features more physically meaningful.

[0028] This embodiment realizes the systematic extraction of pressure features. The pressure features are processed by peak normalization: F_p = P / P_max, where P is the measured pressure and P_max is the designed maximum pressure. The pressure waveform features are considered in the feature extraction process, including multiple dimensions such as rise time, duration and pressure gradient.

[0029] This embodiment innovatively realizes the extraction of charge features. The charge feature normalization adopts the mass standard method: F_m = M / M_s, where M is the actual charge amount and Ms is the standard charge amount. The feature extraction takes into account the influence of charge density and filling coefficient to achieve a comprehensive characterization of the charge working condition.

[0030] This embodiment uses an improved principal component analysis method for feature fusion. The feature weights are determined adaptively using the variance contribution rate, and the principal components with a cumulative contribution rate of 95% are retained. The feature conversion matrix W is obtained by orthogonalizing the eigenvectors to ensure the independence of the converted features.

[0031] This embodiment designs a timing sampling strategy for the system. The sampling interval is determined based on the characteristic time scale of the ablation process to ensure that key state changes are captured. The cumulative ablation amount calculation takes into account the material removal rate and is achieved through an integral method.

[0032] This embodiment implements an innovative stratified sampling scheme. First, the stratification boundary is determined based on the statistical distribution of the ablation amount, and the samples are divided into multiple wear level layers using the K-means clustering method. Random sampling is used within each layer, and the sampling ratio is proportional to the number of samples in the layer.

[0033] This embodiment adopts a data balance assessment method. By calculating the distribution difference of samples in each layer in the training set and the validation set: D = Σ|P_train,i - P_valid,i|, where P_train,i and P_valid,i are the proportions of samples in the i-th layer in the training set and the validation set respectively. When D exceeds the preset threshold, the resampling mechanism is triggered.

[0034] This embodiment establishes a reliable sample set construction framework through systematic data collection and processing. While ensuring data quality, this solution provides comprehensive feature expression through multi-level analysis. The overall design fully considers the characteristics of artillery ablation and realizes the optimization of the entire process from data collection to sample construction.

[0035] The sample set of this embodiment shows excellent representativeness and balance, and achieves accurate characterization of the ablation process through multi-level analysis. This solution shows strong data processing capabilities in practical applications and provides a reliable data basis for subsequent model training. The balanced distribution of samples provides comprehensive training information for model learning.

[0036] The data processing framework of this embodiment has good adaptability and scalability, and can meet the data collection requirements under different working conditions. Through feature adaptation and sample balancing, accurate characterization of complex ablation processes is achieved, providing reliable data support for model training.

[0037] Step S102: establishing a search space of the escape optimization algorithm based on the training sample subset, setting the initial position distribution of the search particle swarm, mapping the position information of the search particle swarm to the prediction model parameters, using the cross-validation method to calculate the prediction error corresponding to each particle, using the prediction error as the particle fitness function, sorting the particle swarm according to the fitness function, selecting the particle with the best fitness as the global optimal solution, updating the speed and position of the particle based on the escape mechanism in each round of iteration, stopping the search when the number of iterations reaches a preset value, and taking the finally converged global optimal solution as the optimal prediction model parameters; Optionally, this embodiment constructs a multidimensional search space based on the feature dimensions of the training samples. The value range of each dimension is determined by feature analysis: Xi_min = μi - 3σi, Xi_max = μi + 3σi, where μi and σi are the mean and standard deviation of the i-th dimension feature, respectively. This boundary setting method ensures that 99.7% of the feature distribution is covered while avoiding an excessively large search space.

[0038] This embodiment uses an improved initialization strategy to generate a search particle swarm. The particle position uses the Latin hypercube sampling method to ensure that the initial particles are evenly distributed in the search space. The velocity of each particle is initialized using a Gaussian distribution: Vi ~N(0, 0.1×|Xi_max - Xi_min|), where Vi is the i-th dimension velocity component.

[0039] This embodiment innovatively implements a parameter mapping mechanism. The mapping of position vectors to model parameters uses a piecewise function: W = sigmoid(X) for weight parameters and b = tanh(X) for bias parameters. This mapping method ensures the rationality of the parameter value range while maintaining the continuity of the parameter space.

[0040] This embodiment designs a complete cross-validation framework. A 5-fold cross-validation method is used to randomly divide the training samples into 5 subsets. During the validation process, one subset is selected as the test set and the rest are used as the training set. The prediction error uses the weighted mean square error: E = Σwi(yi - ŷi)² / N, where wi is the sample weight, reflecting the importance of different wear levels.

[0041] This embodiment implements an innovative fitness evaluation scheme. The fitness function combines prediction accuracy and model complexity: F = E + λ×R, where E is the prediction error, R is the parameter regularization term, and λ is the balance factor. This design avoids overfitting and improves the generalization ability of the model.

[0042] This embodiment adopts an improved escape mechanism update strategy. The distance between the particle and the global optimal solution is calculated by cosine similarity: cos_θ = X·Xg / (|X|·|Xg|), where X is the particle position and Xg is the global optimal solution. The escape mechanism is triggered when cos_θ is greater than a preset threshold.

[0043] This embodiment innovatively designs an escape disturbance term. The disturbance direction is orthogonal to the current velocity direction: D = V×n, where V is the velocity vector and n is a unit orthogonal vector. The disturbance amplitude is related to the current fitness of the particle: |D| = α×exp(-β×F), where F is the fitness value, and α and β are control parameters.

[0044] This embodiment implements an adaptive speed update strategy. The speed update formula is: V = w×V + c1×r1×(PX)+ c2×r2×(GX) + D, where w is the inertia weight, which decays with the number of iterations; c1 and c2 are acceleration coefficients; r1 and r2 are random numbers; P is the individual optimal solution; G is the global optimal solution; and D is the escape disturbance term.

[0045] This embodiment uses a boundary processing mechanism to ensure search stability. When the particle position exceeds the search space, a rebound strategy is adopted: Xi = Xi_min + r×(Xi_max - Xi_min), where r is a random number between (0,1). This processing method maintains the diversity of the particle swarm.

[0046] This embodiment designs multiple convergence criteria. In addition to limiting the number of iterations, the change rate of the global optimal solution is also monitored: ΔG = |Gt - Gt-1| / |Gt-1|, where Gt is the global optimal solution of the current iteration. Convergence is determined when ΔG is less than a threshold for multiple consecutive rounds.

[0047] This embodiment achieves an efficient parameter optimization process through innovative optimization strategies. This solution effectively avoids local optimality through the escape mechanism and improves the global search capability. The overall design fully considers the characteristics of the prediction model and achieves efficient exploration of the parameter space.

[0048] The optimization results of this embodiment show excellent convergence and stability, and accurate parameter selection of the prediction model is achieved through multi-level optimization. This solution shows strong optimization capabilities in practical applications and provides reliable guarantee for model performance.

[0049] The optimization framework of this embodiment has good versatility and robustness, and can adapt to the parameter optimization requirements of different feature spaces. Through the escape mechanism and adaptive strategy, efficient search of complex parameter spaces is achieved, providing reliable parameter support for the prediction model.

[0050] Step S103: read the real-time firing data of the artillery to be predicted, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the real-time firing data, generate a comprehensive load characteristic vector according to the same conversion rule for the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters, input the comprehensive load characteristic vector into the prediction model with the optimal prediction model parameters, calculate the ablation wear prediction values ​​at different shooting moments, perform Bayesian correction on the ablation wear prediction values ​​to obtain the corrected prediction results, construct an ablation wear degradation trend curve of the artillery barrel based on the corrected prediction results, extract the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and issue an early warning for the ablation wear state of the artillery barrel according to the wear change rate characteristics and the wear acceleration characteristics.

