The geometric error determines the product quality and function to a certain extent. Among them, the roundness error is one of the important indicators for evaluating the geometric error of shaft parts. With the increase of industrial requirements, there are higher requirements for the accuracy and efficiency of roundness error evaluation. However, most of the traditional roundness error evaluation models used in the industry can no longer meet the needs of current industrial processing in terms of efficiency and accuracy. This paper proposes a new roundness error evaluation method based on twin support vector machines (TWSVMs). First, according to the roundness error evaluation and the TWSVMs theory, the roundness error evaluation model with the TWSVMs is obtained. Then, experimental research and analysis are carried out, and the accuracy and efficiency of the traditional roundness evaluation method and the new method are compared. The research results show that the new roundness evaluation method based on the TWSVMs proposed in this paper can efficiently and accurately evaluate the roundness error, and can be applied to the online evaluation of the roundness error in industrial processing to improve the processing efficiency.

The roundness error is one of the criteria for evaluating the qualifications of industrial parts. With the development of science and technology—especially the popularity of the new generation of Product Geometry Specifications (GPS), the method of high-precision measurement of roundness error has become a hotspot in the field of industrial metrology. To evaluate roundness error accurately and quickly, a novel algorithm was proposed based on the analysis of the existing evaluation algorithms, which was called the improve zone searching algorithm (IZS). First, the roundness error evaluation model based on the IZS was obtained. Then, the accuracy and efficiency of the algorithm are verified by simulation calculating. According to the experimental researches comparing traditional methods with IZS, it turns out that the speed is faster than the traditional methods and the calculating result is more accurate. Also, this paper improves Pauta criterion so that we can get the measured points correctly and quickly. These studies can be applied to the measurement of roundness error in industrial practice and have very important application value.

Small-modulus worms are widely used in precision transmission mechanisms. The detection of surface defects on processed small-modulus worms mainly relies on manual inspection. However, this detection method has low detection efficiency and low accuracy. In response to this issue, this paper proposed a surface defect detection method for small-modulus worms based on deep learning. Firstly, based on the geometric features and material properties of small-modulus worms, designed a two-cameras worms image acquisition system to capture small-modulus worm images. And the captured images were annotated and classified to construct a small-modulus worms defect dataset. Secondly, the you only look once version 7 (YOLOv7) network model algorithm was studied and improved. A three-stage image preprocessing algorithm was added to the YOLOV7 model to reduce the interference in distorted areas. Next, Ghost-convolution, batch normalization, Silu activation (CBS) structures were employed to replace some of the conventional CBS structures, reducing computational complexity. And we also added a width–height balance weights and alpha structure to the efficient-IoU (EIOU) loss function. Improve the fitting performance of the model on the bounding box with large aspect ratio. Furthermore, three SimAM attention modules were added to the backbone, increasing the network’s focus on key areas. Finally, experimental validation was conducted, and the results demonstrated that the performance of our proposed method is better than other existing methods. The detection accuracy reached 95.7%, with a map@50 of 94.6%. Overall, the performance met the requirements for worm defect detection.

In order to solve the problem that the error evaluation delay and the size and roundness of workpiece can not meet the processing requirements at the same time in online measurement. First this paper proposes an online fusion control method for the size and roundness error of workpiece, which can not only improve the processing efficiency, but also improve the consistency of workpiece quality. Then, the Long Short-Term Memory(LSTM) is used to predict the workpiece information of online measurement, and the error is calibrated according to the predicted value. The LSTM is used to predict the workpiece information in real time, and the process parameters are adjusted in time when the prediction value is out of the theoretical boundary to avoid error accumulation. Finallyr the online grinding measurement experiment based on the LSTM is designed and carried out, and the relationship between the dimension of input tensor and the prediction accuracy is analyzed through the experimental results. The results show that the LSTM can accurately predict the grinding size sequence and roundness sequence, and has good universality. The small batch machining is carried out according to the experimental results. Statistical analysis shows that the grinding accuracy is significantly improved by using the fusion prediction and calibration method.

Laser additive manufacturing shows great promise for repairing high-value carburized gears, but the underlying relationships among thermal history, microstructure, and properties remain insufficiently quantified. This study uniquely integrates finite-element modeling with microstructural mapping to decipher thermo-mechanical coupling during gear repair. A thermal simulation model that combines a double-ellipsoidal heat source with phase-transformation kinetics achieves 91.1% accuracy in predicting melt pool depth and hardened-layer depth. The cladding process induces a substantial increase in subsurface hardness, primarily due to phase-transformation-induced refinement and regeneration of martensite during rapid thermal cycling. This results in a peak hardness of 64 HRC and a tensile strength of 2856 MPa in the secondary-hardened layer, both exceeding those of the original carburized substrate. The presence of beneficial compressive residual stresses further improves fatigue resistance. Spatial gradients in elastic modulus, strength, and hardness, measured by flat indentation and microhardness testing, are quantitatively correlated with simulated peak temperatures and predicted phase distributions. These correlations establish a causal link from the thermal history to phase transformations, microstructural evolution, and the resulting local hardness and strength. These findings provide a mechanistic foundation for precision repair and service-life prediction of high-carbon gear steels using laser additive manufacturing.

Laser cladding is promising for use in repairing carburized gear steels, but parameter selection is challenging when clad geometry, substrate thermal disturbance, and morphological quality must be considered. Fifteen tracks of NHT.22.A01 iron-based powder were deposited on carburized-quenched 18CrNiMo7-6 steel with varying laser powers, scanning speeds, and powder feed rates. A physics-informed cascaded-learning framework predicted track width, height, Ac₁-boundary depth, and quality. Leave-one-out out-of-fold width predictions were transferred to height and depth models to prevent target leakage. An auxiliary continuous quality index enabled the bi-objective optimization of quality and high-hardness layer depth, while multi-indicator process maps with local sensitivity analysis supported rapid parameter adjustment. Leave-one-out (R²) values for three geometric responses ranged from 0.934 to 0.968, and the maximum relative error at an unseen boundary condition was 4.3%. The fitted HAZ-depth/track width scaling coefficient (0.211) lay within the Rosenthal theoretical range (0.15–0.25), confirming the physical consistency of the cascade relationship. The optimization revealed a trade-off between quality and high-hardness layer depth; maximizing predicted high-hardness layer depth under $(Q_{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}}\ge 4.5)$ selected a condition already represented in the training set. Repeatability was supported by an independent batch replicate. By integrating leakage-controlled cascading, physically interpretable validation, constrained optimization, and sensitivity-resolved process mapping, the framework provides a transparent and practically applicable approach to small-sample laser cladding process design.