Skip to main content
main-content

Über dieses Buch

This book describes load modeling approaches for complex work pieces and batch forgings, and demonstrates analytical modeling and data-driven modeling approaches for known and unknown complex forging processes. It overcomes the current shortcomings of modeling, analysis and control approaches, presenting contributions in three major areas: In the first, several novel modeling approaches are proposed: a process/shape-decomposition modeling method to help estimate the deformation force; an online probabilistic learning machine for the modeling of batch forging processes; and several data-driven identification and modeling approaches for unknown forging processes under different work conditions. The second area develops model-based dynamic analysis methods to derive the conditions of stability and creep. Lastly, several novel intelligent control methods are proposed for complex forging processes.

One of the most serious problems in forging forming involves the inaccurate forging conditions, velocity and position offered by the hydraulic actuator due to the complexity of both the deformation process of the metal work piece and the motion process of the hydraulic actuator. The book summarizes the current weaknesses of modeling, analysis and control approaches. are summarized as follows: a) With the current modeling approaches it is difficult to model complex forging processes with unknown parameters, as they only model the dynamics in local working areas but do not effectively model unknown nonlinear systems across multiple working areas; further, they do not take the batch forging process into account, let alone its distribution modeling. b) All previous dynamic analysis studies simplify the forging system to having a single-frequency pressure fluctuation and neglect the influences of non-linear load force. Further, they fail to take the flow equation in both valves and cylinders into account. c) Conventional control approaches only consider the linear deformation force and pay no attention to sudden changes and the motion synchronization for the multi-cylinder system, making them less effective for complex, nonlinear time-varying forging processes subject to sudden changes.

Inhaltsverzeichnis

Frontmatter

Background and Fundamental

Frontmatter

Chapter 1. Introduction

Abstract
This chapter is an introduction of the book. It briefly introduces the background, motivation and objective of the research, followed by a list of contributions and organization of the book.
Xinjiang Lu, Minghui Huang

Modeling of Forging Loads and Processes

Frontmatter

Chapter 2. Process/Shape-Decomposition Modeling for Deformation Force Estimation

Abstract
The deformation force of a forging is crucial for manufacturing high-quality products and for managing the machine’s physical condition. In this chapter, a process/shape-decomposition modeling method is presented to estimate this deformation force in the complex forging process. The complex forging process is first decomposed into a group of simple sub-processes using system knowledge. In each sub-process, the complex geometric shape is then decomposed into many easily modeled sub-units, upon which the deformation force model of each sub-unit is built as the sub-model. All sub-models are further integrated to form a global deformation force model for the whole forging process. The continuity of this global model are also considered and guaranteed.
Xinjiang Lu, Minghui Huang

Chapter 3. Distribution Modeling of Batch Forging Processes

Abstract
An effective model of batch forging processes is crucial in order to ensure the quality conformance control of batch productions. However, obtaining this model has proven difficult due to a variety of the raw forgings produced by manufacturing error, material variation, and geometric defects, etc. In this chapter, an online probabilistic extreme learning machine (ELM) is proposed to model batch forging processes. A probabilistic ELM is first developed to extract the distribution information of the batch forging processes from data. The stochastic property of the batch forging processes is then derived and processed. On this basis, a strategy is further developed to update the distribution model as new forging process data are collected. As a result, the model built is able to represent the distribution behavior of the batch forging processes well.
Xinjiang Lu, Minghui Huang

Chapter 4. Multi-level Parameter Identification Approach

Abstract
In the practical forging system, many unknown parameters differ greatly in magnitude and have different influences on the dynamic response when the system runs under different working conditions, rendering parameter identification difficult. This chapter develops a multi-level identification method for alleviating this difficulty. It divides the complex forging process into many sub-processes according to the process knowledge. Since each sub-process has simpler dynamic behavior and less unknown parameters than the original process, this renders parameter identification in each sub-process easier than that in the whole process. Both simulations and experiments demonstrate and test the effectiveness of the presented method.
Xinjiang Lu, Minghui Huang

Chapter 5. Novel LS-SVM Modeling Method for Forging Processes with Multiple Localized Solutions

Abstract
In this chapter, a novel least squares support vector machine (LS-SVM) method is developed for modeling unknown forging processes across multiple working regions. The proposed method integrates the advantages of local LS-SVM modeling and global regularization. Local LS-SVM modeling is performed to capture the local dynamics of each local working region. Global regularization is performed to minimize the global error and improve the global generalization of the local models. These features guarantee continuity and smoothness between the local LS-SVM models and avoid over-fitting of each local LS-SVM model. The algorithm developed here is simple and can represent the complex forging process across multiple working regions well.
Xinjiang Lu, Minghui Huang

