A method for evaluating the wear state of a transmission case

Abstract

A method for evaluating the wear state of a transmission case, belonging to the technical field of wear state evaluation, includes the following steps: Step S01, collect wear particle data in the transmission case oil; Step S02, establish evaluation indexes for the wear particle data in Step S01; Step S03, use the improved parrot optimization algorithm IPO to optimize the penalty function in the projection pursuit model PPC to obtain the best objective function; Step S04, use the best objective function and evaluation indexes to construct a wear state evaluation model IPO-PPC, etc. The present invention improves the parrot optimization algorithm, so that its convergence speed and accuracy are greatly improved. The improved parrot optimization algorithm is used to optimize the penalty function in the projection pursuit method, establish a wear state evaluation model, and combine six evaluation indexes of oil abrasive particle data to realize the rapid and accurate identification of the wear state of the transmission case.

Description

Technical Field

[0001] The present invention belongs to the technical field of transmission box condition assessment, and particularly relates to a method for assessing the wear condition of a transmission box. Background Art

[0002] Under the complex and changeable operating environment of equipped vehicles, the transmission box bears a high-load and high-intensity working state, and its internal mechanical components are prone to wear, fatigue, and even failure. Real-time assessment of the wear condition of the transmission box can not only effectively predict the service life of key components, avoid mission interruption or combat damage risks caused by sudden failures, but also provide a scientific basis for optimizing the equipment maintenance strategy, thereby improving the combat availability and full-life cycle management level of the vehicle.

[0003] Traditional wear condition assessment methods usually rely on vibration signal analysis or regular disassembly and inspection, but these methods have problems such as high data acquisition cost, poor real-time performance, or insufficient diagnostic accuracy. To solve the above problems, wear condition assessment can adopt an assessment method based on oil particle information, which can more accurately reflect the internal wear condition of the transmission box and has the advantages of non-invasiveness and real-time detection.

[0004] Among many data analysis algorithms, the Projection Pursuit Algorithm PPC shows significant advantages due to its unique dimensionality reduction ability and non-linear feature extraction performance. Compared with traditional principal component analysis or linear regression algorithms, the Projection Pursuit Algorithm PPC can better capture the potential non-linear relationships and complex patterns in oil particle data, improving the accuracy and robustness of wear assessment. In addition, the Projection Pursuit Algorithm PPC shows strong stability when dealing with high-dimensional and noisy datasets, making it particularly suitable for wear condition assessment under complex working conditions of vehicle transmission boxes. However, the Projection Pursuit Algorithm PPC is strongly dependent on initial parameters, such as the penalty function, and is prone to falling into local optimal solutions, and its performance urgently needs to be improved. Summary of the Invention

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method for assessing the wear condition of a transmission box. By improving the Parrot Optimization Algorithm, its convergence speed and accuracy are greatly improved. The penalty function in the Projection Pursuit Method is optimized by the improved Parrot Optimization Algorithm, and a wear condition assessment model is established. Combining six evaluation indexes of oil particle data, the wear condition of the transmission box can be quickly and accurately identified.

[0006] To achieve the above object, the main technical solutions adopted by the present invention include:

[0007] A method for assessing the wear condition of a transmission box, comprising the following steps:

[0008] Step S01, collecting wear particle data in the transmission box oil;

[0009] Step S02: Establish evaluation indicators for the wear particle data in Step S01, and divide the wear particle data into a training data set and a test data set;

[0010] Step S03: Optimize the penalty function in the projection pursuit model PPC using the improved parrot optimization algorithm IPO to obtain the best objective function; the improved parrot optimization algorithm includes introducing an adaptive inertia weight in the foraging behavior and introducing a t-distribution perturbation in the communication behavior;

[0011] Step S04: Introduce the best objective function in Step S03 and the evaluation indicators in Step S02 into the projection pursuit model PPC to construct a wear state evaluation model IPO-PPC;

[0012] Step S05: Train the wear state evaluation model IPO-PPC constructed in Step S04 using the training data set in Step S01;

[0013] Step S06: Test the wear state evaluation model IPO-PPC constructed in Step S05 using the test data set in Step S01;

[0014] Step S07: Use the wear state evaluation model IPO-PPC that passes the test in Step S06 to evaluate the wear state of the transmission case and output the evaluation result.