[0051] Optionally, this embodiment uses a real-time data acquisition system to dynamically monitor the artillery firing process. Temperature acquisition uses a high-response thermocouple array, with 16 measuring points evenly arranged along the axial direction of the barrel, and 4 layers of sensors arranged radially at each measuring point. The sampling frequency is set to 100Hz to ensure that the transient temperature change characteristics are captured. Signal conditioning uses cold-end compensation and digital filtering to improve measurement accuracy.

[0052] This embodiment realizes the real-time acquisition of pressure data. A piezoelectric sensor is used to monitor the pressure at the key position of the barrel, with a sampling frequency of 200kHz, to achieve accurate capture of the pressure waveform. The signal processing uses an adaptive threshold detection method to identify shooting events and ensure the accuracy of data segment division.

[0053] This embodiment designs a real-time monitoring scheme for charge parameters. A high-precision weighing system is used to measure charge quantity, while monitoring charge temperature and environmental parameters. Automatic compensation is achieved during data collection to eliminate the influence of environmental factors such as temperature and humidity.

[0054] This embodiment innovatively implements a feature extraction framework. The temperature feature is obtained by thermal field reconstruction: T(r,z) =ΣΣaij·φij(r,z), where φij is the basis function and aij is the expansion coefficient. The pressure feature extracts key parameters through waveform analysis, including peak value, rise time and waveform integral. The charge feature is combined with the packing density and uniformity index.

[0055] This embodiment uses a consistent conversion strategy to generate feature vectors. The feature conversion uses the principal component analysis matrix obtained in the training phase: F = W·X, where X is the original feature and W is the conversion matrix. This ensures that the features in the prediction phase remain consistent with those in the training phase.

[0056] This embodiment implements an innovative prediction calculation framework. The prediction model adopts a recursive structure: y(t) = f(F(t), y(t-1); θ), where F(t) is the current feature vector, y(t-1) is the predicted value at the previous moment, and θ is the optimized model parameter. This design takes into account the cumulative effect of the ablation process.

[0057] This embodiment designs a Bayesian correction mechanism. The prior distribution is established based on historical data: p(θ) ~ N(μ0, Σ0), where μ0 and Σ0 are the parameter mean and covariance, respectively. The posterior distribution is calculated by Bayesian updating: p(θ|D) ∝ p(D|θ)·p(θ), where D is the observed data.

[0058] This embodiment innovatively implements the correction of prediction results. The correction process adopts a sliding window strategy, and the window length is determined by autocorrelation analysis. Each prediction value is calculated by the posterior distribution expectation: ŷ = E[y|D,θ], and the uncertainty estimate of the prediction is obtained at the same time.

[0059] This embodiment uses the spline interpolation method to construct the degradation trend curve. The curve expression is: S(t) = Σci·Bi,k(t), where Bi,k is the k-order B-spline basis function and ci is the control point parameter. The interpolation nodes are adaptively arranged through curvature analysis to ensure that key features are maintained.

[0060] This embodiment implements a feature extraction strategy. The rate of change feature is calculated by numerical differentiation: v(t) = dS / dt, and the central difference format is used to improve accuracy. The acceleration feature is obtained by the second-order derivative: a(t) = d²S / dt², combined with smoothing to reduce the impact of noise.

[0061] This embodiment designs a multi-level warning mechanism. The warning index comprehensively considers the change rate and acceleration characteristics: W = w1·v(t) + w2·a(t), where wi is the weight coefficient. The warning threshold is determined by statistical analysis of historical data, taking into account the characteristics of different operating conditions.

[0062] This embodiment achieves reliable status assessment through innovative predictive analysis. The solution provides accurate characterization of the ablation process through multi-level analysis. The overall design fully considers the characteristics of artillery use and achieves full process optimization from data collection to status warning.

[0063] The prediction results of this embodiment show excellent accuracy and reliability, and the accurate prediction of the ablation process is achieved through multi-level analysis. This solution has shown strong prediction capabilities in practical applications and provides a reliable quantitative basis for equipment maintenance decisions.

[0064] The prediction framework of this embodiment has good adaptability and scalability, and can meet the prediction requirements under different working conditions. Through feature adaptation and Bayesian correction, accurate prediction of complex ablation processes is achieved, providing reliable decision support for equipment management.

[0065] From the above description, it can be seen that the gun barrel wear prediction method based on escape optimization provided in the embodiment of the present application can construct a sample set of corresponding relationships with the amount of ablation wear by converting temperature, pressure and charge characteristics into comprehensive load characteristic vectors. By designing the search space of the escape optimization algorithm, combined with the cross-validation method and the fitness sorting mechanism, the global optimization of the prediction model parameters is achieved. The system uses Bayesian correction to improve the prediction accuracy, extracts the wear change rate and acceleration characteristics based on the degradation trend curve, and establishes a reliable ablation wear status warning mechanism. This method effectively solves the shortcomings of traditional prediction technology in parameter optimization and status warning, and provides a scientific basis for the use and maintenance of gun barrels.

[0066] In one embodiment of the gun barrel wear prediction method based on escape optimization of the present application, see Figure 2 , and can also include the following: Step S201: during the firing process of multiple guns, barrel temperature data, bore pressure data and charging propellant data are collected, the barrel temperature data are normalized according to the overload factor to obtain a temperature characteristic parameter, the bore pressure data are normalized according to the peak pressure to obtain a pressure characteristic parameter, the charging propellant data are normalized according to the quality standard to obtain a charging characteristic parameter, and the temperature characteristic parameter, the pressure characteristic parameter and the charging characteristic parameter are reduced in dimension by using the principal component analysis method to obtain a characteristic vector reflecting the comprehensive load of the firing process; Step S202: perform time-series sampling on the feature vector and calculate the cumulative ablation wear amount corresponding to each shooting moment, establish a mapping relationship between the feature vector sequence and the ablation wear amount, use a stratified random sampling method to divide the mapping relationship into a training sample set and a verification sample set, normalize and standardize the data in the training sample set and the verification sample set, and construct a sample set of the corresponding relationship between the standardized feature vector and the ablation wear amount.

[0067] Optionally, the embodiment deploys a multi-dimensional sensing system for monitoring the ablation and wear of the gun barrel. Temperature monitoring uses a high-precision thermocouple array, with 16 measuring points arranged along the axial direction, and 4 sensors are set radially at each measuring point to form a 3D temperature field monitoring network. The sampling frequency is set to 100Hz to ensure accurate capture of transient temperature changes. Signal conditioning uses cold-end compensation and adaptive filtering to effectively suppress interference caused by shooting vibration.

[0068] This embodiment uses piezoelectric sensors in pressure data collection and is deployed at key positions in the gun barrel. The sensor range covers 0-600MPa and the sampling frequency is 200kHz. The signal conditioning circuit uses a charge amplifier configuration to ensure measurement stability under high temperature and high pressure environments. The propellant data collection uses a high-precision weighing system and monitors environmental parameters such as the propellant temperature.

[0069] This embodiment innovatively implements feature normalization processing. The temperature feature normalization uses an overload factor: F_t = (T-T_ref) / (T_max-T_ref), where T is the measured temperature, T_ref is the ambient temperature, and T_max is the material tolerance temperature. This processing method associates the temperature feature with the material performance, improving the physical significance of the feature.

