Developing a neural network-based run-to-run process controller for chemical-mechanical planarization

Wang, Gou‐Jen; Yu, Cheng-Hsiong

doi:10.1007/s00170-004-2451-6

Cited by 13 publications

(8 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main consideration of a CMP process is to simultaneously maximize the material removal rate and minimize the within-wafer nonuniformity. Thus, in the process model (1), the output y k can be denoted as an aggregative index of the material removal rate and the within-wafer nonuniformity, the input u k is the relative speed between the plashing pad and the plashed wafer [36].…”

Section: Simulation Studymentioning

confidence: 99%

Takagi–Sugeno Model Based Analysis of EWMA RtR Control of Batch Processes With Stochastic Metrology Delay and Mixed Products

Zheng

Wong

Wang

et al. 2014

IEEE Trans. Cybern.

View full text Add to dashboard Cite

In many batch-based industrial manufacturing processes, feedback run-to-run control is used to improve production quality. However, measurements may be expensive and cannot always be performed online. Thus, the measurement delay always exists. The metrology delay will affect the stability and performance of the process. Moreover, since quality measurements are performed offline, delay is not fixed but is stochastic in nature. In this paper, a modeling approach Takagi-Sugeno (T-S) model is presented to handle stochastic metrology delay in both single-product and mixed-product processes. Based on the Markov characteristics of the delay, the membership of the T-S model is derived. Performance indices such as the mean and the variance of the closed-loop output of the exponentially weighted moving average (EWMA) control algorithm can be derived. A steady-state error of the process output always exists, which leads the output deviating from the target. To remove the steady-state error, an algorithm called compensatory EWMA run-to-run (COM-EWMA-RtR) algorithm is proposed. The validity of the T-S model analysis and the efficiency of the proposed COM-EWMA-RtR algorithm are confirmed by simulation.

show abstract

Section: Simulation Studymentioning

confidence: 99%

Takagi–Sugeno Model Based Analysis of EWMA RtR Control of Batch Processes With Stochastic Metrology Delay and Mixed Products

Zheng

Wong

Wang

et al. 2014

IEEE Trans. Cybern.

View full text Add to dashboard Cite

show abstract

“…The algorithm of BP, consisting of forward information transmission and reverse error propagation, has a hierarchical network that comprises input layer, hidden layer, and output layer. Figure shows the topology of a BP neural network …”

Section: The Establishment Of Gra–ga–bp–mcrc Hybrid Algorithmmentioning

confidence: 99%

A Hybrid Algorithm for the Reliability Evaluation Models of Chemical Systems

Liu

Wei

et al. 2016

Quality & Reliability Eng

View full text Add to dashboard Cite

Chemical processes are complex dynamic systems. With the chemical industry under pressure to introduce improvements through the greater use of automation and intelligence, the need for comprehensive reliability evaluation has become more urgent both theoretically and practically. The employment of intelligent algorithms based on factory data has been the recent research hotspots. But for complex systems with available data on a smaller scale, reliability evaluation models have suffered on such problems as a result of instability and over-fitting, which have to be resolved. The GRA-GA-BP-MCRC hybrid algorithm was proposed. It combined the two-step genetic algorithm (GA)-back propagation (BP) and grey relational analysis (GRA) with Markov chain residual correction (MCRC). Based on the technical characteristics and the management demands, 46 influential factors of process reliability were introduced, which covered man, machine, material, method, and environment. For model convergence to be assured, GRA and attribute reduction rule were introduced. Meanwhile, based on the correlation of the factors, the two-step GA-BP was proposed to resolve the over-fitting problem of artificial neural network with complex input parameters. As well, MCRC was applied to modify the GA-BP error. The resulting average relative error of the hybrid algorithm was 2.36%, while the conventional algorithm was 10.28%.T he Internet of things, big data, artificial intelligence, and mobile intelligent terminals are enabling new operation modes in the chemical industry. Cyber-physical systems allow enterprises to exhibit enhanced perception, analysis, prediction, coordination, and optimization, and the use of intelligence becomes the most important capability to improvements. Some scholars have introduced the concept of an intelligent factory 1,2 in which the online management of health, safety, and environment (HSE) is one of the most important requirements. 3-5 HSE evaluation models have become essential models for enterprises to continuously improve through using intelligent measures.System reliability is one of the important factors to evaluate the HSE state of chemical industries and the system's probability of completing assigned tasks under certain conditions without fault. 6,7 Progress has been made in reliability theories applicable to the chemical industry. Saleh researched reliability theories of complex systems. 8 Gheorghe introduced the integrated regional risk assessment and safety management for energy and several complex industrial systems. 9 Gang presented a novel condition-based maintenance system that used a reliability-centered maintenance mechanism to optimize maintenance cost, and employed data fusion strategy to improve condition monitoring, health assessment, and prognostics. 10 Liu established the fault-oriented reliability model to evaluate chemical systems for safety, environment, and economics. 11 However, these reliability models of chemical systems also have limitations under complex conditions in which there are vari...

show abstract

“…Artificial neural networks have been widely used for nonlinear function approximation, pattern identification, system modeling, and control [25][26][27][28][29][30][31][32][33]. For example, Wu et al employed the neural network to predict molten temperature of blended coal ash and estimate the degree of slagging of the coal-fired boiler in power plant [29,30].…”

Section: Artificial Neural Network For Stribeck Curvementioning

confidence: 99%

Lubrication in Chemical and Mechanical Planarization

Wu¹,

Liao²

2016

Advances in Tribology

View full text Add to dashboard Cite

Chemical mechanical planarization (CMP) has been widely used in integrated circuit (IC) processing to achieve both local and global surface planarity through combined chemical and mechanical actions. The lubrication plays a significant role in CMP and can be determined by the Stribeck curve since it provides direct evidence of the extent of contact among wafer, pad asperities, and slurry particles. The advancements in the construction of the Stribeck curve are highlighted in this chapter. Traditionally, the procedure for constructing the Stribeck curve is as follows: (1) polish wafers at various pressures and sliding velocities to obtain the coefficient of friction (COF) values; (2) plot the experimental data as COF vs. Sommerfeld number; (3) construct the Stribeck curve via curve fitting. Recently, an alternative method was presented to construct the Stribeck curve via only performing one wafer polishing experiment. Pressure and sliding velocity are varied separately or together for a desired length of time, so that multiple measurements can be taken within one run. In this study, a back-propagation (BP) neural network is proposed to construct the Stribeck curve. Results show that the BP neural network could construct a more accurate Stribeck curve and thus could better provide insight into the lubrication mechanism of CMP processes.

show abstract

Developing a neural network-based run-to-run process controller for chemical-mechanical planarization

Cited by 13 publications

References 17 publications

Takagi–Sugeno Model Based Analysis of EWMA RtR Control of Batch Processes With Stochastic Metrology Delay and Mixed Products

Takagi–Sugeno Model Based Analysis of EWMA RtR Control of Batch Processes With Stochastic Metrology Delay and Mixed Products

A Hybrid Algorithm for the Reliability Evaluation Models of Chemical Systems

Lubrication in Chemical and Mechanical Planarization

Contact Info

Product

Resources

About