[0015] Furthermore, in Step S03, the adaptive inertia weight gradually decreases in a non-linear manner according to the increase in the number of iterations. The improved formula of the parrot optimization algorithm IPO after introducing the adaptive inertia weight in the foraging behavior is as follows:

[0016] ;

[0017] where t is the current number of iterations; T is the maximum number of iterations; is the position of the i-th individual at the t-th iteration; is the position of the i-th individual at the (t + 1)-th iteration; X best is the best position from initialization to the current search; is the average position within the current population; Levy(dim) is the Levy distribution; dim is the number of problem dimensions; η is a cheap position coefficient that changes with the increase in the number of iterations; δ is the first weight coefficient, β is the second weight coefficient; ξ is the adaptive inertia weight.

[0018] Furthermore, the improved formula of the improved parrot optimization algorithm for introducing a t-distribution perturbation in the communication behavior is as follows:

[0019] ;

[0020] Among them, rand(0,1) is a random number uniformly distributed between [0,1], t(θ) is a random perturbation value following a t-distribution with degree of freedom θ, C Iter is the number of iterations, and P is a random number between [0,1].

[0021] Furthermore, in the step S02, six evaluation indexes are established according to the size information in the wear particle data, including 0 - 100μm, 100μm - 200μm, 200μm - 300μm, 300μm - 450μm, 450μm - 600μm, and above 600μm.

[0022] The beneficial effects of the present invention are as follows:

[0023] 1. The projection pursuit algorithm is used to analyze the collected data, reducing the dimension of high-dimensional data, reducing data redundancy, and improving data quality. This algorithm is efficient and practical, and is suitable for the wear state monitoring scenario;

[0024] 2. The foraging behavior and communication behavior of the parrot optimization algorithm are improved by introducing an adaptive inertia weight and a t-disturbance distribution, solving the problem of its weak convergence, improving the convergence speed and expanding the search range;

[0025] An adaptive inertia weight is introduced into the foraging behavior of the parrot optimization algorithm, making the algorithm have stronger global search ability and a wider search range, which helps the algorithm break away from the bondage of local optimal solutions;

[0026] A t-distribution perturbation is introduced into the communication behavior of the parrot optimization algorithm. The t-distribution combines the advantages of the Gaussian distribution and the Cauchy distribution, that is, the convergence speed in the later stage of the algorithm can be guaranteed through Gaussian mutation, while the Cauchy mutation maintains the population diversity. In the communication stage, the information of other individuals in the population is introduced into the global search update mechanism, enhancing the diversity of the search space;

[0027] 3. The penalty function in the projection pursuit algorithm is optimized by the improved parrot optimization algorithm, which can determine the optimal projection vector more accurately and quickly, enhance the generalization performance, realize the rapid and accurate identification of the wear state of the transmission box, and provide a basis for the offline evaluation of future equipment. Description of the Drawings

[0028] Figure 1 It is a flowchart of a method for evaluating the wear state of a transmission box according to the present invention. Detailed Embodiments

[0029] In order to better explain the present invention for easy understanding, the present invention will be described in detail below with reference to the drawings through specific embodiments.

[0030] As Figure 1 shown, an embodiment of the present invention provides a method for evaluating the wear state of a transmission case, including the following steps:

[0031] Step S01: Collect wear particle data in the transmission case oil. Specifically, it is collected through an oil sensor.

[0032] Step S02: Establish evaluation indexes for the wear particle data in step S01, and divide the wear particle data into a training data set and a test data set.

[0033] Specifically, six evaluation indexes are established according to the size information in the wear particle data, including 0 - 100μm, 100μm - 200μm, 200μm - 300μm, 300μm - 450μm, 450μm - 600μm, and above 600μm. The wear particle data is divided with 70% as the training data set and 30% as the test data set.

[0034] Step S03: Use the improved parrot optimization algorithm IPO to optimize the penalty function in the projection pursuit model PPC to obtain the best objective function; the improved parrot optimization algorithm includes introducing an adaptive inertia weight in the foraging behavior and introducing a t-distribution perturbation in the communication behavior.

[0035] The specific algorithm and improvement of the parrot optimization algorithm IPO are as follows:

[0036] (1) Population initialization, and the initialization formula is:

[0037]

[0038] where ub and lb are the lower and upper bounds of the space search; is the position of the i-th parrot at the initial stage; rand(0,1) is a random number uniformly distributed between [0,1].