[0070] This embodiment designs a normalization scheme for pressure characteristics. The pressure characteristics are normalized by peak value: F_p = P / P_max, where P is the measured pressure and P_max is the designed maximum pressure. At the same time, the pressure waveform characteristics, including rise time, duration and pressure gradient, are considered to form a multi-dimensional pressure characteristic vector.

[0071] This embodiment realizes the normalization of charge characteristics. The charge characteristics are standardized by mass: F_m = M / M_s, where M is the actual charge amount and M_s is the standard charge amount. The packing density and uniformity index are considered at the same time to build a complete charge characteristic system.

[0072] This embodiment uses an improved principal component analysis method for feature fusion. First, the feature matrix is ​​standardized to eliminate the dimension effect. The feature covariance matrix is ​​decomposed by SVD to obtain eigenvalues ​​and eigenvectors. The principal component selection is based on the cumulative contribution rate, and the principal component with a contribution rate of 95% is retained.

[0073] This embodiment innovatively designs a time-series sampling strategy. The sampling interval is determined based on the characteristic time scale of the ablation process to ensure that key state changes are captured. The cumulative ablation amount calculation takes into account the nonlinear characteristics of the material removal rate and is achieved through an improved integration method.

[0074] This embodiment implements an innovative stratified sampling scheme. First, based on the statistical distribution characteristics of the ablation amount, K-means clustering is used to divide the samples into multiple wear level layers. Random sampling is used within each layer, and the sampling ratio is proportional to the number of samples in the layer to ensure the representativeness of the samples.

[0075] This embodiment designs a data standardization processing framework. Standardization uses the z-score method: Z = (X-μ) / σ, where X is the original data, μ is the mean, and σ is the standard deviation. This processing eliminates the scale differences between different features and improves the stability of subsequent modeling.

[0076] This embodiment implements a sample set construction strategy. The corresponding relationship sample set is organized in a matrix form, with rows representing samples at different times and columns representing features and labels. The data structure design takes into account time series correlation, which facilitates subsequent model training.

[0077] This embodiment establishes a reliable sample construction framework through systematic data processing. While ensuring data quality, this solution provides comprehensive feature expression through multi-level analysis. The overall design fully considers the characteristics of artillery ablation and realizes the optimization of the entire process from data collection to sample construction.

[0078] The data processing results of this embodiment show excellent representativeness and balance, and achieve accurate characterization of the ablation process through multi-level analysis. This solution has shown strong data processing capabilities in practical applications and provides a reliable data basis for subsequent model training.

[0079] The processing framework of this embodiment has good adaptability and scalability, and can meet the data collection requirements under different working conditions. Through feature adaptation and sample balancing, accurate characterization of complex ablation processes is achieved, providing reliable data support for model training.

[0080] This embodiment innovatively solves key problems such as data imbalance and feature redundancy, and significantly improves sample quality. Through stratified sampling and feature dimensionality reduction, high-quality training samples are constructed, laying the foundation for improving the performance of the prediction model.

[0081] In one embodiment of the gun barrel wear prediction method based on escape optimization of the present application, see Figure 3 , and can also include the following: Step S301: dividing the sample data into multiple wear level layers based on the distribution characteristics of the ablation wear amount in the corresponding relationship sample set, calculating the proportion of the sample data in each wear level layer, randomly shuffling the sample data in the wear level layer, extracting a corresponding number of sample data from each wear level layer according to a preset ratio of the training set to the validation set, and constructing a training sample subset and a validation sample subset from the extracted sample data respectively; Step S302: Perform a data balance check on the training sample subset and the verification sample subset, calculate the distribution difference of each wear level layer sample in the training set and the verification set, and re-perform stratified sampling when the distribution difference exceeds a preset threshold until the data distribution of the training sample subset and the verification sample subset meets the balance requirement.

[0082] Optionally, this embodiment first analyzes the distribution characteristics of the ablation wear amount, and uses a data clustering method to divide the samples into multiple wear level layers. Specifically, an improved K-means clustering algorithm is used to determine the optimal number of level layers through the silhouette coefficient and inter-cluster distance evaluation. In the actual application of artillery, the wear state is usually divided into three main level layers: slight wear, moderate wear, and severe wear. Each main level layer is further subdivided into several sub-layers according to the slight differences in the degree of wear.

[0083] This embodiment performs statistical analysis on the sample data of each wear level layer and calculates the proportion of each layer of samples in the total. This analysis fully considers the frequency characteristics of different wear states during the use of artillery and reflects the distribution law of wear states in actual application scenarios. For artillery with high frequency of use, slightly worn samples usually account for a large proportion, while severely worn samples are relatively small.

[0084] This embodiment uses an improved Fisher-Yates algorithm in the sample shuffling process to ensure randomness while maintaining the temporal correlation characteristics of the data. The shuffling process takes into account the correlation between adjacent shots and avoids dispersing strongly correlated samples into different data sets. This processing method ensures that the model can learn the continuity characteristics of wear evolution.

[0085] This embodiment adopts a stratified proportional sampling strategy in the sample extraction process. First, the overall ratio of the training set to the validation set is determined, usually set to 7:3 or 8:2. Then, sampling is performed in each wear level layer according to the same ratio to ensure that the ratio of samples in each level layer in different data sets is consistent. This sampling method avoids over-representation or under-representation of certain wear states in a specific data set.

[0086] This embodiment innovatively implements a sample balancing mechanism. For wear level layers with a small number of samples, a data enhancement method based on a physical model is used to generate synthetic samples. The enhancement method takes into account the physical constraints of the shooting conditions to ensure that the generated samples conform to the actual wear evolution law. This processing effectively improves the imbalance of sample distribution.

[0087] This embodiment designs a complete data balance test scheme. By calculating the JS divergence of the distribution of samples of each wear level layer in the training set and the validation set, the distribution consistency of the two data sets is evaluated. When the distribution difference exceeds the preset threshold, the resampling mechanism is triggered. This dynamic adjustment ensures the distribution similarity between different data sets.

[0088] This embodiment adopts an adaptive adjustment strategy in the resampling process. The sampling weights of each level are dynamically adjusted according to the distribution difference characteristics of the previous sampling. The sampling weights of the level with large distribution differences are increased to improve its selection probability in the resampling process. This method speeds up the convergence speed of distribution balance.

[0089] This embodiment implements an iteratively optimized sampling framework. After each round of resampling, the representativeness and diversity of the samples are evaluated simultaneously. The representativeness is evaluated by the intra-layer sample variance, and the diversity is evaluated by the inter-layer sample distance. This multi-dimensional evaluation ensures the quality of the sampling results.

[0090] The stratified sampling scheme of this embodiment shows excellent performance in practical applications. By considering the distribution characteristics and evolution laws of wear states, high-quality training data can be constructed. This scheme effectively solves the problems of sample imbalance and insufficient representativeness in traditional random sampling.

[0091] The data balance check mechanism of this embodiment ensures the reliability of model training. Through strict distribution consistency control, the negative impact of the distribution difference between the training set and the validation set on the model performance is avoided. This mechanism provides an important guarantee for the generalization ability of the model.

[0092] This embodiment significantly improves the sample quality through innovative sampling strategies and balance control. While ensuring data representativeness, this solution achieves a balanced distribution of samples of different wear states, laying a solid foundation for subsequent model training. The balanced distribution of samples enables the model to fully learn the characteristics of different wear states, improving the accuracy and reliability of prediction.

