关键词： 齿轮故障   小波阈值   一维卷积神经网络   脉冲神经网络   诊断分类   抗噪性能  

摘要： 针对提高齿轮在少训练集条件下故障检测的准确率和在复杂噪声环境中的抗噪性能,提出一种基于小波阈值、一维卷积神经网络(1DCNN)和脉冲神经网络(SNN)进行齿轮故障诊断分类的设计方案。首先,将齿轮故障信号通过小波阈值去噪处理;然后,利用1DCNN从处理过后的信号中提取特征;最后,利用多层脉冲神经网络对去噪后的数据进行训练得到诊断分类结果。为了检验该方案的诊断效果,利用美国康涅狄格大学齿轮数据集进行验证,并与其他方法进行对比分析。实验结果表明,该方法在少训练集和加入噪声的条件下具有较好的识别准确率和抗噪性能。

关键词： Image reconstruction  

摘要： To address the issues with complex and variable vibration signals in gearboxes, which led to low diagnostic accuracy of existing gearbox fault diagnosis methods and the risk of weak fault features being overwhelmed by noise, a new fault diagnosis method was proposed. Firstly, entropy weight fusion algorithm was used to fusion vibration sensor signals at different locations, and vector weighted mean value algorithm (INFO) was used to optimize parameters in the variational mode decomposition (VMD) algorithm. A composite evaluation index was designed as the evaluation standard for parameter optimization. The Singular Kurtosis Differential Spectrum (SKD) method was employed to reconstruct the sensitive components. What's more, the time domain and frequency domain features were extracted from the reconstructed signals and fed into the CNN model for classification. Finally, Shapley Additive Explanations (Shap) value method was used to rank the importance of input features. The impact of different feature combinations on model classification and specific fault identification was analyzed. The proposed method was validated on the Planetary Gearbox Dataset from Southeast University. It was shown that using the proposed feature combination for fault diagnosis, the CNN model achieves an accuracy of 98.24%, which is higher than other combinations. This provides an effective set of feature indicators for planetary gearbox fault diagnosis. © 2025 Chinese Vibration Engineering Society. All rights reserved.

关键词： Genetic algorithms  

摘要： In order to meet the lightweight requirements of high-power density spiral bevel gear transmission，a multi-parameter optimization technology for spiral bevel gear blank was proposed. Taking the tooth number，modulus，and tooth width of the gear as design variables，the minimum sum of the volume of the gear pair as the objective function， and the working condition limitations， installation conditions，and multiple strength requirements as constraints，the genetic algorithm was synthesized to realize the optimization design of spiral bevel gear blank parameters. Examples with power of 0.75 MW and 5 MW were used to verify the effectiveness of the proposed technique. The results showed that: compared with traditional design methods such as adaptive method and nonlinear mathematical programming method，the proposed technique could achieve global optimization with a success rate of 100% thanks to the adaptability and parallel ability of genetic algorithm;in addition，compared with the design requirements，the redundancy of pitch diameter and tooth width in the 0.75 MW case was 31% and 19.23%，respectively，while that in the 5 MW case was slightly more than 4.69% and 2.65%，showing that the proposed technique could provide effective design margin for designers under different design conditions. © 2025 Beijing University of Aeronautics and Astronautics (BUAA). All rights reserved.

关键词： 深度迁移学习   Resnet50 CNN   故障识别模型   齿轮   轴承  

摘要： 目前在矿井生产过程中,采煤机摇臂故障率高,故障识别慢,给井下生产带来了一定的安全隐患,为了解决此问题,本文以传动系统故障高发的齿轮与轴承为研究对象,从故障的原理出发,构建采煤机摇臂故障识别模型,通过深度迁移学习,确定了卷积层网络的全局参数,测试模型的故障识别准确率达到95%,能够对采煤机安全运行以及矿井安全生产起到一定的保障作用。

