Prediction and optimization of hobbing gear geometric deviations

•The PSO-BP algorithm is improved by considering flexible inertia weights.•The improved PSO-BP algorithm is applied to predict the gear geometric errors accurately.•The optimization of gear hobbing processing parameters is achieved by the predicted results. Hobbing is a precision gear manufacturing...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Mechanism and machine theory Jg. 120; S. 288 - 301
Hauptverfasser: Sun, Shouli, Wang, Shilong, Wang, Yawen, Lim, Teik C., Yang, Yong
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.02.2018
Schlagworte:
Gear geometric deviation
Gear hobbing process
IPSO-BP neural network algorithm
Parameters optimization
Precision prediction
IPSO-BP neural network algorithm
Precision prediction
Gear hobbing process
Parameters optimization
Gear geometric deviation
ISSN:0094-114X, 1873-3999
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•The PSO-BP algorithm is improved by considering flexible inertia weights.•The improved PSO-BP algorithm is applied to predict the gear geometric errors accurately.•The optimization of gear hobbing processing parameters is achieved by the predicted results. Hobbing is a precision gear manufacturing process with high efficiency and low cost. High precision gears are essential components for high-end equipment to meet the requirement of extreme operation conditions. In order to further improve the precision of gear hobbing process as well as lower the gear manufacturing cost, this paper proposes a model for predicting the hobbing gear geometric deviations and optimizing the hobbing processing technique. The relationship between gear hobbing processing technique and gear geometric deviation is modeled applying the improved Particle Swarm Optimization and Back Propagation algorithm. The performance of the proposed method is compared with the existing optimization and back propagation method and validated by experiments. The accuracy of both algorithms is evaluated by the Root Mean Square Error between the predicted and experimental values. The result shows that the gear geometric deviations predicted by the proposed algorithm yields better performance and are in reasonably good agreement with experimental data. Employing the proposed model, the gear hobbing process parameters can be optimized to minimize gear geometric errors, and thus improve the gear manufacturing precision.
ISSN:0094-114X
1873-3999
DOI:10.1016/j.mechmachtheory.2017.09.002