[0093] In one embodiment of the gun barrel wear prediction method based on escape optimization of the present application, see Figure 4 , and can also include the following: Step S401: determining the boundary range of the search space according to the characteristic dimension of the training sample subset, randomly generating an initial particle swarm in the search space, mapping each dimensional component of the initial particle swarm to a weight parameter and a bias parameter of a prediction model, constructing a parameter matrix of the prediction model based on the weight parameter and the bias parameter, and dividing the training sample subset into multiple training sets and test sets using a K-fold cross-validation method; Step S402: input the training set into the prediction model for training, use the test set to calculate the prediction error of the model, repeat the cross-validation process for each particle to obtain the average prediction error, use the average prediction error as the fitness value of the particle, sort the particle group in ascending order based on the fitness value, and select the particle position with the smallest fitness value as the global optimal solution of the current iteration.

[0094] Optionally, this embodiment determines the search space boundary based on the feature distribution of the training samples. For each feature dimension, its variation range is determined through statistical analysis, and an appropriate margin is left when setting the search boundary to prevent valid solutions from being truncated by the boundary. This boundary setting method fully considers the physical characteristics of the artillery ablation process and ensures that the search space covers all possible valid solutions.

[0095] This embodiment uses an improved Latin hypercube sampling method to generate the initial particle swarm. This method divides the search space into multiple subspaces to ensure that the initial particles are evenly distributed throughout the search space. The number of particles is dynamically determined based on the problem dimension, which not only ensures the coverage of the search space, but also avoids excessive computational burden.

[0096] This embodiment innovatively designs a parameter mapping mechanism. The mapping of particle positions to model parameters uses nonlinear transformation, and the weight parameters are mapped using a sigmoid function to ensure that the weight parameters are distributed within a reasonable range. The bias parameters are mapped using a tanh function. This design takes into account the role characteristics of different parameters in the prediction model.

[0097] This embodiment realizes the dynamic construction of the parameter matrix. According to the network structure of the prediction model, the mapped parameters are organized into a matrix form according to the hierarchical relationship. The matrix structure design takes into account the correlation between parameters, which is convenient for subsequent model calculation and gradient propagation.

[0098] This embodiment adopts an improved K-fold cross validation method. First, the training sample subsets are divided into layers to ensure that the distribution of samples with different wear states in each compromise is similar. During the validation process, a rotation method is adopted so that each sample has the opportunity to be used as test data, thereby improving the reliability of the validation results.

[0099] This embodiment innovatively implements the model training strategy. In each training compromise, a batch training method is adopted, and the batch size is dynamically adjusted according to the number of samples. The learning rate adaptive adjustment mechanism is adopted during the training process to ensure the stability of model convergence.

[0100] This embodiment designs a complete prediction error evaluation framework. The error calculation takes into account the importance weights of different wear states, and gives a higher weight to the prediction error of severe wear states. This weighting method is consistent with the degree of attention paid to different wear states in actual applications.

[0101] This embodiment implements an innovative error aggregation method. In the aggregation of multi-fold verification results, the stability of the prediction error of each fold is taken into account. The verification results with large fluctuations are screened through variance analysis to improve the representativeness of the average prediction error.

[0102] This embodiment adopts an improved fitness evaluation scheme. The fitness calculation not only considers the prediction error, but also introduces the model complexity penalty term to avoid overfitting. This design improves the generalization ability of the model while ensuring the prediction accuracy.

[0103] This embodiment realizes an efficient particle sorting mechanism. The sorting process uses an improved quick sorting algorithm to carefully compare particles with similar fitness. The selection of the global optimal solution takes into account the stability of the solution to avoid interference by local noise.

[0104] This embodiment achieves efficient model parameter search through innovative parameter optimization strategies. This solution provides reliable parameter evaluation through multi-level verification while ensuring search space coverage. The overall design fully considers the characteristics of the prediction model and achieves efficient exploration of the parameter space.

[0105] The optimization results of this embodiment show excellent performance and stability. Through multi-level verification and evaluation, it is ensured that the selected parameters have good generalization ability. This solution shows strong parameter optimization ability in practical applications and provides reliable parameter support for the prediction model.

[0106] The optimization framework of this embodiment has good versatility and robustness, and can adapt to the parameter optimization requirements of different feature spaces. Through innovative parameter mapping and evaluation mechanisms, efficient search of complex parameter spaces is achieved, providing reliable performance guarantee for the prediction model.

[0107] In one embodiment of the gun barrel wear prediction method based on escape optimization of the present application, see Figure 5 , and can also include the following: Step S501: Calculate the Euclidean distance between each particle and the global optimal solution. When the Euclidean distance is less than a preset escape threshold, trigger the escape mechanism, add an escape disturbance term in the direction of the current velocity of the particle, update the velocity vector of the particle according to the escape disturbance term and the inertia weight, calculate the new position of the particle based on the velocity vector, map the position components that exceed the boundary of the search space, and convert the mapped position information into new prediction model parameters; Step S502: recalculate the fitness value of the particle swarm after each round of iteration and update the global optimal solution, record the historical change trajectory of the global optimal solution, stop searching when the change of the global optimal solution in multiple consecutive rounds of iterations is less than the convergence threshold or the number of iterations reaches the preset upper limit, and use the global optimal solution at the final convergence as the optimized prediction model parameter.

[0108] Optionally, this embodiment implements particle escape judgment based on Euclidean distance calculation. For each particle, the distance between it and the current global optimal solution in the feature space is calculated. This distance calculation takes into account the weights of the features of each dimension and reflects the actual distribution state of the particles in the search space. The setting of the distance threshold is dynamically adjusted based on the scale of the search space to ensure the reasonable triggering of the escape mechanism.

[0109] This embodiment innovatively designs an escape disturbance mechanism. When a particle falls into a local optimal region, it escapes by superimposing disturbances in the velocity direction. The disturbance amplitude is related to the current fitness of the particle. The better the fitness, the smaller the disturbance. This design ensures a detailed search for potential high-quality solutions.

[0110] This embodiment implements an adaptive speed update strategy. The inertia weight is dynamically adjusted during the iteration process, maintaining a larger value in the early stage to promote global search, and reducing it in the later stage to strengthen local refined search. The speed update takes into account the influence of both individual experience and group experience, and realizes intelligent adjustment of the search direction.

[0111] This embodiment adopts an innovative boundary processing method. When the particle position exceeds the search space, it is not simply truncated or rebounded, but a periodic mapping method is adopted. This processing not only maintains the continuity of the search, but also avoids the severe interference of boundary processing on the search trajectory.

[0112] This embodiment designs a smoothing mechanism for parameter conversion. The conversion of position information to model parameters uses a continuously differentiable mapping function to ensure the smoothness of the parameter space. This design is conducive to the transmission of gradient information and improves the efficiency of parameter optimization.

[0113] This embodiment implements an innovative fitness evaluation framework. The fitness calculation after each iteration not only considers the current prediction error, but also introduces a smoothing factor of historical performance to avoid drastic fluctuations in the evaluation results. This stable evaluation mechanism helps the search process converge.

[0114] This embodiment adopts a multi-level optimal solution update strategy. The update of the global optimal solution takes into account the stability of the solution and avoids being misled by temporary high-quality solutions through continuous multiple rounds of verification. This strategy ensures the stability and reliability of the optimization process.

[0115] This embodiment innovatively designs a convergence criterion. By monitoring the change trend of the global optimal solution, the convergence state is judged by combining the dual indicators of change amount and change rate. This design not only ensures the quality of the optimization result, but also avoids the computational overhead caused by excessive iteration.

[0116] This embodiment realizes dynamic monitoring of the optimization process. By recording the historical trajectory of the global optimal solution, the convergence characteristics of the optimization process are analyzed. This information can be used to dynamically adjust the optimization strategy and improve the search efficiency.