[0039] (2) Foraging behavior

[0040] When looking for food, they mainly estimate the approximate position of the food by observing the position of the food or considering the position of the owner, and its mathematical formula is:

[0041] ;

[0042] where t is the current iteration number; T is the maximum iteration number; is the position of the i-th individual at the t-th iteration; is the position of the i-th individual at the (t + 1)-th iteration; X best is the best position from initialization to the current search; is the average position within the current population; Levy(dim) represents the Levy distribution; dim is the number of problem dimensions; η represents the cheap position coefficient that changes with the increase in the number of iterations; N is the number of populations; μ, ν are positive integers within (0, dim); Γ is the Gamma function; γ is assigned a value of 1.5; σ is the probability density of the Levy flight distribution.

[0043] An adaptive inertia weight is introduced in the foraging behavior. The weight gradually decreases in a non-linear manner according to the increase in the number of iterations. At the initial stage of iteration, a larger weight value gives the algorithm stronger global search ability and a wider search range, which helps the algorithm break free from the bondage of local optimal solutions. As the number of iterations continues to increase, the value of the weight will gradually decrease, enabling a more detailed exploration of the area near the optimal solution, thereby improving the optimization accuracy of the algorithm and accelerating the convergence speed. The improved formula is:

[0044]

[0045] Among them, δ is the first weight coefficient, β is the second weight coefficient; ξ is the adaptive inertia weight.

[0046] (3)Stay behavior

[0047] The stay behavior of parrots mainly includes suddenly flying to any part of the owner's body and staying there for a period of time. The mathematical expression of this process is:

[0048]

[0049] Among them, ones(1, dim) is a vector of all 1s in dimension dim.

[0050] (4)Communication behavior

[0051] Parrots will communicate closely within the group. This communication behavior includes communication when flying towards the flock and not flying towards the flock. Assume that the probabilities of these two behaviors are equal, and the average position of the current population is used to symbolize the center of the group. The mathematical expression of this process is:

[0052]

[0053] A t-distribution perturbation is introduced in the communication behavior. The t-distribution combines the advantages of the Gaussian distribution and the Cauchy distribution, that is, the convergence speed of the algorithm in the later stage can be guaranteed through Gaussian mutation, while the Cauchy mutation maintains the population diversity. During the communication stage, the information of other individuals in the population is introduced into the global search update mechanism, enhancing the diversity of the search space. The improved new formula is:

[0054]

[0055] where \(t(\theta)\) is a random perturbation value following a t-distribution with degree of freedom \(\theta\); \(C\) Iter is the number of iterations; \(P\) is a random number between \([0, 1]\).

[0056] (5)Fear behavior towards strangers

[0057] Generally speaking, birds show natural fear towards strangers. Their behavior of keeping a distance from unfamiliar individuals and seeking a safe environment with the owner can be expressed by the following formula:

[0058]

[0059] In the formula, the second term represents the process of reorienting to fly towards the owner, and the last term represents the process of moving away from strangers.

[0060] The specific steps of the projection pursuit algorithm PPC are as follows:

[0061] (1)Normalization of sample evaluation index data:

[0062] To eliminate the influence of different dimensions of abrasive particle concentration, the wear particle data collected in step S01 are divided into multiple groups of wear particle data according to the evaluation indexes in step S02, and the wear particle data are normalized. The formula is as follows:

[0063]

[0064] In the formula, \(x\) * i is the \(i\)-th data index value; \(x\) imax， \(x\) imin are the maximum and minimum values of the \(i\)-th data index value respectively; \(x\) i is the eigenvalue normalization sequence.

[0065] (2)Construct a projection pursuit model

[0066] Construct an objective function \(J(a)\) in the form of:

[0067]

[0068] where \(Q(a)\) is the main function used to reflect the distribution characteristics of the projection data; \(y\) i is the projection value of the \(i\)-th data point; is the mean value of the projection values; \(Var(y)\) is the projection variance used to maximize the projection information content of the data; \(P(a)\) is the penalty function used to constrain the projection direction; \(\lambda\) is the penalty function coefficient; \(d\) is the dimension of the original high-dimensional space; \(a\) is the projection vector; \(\|a\|\) is the two-norm of the projection vector; \(a\) j is the \(j\)-th component of the projection vector \(a\); \(a\) T is the transpose of the projection vector.