关键词： 牵引电机   小齿轮轴   裂损   质量卡控  

摘要： 运用机车牵引电机小齿轮轴存在裂损质量故障问题,严重威胁机车运用安全。针对HXD_(1)型牵引电机小齿轮轴裂损故障问题,通过从结构受力、宏观及微观分析、线路及载荷的影响,对裂损故障进行原因分析,提出了强化轴报分析及优化探伤管理2个方面针对性的重要质量卡控对策,保证牵引电机小齿轮轴运用质量,通过现场运用实施效果明显,对及时发现牵引电机小齿轮轴裂损、防止蔓延势头有重要指导意义。

关键词： CAK4085D数控车床   系统建模   强度计算   应力分析  

摘要： 大功率数控车床齿轮传动系统在工作过程中精度要求相对较高,在不同的加工工况情况下容易产生加工误差,影响加工精度。以CAK4085D机床主传动系统为例,建立3种工况下的数学及三维模型,分析齿轮结构的工作过程,并进行强度理论计算,验证模型的可靠性。借助Romax Designer软件分析主轴箱齿轮在不同工况下的应力、轴的偏移、啮合错位的具体情况,获取结构的静态特性,为机床主传动系统优化设计提供重要理论依据。

关键词： 矿井提升绞车   行星齿轮变速机构   有限元分析  

摘要： 根据工作条件及技术要求设计了矿井作业提升绞车2K-H行星齿轮变速机构并以此为研究对象,采用有限元分析方法进行受力分析、可靠性分析,并将仿真分析结果与理论计算值进行比较。结果表明,在满载情况下,行星轮系中齿根最大弯曲应力与齿面最大接触应力均小于理论计算值,均满足作业要求,可以为进一步轻量化设计绞车变速机构和动力输出装置提供参考依据。

关键词： 变速箱   拉维纳行星齿轮机构   大马力无级变速拖拉机  

摘要： 大马力无级变速拖拉机作为高端农业装备的代表,近几年的市场占有率在逐年提升。变速器作为大马力无级变速拖拉机的核心部件至关重要,其中无级变速主要依靠行星轮系,大马力无级变速拖拉机变速箱采用的行星齿轮结构主要有两类:辛普森式行星齿轮机构和拉维纳式行星齿轮机构。拉维纳行星轮系是一种复合行星齿轮机构,它由一个单行星轮式行星排和一个双行星轮式行星排组合而成[1]。

关键词： 过盈配合   高精重载   齿轮精度  

摘要： 过盈配合由于其没有切口效应、不影响零件的承载能力和使用寿命的特点,普遍应用于齿轮与轴间的连接,但过盈配合产生的装配应力与变形易导致轮齿微观形貌偏离设计基准。高精重载齿轮的几何精度稳定性是保障传动系统可靠性的核心要素,针对过盈引发的齿轮几何精度变化缺乏系统性解析的问题,文章以重载渐开线齿轮副为对象,采用理论建模、数值仿真与实物检测相结合的方法,揭示过盈配合对齿形及齿向精度的作用规律。

关键词： Image segmentation  

摘要： Combined with the advantages of the reconstructed gear test bench to obtain the working tooth surface image of the gear online, the method of gear pitting identification based on machine vision was discussed, and the experimental research was carried out. In view of the scarcity of gear pitting data, the deep convolutional generative adversarial network (DCGAN) model was used to realize the diversification and high-quality augmentation of the gear pitting samples. Based on the previous researeh by the authors, the effective working tooth surface area of the gear was extracted, and the tooth surface tilt correction were well as distortion correction are realized. By introducing the efficient channel attention, the U -Net model was improved, and the accurate segmentation of the interested region of the gear pitting image was realized. On this basis, by counting the historical pitting rate of gears, a gear pitting identification model based on image signals was constructed, and the gear pitting identification was realized. The results show that the gear pitting identification method based on machine vision technology is feasible, and the recognition accuracy based on the DCGAN and U -Net models can reach 93. 56%. The research provides a more direct and reliable method for gear pitting identification, and has certain reference value for the condition monitoring of mechanical equipments. © 2025 Chinese Vibration Engineering Society. All rights reserved.