[0117] This embodiment adopts an improved termination judgment mechanism. In addition to the basic iteration number limit, a stability judgment of the solution is also introduced. When the optimal solution changes very little in multiple consecutive rounds, it is considered that a stable solution has been found, and the search is terminated in time to avoid invalid calculations.

[0118] This embodiment achieves an efficient parameter search process through innovative optimization strategies. The solution effectively avoids local optimality through the escape mechanism and improves the global search capability. The overall design fully considers the characteristics of the prediction model and achieves efficient exploration of the parameter space.

[0119] The optimization results of this embodiment show excellent convergence and stability. Accurate parameter selection of the prediction model is achieved through multi-level optimization. This solution shows strong optimization capabilities in practical applications and provides reliable guarantee for model performance.

[0120] The optimization framework of this embodiment has good versatility and robustness, and can adapt to the parameter optimization requirements of different feature spaces. Through the escape mechanism and adaptive strategy, efficient search of complex parameter space is achieved, providing reliable parameter support for the prediction model. This solution significantly improves the performance of the model in the prediction of artillery ablation and wear.

[0121] In one embodiment of the gun barrel wear prediction method based on escape optimization of the present application, see Figure 6 , and can also include the following: Step S601: Substitute the temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters collected in real time into the principal component analysis model, perform dimensionality reduction transformation on the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters according to the characteristic conversion matrix obtained in the training stage, generate a characteristic vector reflecting the comprehensive load of the current shooting process, input the characteristic vector into the prediction model, and calculate the ablation wear prediction value at the current moment based on the optimal prediction model parameters; Step S602: Construct a Bayesian prior distribution model, use historical prediction data as prior information, calculate the posterior probability distribution of the ablation wear prediction value, correct the prediction result based on the mean and variance of the posterior probability distribution, and use the sliding time window method to calculate the corrected prediction results at different shooting times in turn.

[0122] Optionally, this embodiment adopts an efficient feature extraction scheme in real-time data processing. The temperature characteristic parameters are collected by a high-precision thermocouple array to capture the temperature distribution of each measuring point of the barrel in real time. Real-time filtering and compensation are performed during the data collection process to ensure the accuracy of the characteristic parameters. For transient temperature changes during the artillery firing process, an adaptive sampling strategy is adopted to increase the sampling frequency at critical moments.

[0123] This embodiment uses an intelligent trigger mechanism in the pressure feature extraction. The piezoelectric sensor array is used to monitor the pressure fluctuation of the gun chamber in real time, and a high-speed data acquisition system is used to accurately capture the pressure waveform characteristics. During the data processing process, the physical model is combined to reconstruct the signal to improve the reliability of feature extraction.

[0124] This embodiment realizes real-time monitoring of charge characteristics. Through high-precision weighing system and environmental parameter monitoring, the charge quantity and filling status information are obtained. The feature extraction process takes into account the influence of environmental factors such as temperature and humidity, and improves the accuracy of feature parameters through compensation algorithm.

[0125] This embodiment innovatively implements a feature dimension reduction strategy. The feature conversion matrix obtained in the training phase is used to perform a consistent conversion on real-time features. The conversion process maintains the correlation between features, while significantly reducing the data dimension and improving the efficiency of subsequent predictions.

[0126] This embodiment designs a real-time computing framework for the prediction model. The model input layer adopts a dynamic cache mechanism to ensure the continuous input of feature vectors. The parallel processing strategy is adopted in the calculation process to improve the prediction speed. The prediction results are evaluated by confidence to ensure the reliability of the output.

[0127] This embodiment innovatively constructs a Bayesian prior distribution model. The prior distribution is established based on the statistical characteristics of historical data, and fully considers the laws of wear evolution under different working conditions. The model parameters are determined by maximum likelihood estimation to ensure the accuracy of the prior information.

[0128] This embodiment realizes the dynamic update of the posterior probability distribution. Combining the current prediction value and historical data, Bayesian inference is used to calculate the posterior distribution. The update of the distribution parameters takes into account the timeliness of the data, and recent data has a higher weight.

[0129] This embodiment designs a correction mechanism for prediction results. The correction process not only considers the mean of the posterior distribution, but also uses variance information to evaluate the uncertainty of the prediction. The correction parameters are adjusted through an adaptive algorithm to ensure the stability of the correction effect.

[0130] This embodiment adopts an innovative sliding window strategy. The window size is determined based on the time scale of the wear evolution characteristics, and the window sliding step is dynamically adjusted according to the shooting frequency. This processing method not only ensures the continuity of the prediction, but also improves the calculation efficiency.

[0131] This embodiment realizes the time series processing of the prediction results. The key features of wear evolution are identified through the trend analysis of the data in the window. The time series association of the prediction results is described by the state transition model, which improves the accuracy of the prediction.

[0132] This embodiment achieves real-time assessment of gun ablation and wear through an innovative prediction framework. The solution combines physical models and data-driven methods to provide reliable prediction capabilities. The overall design fully considers actual application needs and achieves full process optimization from data collection to prediction output.

[0133] The prediction results of this embodiment show excellent real-time performance and accuracy. Through multi-level analysis and correction, accurate prediction of the ablation and wear process is achieved. This solution has shown strong prediction capabilities in practical applications and provides a reliable basis for equipment maintenance decisions.

[0134] The prediction framework of this embodiment has good adaptability and scalability, and can meet the prediction requirements under different working conditions. Through feature adaptation and Bayesian correction, accurate prediction of complex ablation processes is achieved, providing reliable decision support for equipment management.

[0135] In one embodiment of the gun barrel wear prediction method based on escape optimization of the present application, see Figure 7 , and can also include the following: Step S701: performing time series fitting processing on the corrected prediction result sequence, using a polynomial regression method to construct a degradation trend curve reflecting the ablation wear evolution process, calculating the first-order derivative of the degradation trend curve at different times to obtain the wear change rate characteristics, calculating the second-order derivative of the degradation trend curve at different times to obtain the wear acceleration characteristics, and establishing a quantitative evaluation index of the ablation wear state of the gun barrel based on the wear change rate characteristics and the wear acceleration characteristics; Step S702: Determine the wear change rate threshold and the wear acceleration threshold based on statistical analysis of historical data. When the wear change rate characteristic exceeds the wear change rate threshold or the wear acceleration characteristic exceeds the wear acceleration threshold, trigger the warning mechanism, and determine the warning level of the ablation wear state of the gun barrel based on the preset warning level classification standard.

[0136] Optionally, this embodiment uses a polynomial regression method to fit the corrected prediction sequence. During the fitting process, the polynomial order is first determined based on the characteristics of the shooting data, and the optimal fitting parameters are selected through cross-validation. For the wear characteristics of the artillery at different use stages, a segmented fitting strategy is adopted to ensure that the degradation trend curve can accurately reflect the wear evolution process.

[0137] This embodiment innovatively realizes the extraction of wear change rate features. By taking the first-order derivative of the degradation trend curve, the wear change rate at different times is obtained. The calculation process uses the numerical differentiation method and introduces a smoothing mechanism to reduce the impact of noise. The change rate feature directly reflects the evolution speed of the wear process.

[0138] This embodiment designs a wear acceleration feature extraction scheme. The second-order derivative of the degradation trend curve is calculated to obtain the acceleration characteristics of wear changes. An adaptive step size strategy is used in the calculation process to ensure the stability of feature extraction. The acceleration feature reflects the development trend of the wear process.

[0139] This embodiment implements an innovative state evaluation index system. The evaluation index comprehensively considers the change rate and acceleration characteristics to construct a multi-dimensional state representation. The index weight is determined by combining expert experience and data analysis to ensure the scientific nature of the evaluation results.