Chapter 6. Forging Process Modeling via Multi-experiment Data

Abstract
As forging processes require to working across a large operation region, input/output samples do not easily satisfy the requirement of data-driven modeling because of many practical constraints involved. This renders forging processes difficult to model accurately. In this chapter, an operation-region-decomposition-based SVD/NN modeling method is presented for modeling of this type of processes. Because the complexity of the system at the local region is much lower than the original system throughout the operation region, the required input signal for modeling at a local region is easier to obtain than the one suitable for the whole region. An SVD/NN modeling method is then proposed to produce a low-order global model from these experiments at all local operation regions. The practical forging experiment finally demonstrates the effectiveness of the proposed method.
Xinjiang Lu, Minghui Huang

Chapter 7. Online Modeling Approach for Time-Varying Forging Processes

Abstract
The previous two chapters mainly discussed about off-line modeling as forging processes are time-invariant. In this chapter, a simple and effective online modeling approach is presented to model time-varying forging processes. This proposed method first constructs a model set for the time-varying forging process. All parameters in the model set are then identified online by using process data. An error minimization based match method is further developed to select a suitable model from the model set to reflect the present dynamic behavior of the forging process. Numerical cases and practical forging cases finally demonstrate the effectiveness of the proposed method.
Xinjiang Lu, Minghui Huang

Dynamic Analysis of Forging Processes

Frontmatter

Chapter 8. Model-Based Estimation and Prediction of System Dynamics

Abstract
The dynamic behavior of the forging process is crucial to fabrication of high-quality products and management of the machine’s physical condition. Estimating this dynamic behavior is difficult due to the complexity and strong nonlinearity of the forging process. In this chapter, a model-based dynamic analysis method is proposed to meet this challenge. A model of the complex forging process is first derived and a solving method is then developed to determine the model solution. Using this solution, the conditions of stable run, vibration, and creep are further derived. Experiments and simulations on a practical hydraulic driving process are finally performed to demonstrate and test the effectiveness of these analytical results.
Xinjiang Lu, Minghui Huang

Chapter 9. Dynamic Analysis of Closed-Loop Forging System

Abstract
The previous chapter mainly considers the dynamic analysis of the open-loop forging system. This chapter will develop an approach to estimate the dynamic behavior of the closed-loop forging system. The model of the closed-loop forging system is first derived and a solving method is then developed in order to find the velocity expression of the closed-loop forging system. Using this velocity expression, the dynamics of the closed-loop forging system is further estimated and the conditions of stability, vibration, and creep, as well as the relationships between the controller parameters and the constraints are also derived. These derived dynamic characteristics, conditions and relationships for different workpieces are further integrated and used to design the controller.
Xinjiang Lu, Minghui Huang

Intelligent Control of Complex Forging Processes

Frontmatter

Chapter 10. System-Decomposition Based Multi-level Control Approach

Abstract
A system-decomposition based multi-level control method is developed for complex forging processes with uncertainty in this chapter. The key idea in this proposed method is to decompose the system complexity into a group of simple sub-systems and the control task is shared by a group of simple sub-controllers. Under this framework, a sequence control strategy is developed to help these sub-controllers to handle the coupling between sub-systems, which can ensure the desirable global control performance for the complex forging process.
Xinjiang Lu, Minghui Huang

Chapter 11. Intelligent Integration Control for Time-Varying Forging Processes

Abstract
Time-varying forging process, big uncertainties and sudden changes from deformation force or driving force bring a great challenge to the high-quality forging control. In this chapter, a two-level modeling based intelligent integration control approach is proposed to meet this challenge. A two-level modeling method is first developed to take the time-varying forging process and the unpredictable sudden changes into account. An intelligent integration control method is then proposed to ensure the continuity and smoothness between the multiple localized nonlinear dynamics even if the forging processes have big uncertainties and sudden changes. The effectiveness of the proposed method is verified by both numerical simulations and experimental tests.
Xinjiang Lu, Minghui Huang

Chapter 12. Conclusion and Challenge

Abstract
This chapter summarizes all methods introduced in the book, and discusses future challenges in this area.
Xinjiang Lu, Minghui Huang
Weitere Informationen

Premium Partner

    Bildnachweise