[0069] (3)Determine the projection vector

[0070] Through the objective function J(a), obtain the optimal projection vector a*:

[0071]

[0072] (4)Project onto the low-dimensional space

[0073] Project the wear particle data onto the low-dimensional space through the optimal projection vector a*:

[0074]

[0075] (5)Analyze the projection boundary

[0076] Through historical data analysis, determine the distribution range of the projection values and determine the operating projection range (such as boundary values). Compare whether the projection values of the real-time data fall within the normal range to judge the wear state.

[0077] Use the improved parrot optimization algorithm to optimize the penalty function in the projection pursuit model PPC to obtain the optimal objective function. The specific execution process of the optimization can be obtained by those skilled in the art according to the content of the present invention and the prior art, and will not be elaborated here.

[0078] Step S04: Introduce the optimal objective function in step S03 and the evaluation index in step S02 into the projection pursuit model PPC to construct a wear state evaluation model IPO-PPC;

[0079] Step S05: Use the training data set in step S01 to train the wear state evaluation model IPO-PPC constructed in step S04;

[0080] Step S06: Use the test data set in step S01 to test the wear state evaluation model IPO-PPC constructed in step S05;

[0081] Step S07: Use the wear state evaluation model IPO-PPC qualified in step S06 to evaluate the wear state of the transmission case and output the evaluation result.

[0082] The specific steps for evaluating the transmission case with the wear state evaluation model IPO-PPC are as follows:

[0083] (1)Establish an evaluation index based on the size information of the wear particle data of the transmission case;

[0084] (2)Normalize the wear particle data according to the evaluation index to generate an initial population;

[0085] (3) The penalty function of the projection pursuit model PPC is optimized by using the improved parrot optimization algorithm IPO to obtain the optimal objective function;

[0086] (4) Construct a wear state evaluation model IPO-PPC;

[0087] (5) Set the boundary value through historical data, compare the projection vector in the wear state evaluation model IPO-PPC with the boundary value, and determine the optimal projection vector;

[0088] (6) Determine the optimal projection value and conduct wear state evaluation.

[0089] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Modifications, alterations, substitutions, and variations made by those of ordinary skill in the art to the above embodiments all fall within the scope of the present invention.

Claims

1. A method for evaluating the wear state of a transmission case, characterized in that, It includes the following steps: Step S01, collect the wear particle data in the transmission oil; Step S02, establish evaluation indexes for the wear particle data in Step S01, and divide the wear particle data into a training data set and a test data set; Step S03, optimize the penalty function in the projection pursuit model PPC by using the improved parrot optimization algorithm IPO to obtain the best objective function; the improved parrot optimization algorithm includes introducing an adaptive inertia weight in the foraging behavior and introducing a t-distribution perturbation in the communication behavior; Step S04, introduce the best objective function in Step S03 and the evaluation indexes in Step S02 into the projection pursuit model PPC to construct a wear state evaluation model IPO-PPC; Step S05, train the wear state evaluation model IPO-PPC constructed in Step S04 by using the training data set in Step S01; Step S06, test the wear state evaluation model IPO-PPC constructed in Step S05 by using the test data set in Step S01; Step S07, use the wear state evaluation model IPO-PPC qualified in Step S06 to evaluate the wear state of the transmission and output the evaluation result; In Step S03, the adaptive inertia weight gradually decreases in a non-linear manner according to the increase of the iteration times. The improved formula of the parrot optimization algorithm IPO after introducing the adaptive inertia weight in the foraging behavior is as follows: ； where, t is the current iteration number; T is the maximum iteration number; is the position of the i-th individual at the t-th iteration; is the position of the i-th individual at the (t + 1)-th iteration; X best is the best position from initialization to the current search; is the average position within the current population; Levy(dim) is the Levy distribution; dim is the number of problem dimensions; η is a cheap position coefficient that changes with the increase of the iteration number; δ is the first weight coefficient, β is the second weight coefficient; ξ is the adaptive inertia weight; The improved formula of the improved parrot optimization algorithm for introducing the t-distribution perturbation in the communication behavior is as follows: ； where rand(0,1) is a random number uniformly distributed between [0,1], t(θ) is a random perturbation value following a t-distribution with degrees of freedom θ, C Iter is the number of iterations, and P is a random number between [0,1].

2. The wear state evaluation method of a transmission case according to claim 1, wherein: In Step S02, six evaluation indexes are established according to the size information in the wear particle data, including 0-100μm, 100μm-200μm, 200μm-300μm, 300μm-450μm, 450μm-600μm, and above 600μm.

Images