[0140] This embodiment adopts a data-driven threshold determination method. By statistically analyzing historical data, a distribution model of wear characteristics is established. The threshold setting takes into account the distribution of wear characteristics under different working conditions, and the quantile method is used to determine a reasonable threshold range.

[0141] This embodiment innovatively designs a dynamic threshold adjustment mechanism. The feature distribution model is continuously updated based on the newly added data to achieve adaptive adjustment of the threshold. This dynamic mechanism ensures that the warning standard matches the actual usage status.

[0142] This embodiment implements a multi-level warning mechanism. The warning level division takes into account the two dimensions of wear degree and development trend, and constructs a two-dimensional warning matrix. Different warning levels correspond to different management measures, providing a decision-making basis for equipment maintenance.

[0143] This embodiment designs a process for determining the warning level. The determination process adopts a fuzzy decision method and comprehensively considers multiple characteristic indicators. By establishing a fuzzy rule base, intelligent determination of the warning level is achieved.

[0144] This embodiment innovatively implements dynamic updating of warning information. As new monitoring data is continuously input, the warning status is updated in real time. The update process takes into account the hysteresis characteristics of state transitions to avoid frequent fluctuations in the warning level.

[0145] This embodiment adopts a hierarchical early warning processing strategy. Different early warning levels trigger different response mechanisms, which gradually increase from prompt early warning to mandatory early warning. Early warning information includes wear status assessment and development trend prediction, providing comprehensive support for maintenance decisions.

[0146] This embodiment achieves accurate assessment of the ablation and wear state of artillery through an innovative assessment and early warning framework. The solution combines physical characteristics and statistical analysis to provide reliable early warning capabilities. The overall design fully considers the actual application needs and achieves full process optimization from state assessment to early warning output.

[0147] The evaluation and early warning results of this embodiment show excellent accuracy and reliability. Through multi-dimensional analysis and dynamic adjustment, accurate evaluation of the ablation and wear state is achieved. This solution has shown strong early warning capabilities in practical applications and provides a scientific basis for equipment maintenance.

[0148] The evaluation and early warning framework of this embodiment has good adaptability and scalability, and can meet the evaluation requirements under different usage conditions. Through feature adaptation and dynamic adjustment of thresholds, accurate evaluation of complex ablation processes is achieved, providing reliable decision support for equipment management. This solution significantly improves the scientificity and effectiveness of artillery maintenance and support.

[0149] In order to establish a reliable ablation wear state warning mechanism, effectively solve the deficiencies of traditional prediction technology in parameter optimization and state warning, and provide a scientific basis for the use and maintenance of gun barrels, the present application provides an embodiment of a gun barrel wear prediction device based on escape optimization for realizing all or part of the contents of the gun barrel wear prediction method based on escape optimization, see Figure 8 The gun barrel wear prediction device based on escape optimization specifically includes the following contents: The data set construction module 10 is used to collect historical ablation and wear data of multiple guns at different firing stages, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, convert the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, construct a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation and wear amount, and use a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; The optimal parameter determination module 20 is used to establish a search space for the escape optimization algorithm based on the training sample subset, set the initial position distribution of the search particle group, map the position information of the search particle group to the prediction model parameters, use the cross-validation method to calculate the prediction error corresponding to each particle, use the prediction error as the particle fitness function, sort the fitness of the particle group according to the fitness function, select the particle with the best fitness as the global optimal solution, update the speed and position of the particle based on the escape mechanism in each round of iteration, stop the search when the number of iterations reaches a preset value, and use the finally converged global optimal solution as the optimal prediction model parameter; The wear state prediction module 30 is used to read the real-time shooting data of the artillery to be predicted, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the real-time shooting data, generate a comprehensive load characteristic vector according to the same conversion rule for the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters, input the comprehensive load characteristic vector into the prediction model with the optimal prediction model parameters, calculate the ablation wear prediction values ​​at different shooting moments, perform Bayesian correction on the ablation wear prediction values ​​to obtain the corrected prediction results, construct an ablation wear degradation trend curve of the artillery barrel based on the corrected prediction results, extract the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and issue an early warning for the ablation wear state of the artillery barrel according to the wear change rate characteristics and the wear acceleration characteristics.

[0150] From the above description, it can be seen that the gun barrel wear prediction device based on escape optimization provided in the embodiment of the present application can construct a sample set of corresponding relationships with the amount of ablation wear by converting temperature, pressure and charge characteristics into comprehensive load characteristic vectors. By designing the search space of the escape optimization algorithm, combined with the cross-validation method and fitness sorting mechanism, global optimization of the prediction model parameters is achieved. The system uses Bayesian correction to improve prediction accuracy, extracts wear change rate and acceleration characteristics based on the degradation trend curve, and establishes a reliable ablation wear status warning mechanism. This method effectively solves the shortcomings of traditional prediction technology in parameter optimization and status warning, and provides a scientific basis for the use and maintenance of gun barrels.

[0151] From the hardware level, in order to establish a reliable ablation wear state warning mechanism, effectively solve the deficiencies of traditional prediction technology in parameter optimization and state warning, and provide a scientific basis for the use and maintenance of gun barrels, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the gun barrel wear prediction method based on escape optimization, 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 gun barrel wear prediction device based on escape optimization and related equipment such as core business system, user terminal and related database; the logic controller can be a desktop computer, tablet computer and 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 gun barrel wear prediction method based on escape optimization and the embodiment of the gun barrel wear prediction device based on escape optimization in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0152] 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.

[0153] In practical applications, part of the gun barrel wear prediction method based on escape optimization 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.

[0154] 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.

[0155] 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.

[0156] In one embodiment, the gun barrel wear prediction method function based on escape optimization can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control: Step S101: collecting historical ablation and wear data of multiple guns at different firing stages, extracting temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, converting the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, constructing a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and using a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; Step S102: establishing a search space of the escape optimization algorithm based on the training sample subset, setting the initial position distribution of the search particle swarm, mapping the position information of the search particle swarm to the prediction model parameters, using the cross-validation method to calculate the prediction error corresponding to each particle, using the prediction error as the particle fitness function, sorting the particle swarm according to the fitness function, selecting the particle with the best fitness as the global optimal solution, updating the speed and position of the particle based on the escape mechanism in each round of iteration, stopping the search when the number of iterations reaches a preset value, and taking the finally converged global optimal solution as the optimal prediction model parameters; Step S103: read the real-time firing data of the artillery to be predicted, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the real-time firing data, generate a comprehensive load characteristic vector according to the same conversion rule for the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters, input the comprehensive load characteristic vector into the prediction model with the optimal prediction model parameters, calculate the ablation wear prediction values ​​at different shooting moments, perform Bayesian correction on the ablation wear prediction values ​​to obtain the corrected prediction results, construct an ablation wear degradation trend curve of the artillery barrel based on the corrected prediction results, extract the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and issue an early warning for the ablation wear state of the artillery barrel according to the wear change rate characteristics and the wear acceleration characteristics.

[0157] From the above description, it can be seen that the electronic device provided in the embodiment of the present application constructs a sample set of corresponding relationships with the amount of ablation wear by converting temperature, pressure and charge characteristics into comprehensive load characteristic vectors. By designing the search space of the escape optimization algorithm, combined with the cross-validation method and the fitness ranking mechanism, global optimization of the prediction model parameters is achieved. The system uses Bayesian correction to improve prediction accuracy, extracts wear change rate and acceleration characteristics based on the degradation trend curve, and establishes a reliable ablation wear status warning mechanism. This method effectively solves the shortcomings of traditional prediction technology in parameter optimization and status warning, and provides a scientific basis for the use and maintenance of artillery barrels.

[0158] In another embodiment, the gun barrel wear prediction device based on escape optimization can be configured separately from the central processing unit 9100. For example, the gun barrel wear prediction device based on escape optimization can be configured as a chip connected to the central processing unit 9100, and the function of the gun barrel wear prediction method based on escape optimization can be realized through the control of the central processing unit.

[0159] like Fig. 9As 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.

[0160] 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.

[0161] 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.

[0162] 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.

[0163] 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.

[0164] 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.).

[0165] 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.

[0166] 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.

[0167] The embodiment of the present application also provides a computer-readable storage medium capable of implementing all the steps of the gun barrel wear prediction method based on escape optimization 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 gun barrel wear prediction method based on escape optimization in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented: Step S101: collecting historical ablation and wear data of multiple guns at different firing stages, extracting temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, converting the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, constructing a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and using a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; Step S102: establishing a search space of the escape optimization algorithm based on the training sample subset, setting the initial position distribution of the search particle swarm, mapping the position information of the search particle swarm to the prediction model parameters, using the cross-validation method to calculate the prediction error corresponding to each particle, using the prediction error as the particle fitness function, sorting the particle swarm according to the fitness function, selecting the particle with the best fitness as the global optimal solution, updating the speed and position of the particle based on the escape mechanism in each round of iteration, stopping the search when the number of iterations reaches a preset value, and taking the finally converged global optimal solution as the optimal prediction model parameters; Step S103: read the real-time firing data of the artillery to be predicted, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the real-time firing data, generate a comprehensive load characteristic vector according to the same conversion rule for the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters, input the comprehensive load characteristic vector into the prediction model with the optimal prediction model parameters, calculate the ablation wear prediction values ​​at different shooting moments, perform Bayesian correction on the ablation wear prediction values ​​to obtain the corrected prediction results, construct an ablation wear degradation trend curve of the artillery barrel based on the corrected prediction results, extract the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and issue an early warning for the ablation wear state of the artillery barrel according to the wear change rate characteristics and the wear acceleration characteristics.

[0168] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application constructs a sample set of corresponding relationships with the amount of ablation wear by converting temperature, pressure and charge characteristics into comprehensive load characteristic vectors. By designing the search space of the escape optimization algorithm, combined with the cross-validation method and the fitness sorting mechanism, global optimization of the prediction model parameters is achieved. The system uses Bayesian correction to improve prediction accuracy, extracts wear change rate and acceleration characteristics based on the degradation trend curve, and establishes a reliable ablation wear status warning mechanism. This method effectively solves the shortcomings of traditional prediction technology in parameter optimization and status warning, and provides a scientific basis for the use and maintenance of artillery barrels.

[0169] The embodiments of the present application also provide a computer program product capable of implementing all the steps of the gun barrel wear prediction method based on escape optimization in the above embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the gun barrel wear prediction method based on escape optimization are implemented. For example, the computer program / instruction implements the following steps: Step S101: collecting historical ablation and wear data of multiple guns at different firing stages, extracting temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, converting the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, constructing a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and using a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; Step S102: establishing a search space of the escape optimization algorithm based on the training sample subset, setting the initial position distribution of the search particle swarm, mapping the position information of the search particle swarm to the prediction model parameters, using the cross-validation method to calculate the prediction error corresponding to each particle, using the prediction error as the particle fitness function, sorting the particle swarm according to the fitness function, selecting the particle with the best fitness as the global optimal solution, updating the speed and position of the particle based on the escape mechanism in each round of iteration, stopping the search when the number of iterations reaches a preset value, and taking the finally converged global optimal solution as the optimal prediction model parameters; Step S103: read the real-time firing data of the artillery to be predicted, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the real-time firing data, generate a comprehensive load characteristic vector according to the same conversion rule for the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters, input the comprehensive load characteristic vector into the prediction model with the optimal prediction model parameters, calculate the ablation wear prediction values ​​at different shooting moments, perform Bayesian correction on the ablation wear prediction values ​​to obtain the corrected prediction results, construct an ablation wear degradation trend curve of the artillery barrel based on the corrected prediction results, extract the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and issue an early warning for the ablation wear state of the artillery barrel according to the wear change rate characteristics and the wear acceleration characteristics.

[0170] From the above description, it can be seen that the computer program product provided in the embodiment of the present application constructs a sample set of corresponding relationships with the amount of ablation wear by converting temperature, pressure and charge characteristics into comprehensive load characteristic vectors. By designing the search space of the escape optimization algorithm, combined with the cross-validation method and the fitness ranking mechanism, global optimization of the prediction model parameters is achieved. The system uses Bayesian correction to improve prediction accuracy, extracts wear change rate and acceleration characteristics based on the degradation trend curve, and establishes a reliable ablation wear status warning mechanism. This method effectively solves the shortcomings of traditional prediction technology in parameter optimization and status warning, and provides a scientific basis for the use and maintenance of artillery barrels.

[0171] 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.

[0172] 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.

[0173] 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.

[0174] 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.

[0175] 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 gun barrel wear prediction method based on escape optimization, characterized in that: The method comprises: Collect historical ablation and wear data of multiple guns at different firing stages, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, convert the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, construct a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and use a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; A search space of an escape optimization algorithm is established based on the training sample subset, an initial position distribution of a search particle swarm is set, position information of the search particle swarm is mapped to prediction model parameters, a prediction error corresponding to each particle is calculated using a cross-validation method, the prediction error is used as a particle fitness function, the particle swarm is sorted according to the fitness function, the particle with the best fitness is selected as the global optimal solution, the speed and position of the particle are updated based on the escape mechanism in each round of iteration, the search is stopped when the number of iterations reaches a preset value, and the global optimal solution that finally converges is used as the optimal prediction model parameter; The real-time firing data of the artillery to be predicted is read, and temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters are extracted from the real-time firing data. The temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters are converted into a comprehensive load characteristic vector according to the same conversion rule. The comprehensive load characteristic vector is input into a prediction model with the optimal prediction model parameters, and the ablation wear prediction values ​​at different shooting moments are calculated. The ablation wear prediction values ​​are subjected to Bayesian correction to obtain the corrected prediction results. Based on the corrected prediction results, an ablation wear degradation trend curve of the artillery barrel is constructed, and a wear change rate characteristic and a wear acceleration characteristic are extracted from the degradation trend curve. According to the wear change rate characteristic and the wear acceleration characteristic, an early warning is given to the ablation wear state of the artillery barrel.

2. The gun barrel wear prediction method based on escape optimization according to claim 1 is characterized in that: The method collects historical ablation and wear data of multiple guns at different shooting stages, extracts temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, converts the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, and constructs a corresponding relationship sample set between the comprehensive load characteristic vectors and the ablation wear amount, including: During the firing process of multiple guns, barrel temperature data, bore pressure data and charging propellant data are collected, the barrel temperature data are normalized according to the overload factor to obtain a temperature characteristic parameter, the bore pressure data are normalized according to the peak pressure to obtain a pressure characteristic parameter, the charging propellant data are normalized according to the quality standard to obtain a charging characteristic parameter, and the temperature characteristic parameter, the pressure characteristic parameter and the charging characteristic parameter are reduced in dimension by using a principal component analysis method to obtain a characteristic vector reflecting the comprehensive load of the firing process; The feature vector is sampled in time series and the accumulated ablation wear amount corresponding to each shooting moment is calculated, a mapping relationship between the feature vector sequence and the ablation wear amount is established, a stratified random sampling method is used to divide the mapping relationship into a training sample set and a verification sample set, the data in the training sample set and the verification sample set are normalized and standardized, and a corresponding relationship sample set between the standardized feature vector and the ablation wear amount is constructed.

3. The gun barrel wear prediction method based on escape optimization according to claim 1 is characterized in that: The step of using a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset includes: Based on the distribution characteristics of the ablation wear amount in the corresponding relationship sample set, the sample data is divided into multiple wear level layers, the proportion of sample data in each wear level layer is calculated, the sample data in the wear level layer is randomly shuffled, and a corresponding number of sample data are extracted from each wear level layer according to a preset ratio of training set to verification set, and the extracted sample data are respectively used to construct a training sample subset and a verification sample subset; A data balance check is performed on the training sample subset and the verification sample subset, and the distribution difference of each wear level layer sample in the training set and the verification set is calculated. When the distribution difference exceeds a preset threshold, stratified sampling is performed again until the data distribution of the training sample subset and the verification sample subset meets the balance requirement.

4. The gun barrel wear prediction method based on escape optimization according to claim 1 is characterized in that: The method includes establishing a search space of the escape optimization algorithm based on the training sample subset, setting an initial position distribution of a search particle swarm, mapping the position information of the search particle swarm to a prediction model parameter, calculating a prediction error corresponding to each particle by a cross-validation method, using the prediction error as a particle fitness function, ranking the particle swarm according to the fitness function, and selecting the particle with the best fitness as the global optimal solution, including: Determining the boundary range of the search space according to the characteristic dimension of the training sample subset, randomly generating an initial particle swarm in the search space, mapping each dimensional component of the initial particle swarm to a weight parameter and a bias parameter of a prediction model, constructing a parameter matrix of the prediction model based on the weight parameter and the bias parameter, and dividing the training sample subset into a plurality of training sets and a test set using a K-fold cross validation method; The training set is input into the prediction model for training, the prediction error of the model is calculated using the test set, the cross-validation process is repeated for each particle to obtain the average prediction error, the average prediction error is used as the fitness value of the particle, the particle group is sorted in ascending order based on the fitness value, and the particle position with the smallest fitness value is selected as the global optimal solution of the current iteration.

5. The gun barrel wear prediction method based on escape optimization according to claim 1 is characterized in that: The particle speed and position are updated based on the escape mechanism in each round of iteration, the search is stopped when the number of iterations reaches a preset value, and the final converged global optimal solution is used as the optimal prediction model parameter, including: Calculate the Euclidean distance between each particle and the global optimal solution, trigger the escape mechanism when the Euclidean distance is less than a preset escape threshold, add an escape disturbance term in the direction of the particle's current velocity, update the particle's velocity vector according to the escape disturbance term and the inertia weight, calculate the particle's new position based on the velocity vector, map the position components that exceed the search space boundary, and convert the mapped position information into new prediction model parameters; The fitness value of the particle swarm after each round of iteration is recalculated and the global optimal solution is updated. The historical change trajectory of the global optimal solution is recorded. When the change of the global optimal solution in multiple consecutive rounds of iterations is less than the convergence threshold or the number of iterations reaches the preset upper limit, the search is stopped and the global optimal solution at the final convergence is used as the optimized prediction model parameter.

6. The gun barrel wear prediction method based on escape optimization according to claim 1 is characterized in that: The step of converting the temperature characteristic parameter, the pressure characteristic parameter and the charge characteristic parameter into a comprehensive load characteristic vector according to the same conversion rule, inputting the comprehensive load characteristic vector into a prediction model with the optimal prediction model parameters, calculating ablation wear prediction values ​​at different shooting moments, and performing Bayesian correction on the ablation wear prediction values ​​to obtain a corrected prediction result includes: Substituting the temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters collected in real time into the principal component analysis model, performing dimension reduction transformation on the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters according to the characteristic conversion matrix obtained in the training phase, generating a characteristic vector reflecting the comprehensive load of the current shooting process, inputting the characteristic vector into the prediction model, and calculating the ablation wear prediction value at the current moment based on the optimal prediction model parameters; A Bayesian prior distribution model is constructed, and historical prediction data is used as prior information to calculate the posterior probability distribution of the ablation wear prediction value. The prediction result is corrected based on the mean and variance of the posterior probability distribution, and the corrected prediction results at different shooting moments are calculated in sequence using a sliding time window method.

7. The gun barrel wear prediction method based on escape optimization according to claim 1 is characterized in that: The method comprises: constructing a gun barrel ablation wear degradation trend curve based on the corrected prediction result, extracting a wear change rate feature and a wear acceleration feature from the degradation trend curve, and providing an early warning of the ablation wear state of the gun barrel according to the wear change rate feature and the wear acceleration feature, including: The corrected prediction result sequence is subjected to time series fitting processing, and a degradation trend curve reflecting the ablation wear evolution process is constructed by using a polynomial regression method. The first-order derivative of the degradation trend curve at different times is calculated to obtain the wear change rate characteristics, and the second-order derivative of the degradation trend curve at different times is calculated to obtain the wear acceleration characteristics. Based on the wear change rate characteristics and the wear acceleration characteristics, a quantitative evaluation index of the ablation wear state of the gun barrel is established. The wear change rate threshold and the wear acceleration threshold are determined based on statistical analysis of historical data. When the wear change rate characteristic exceeds the wear change rate threshold or the wear acceleration characteristic exceeds the wear acceleration threshold, the early warning mechanism is triggered, and the early warning level of the ablation wear state of the gun barrel is determined based on the preset early warning level classification standard.

8. A gun barrel wear prediction device based on escape optimization, characterized in that: The device comprises: A data set construction module is used to collect historical ablation and wear data of multiple guns at different firing stages, extract temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the historical ablation and wear data, convert the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters into comprehensive load characteristic vectors, construct a corresponding relationship sample set between the comprehensive load characteristic vector and the ablation wear amount, and use a stratified sampling method to divide the corresponding relationship sample set into a training sample subset and a verification sample subset; An optimal parameter determination module is used to establish a search space for the escape optimization algorithm based on the training sample subset, set the initial position distribution of the search particle group, map the position information of the search particle group to the prediction model parameters, use the cross-validation method to calculate the prediction error corresponding to each particle, use the prediction error as the particle fitness function, sort the fitness of the particle group according to the fitness function, select the particle with the best fitness as the global optimal solution, update the speed and position of the particle based on the escape mechanism in each round of iteration, stop the search when the number of iterations reaches a preset value, and use the finally converged global optimal solution as the optimal prediction model parameter; The wear state prediction module is used to read the real-time shooting data of the artillery to be predicted, extract the temperature characteristic parameters, pressure characteristic parameters and charge characteristic parameters from the real-time shooting data, generate a comprehensive load characteristic vector according to the same conversion rule for the temperature characteristic parameters, the pressure characteristic parameters and the charge characteristic parameters, input the comprehensive load characteristic vector into the prediction model with the optimal prediction model parameters, calculate the ablation wear prediction values ​​at different shooting moments, perform Bayesian correction on the ablation wear prediction values ​​to obtain the corrected prediction results, construct the ablation wear degradation trend curve of the artillery barrel based on the corrected prediction results, extract the wear change rate characteristics and the wear acceleration characteristics from the degradation trend curve, and issue an early warning for the ablation wear state of the artillery barrel according to the wear change rate characteristics and the wear acceleration characteristics.

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 gun barrel wear prediction method based on escape optimization 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 gun barrel wear prediction method based on escape optimization as described in any one of claims 1 to 7 are implemented.

Images