Matching Items (129)
Filtering by
- Language: English
Description
A good production schedule in a semiconductor back-end facility is critical for the on time delivery of customer orders. Compared to the front-end process that is dominated by re-entrant product flows, the back-end process is linear and therefore more suitable for scheduling. However, the production scheduling of the back-end process is still very difficult due to the wide product mix, large number of parallel machines, product family related setups, machine-product qualification, and weekly demand consisting of thousands of lots. In this research, a novel mixed-integer-linear-programming (MILP) model is proposed for the batch production scheduling of a semiconductor back-end facility. In the MILP formulation, the manufacturing process is modeled as a flexible flow line with bottleneck stages, unrelated parallel machines, product family related sequence-independent setups, and product-machine qualification considerations. However, this MILP formulation is difficult to solve for real size problem instances. In a semiconductor back-end facility, production scheduling usually needs to be done every day while considering updated demand forecast for a medium term planning horizon. Due to the limitation on the solvable size of the MILP model, a deterministic scheduling system (DSS), consisting of an optimizer and a scheduler, is proposed to provide sub-optimal solutions in a short time for real size problem instances. The optimizer generates a tentative production plan. Then the scheduler sequences each lot on each individual machine according to the tentative production plan and scheduling rules. Customized factory rules and additional resource constraints are included in the DSS, such as preventive maintenance schedule, setup crew availability, and carrier limitations. Small problem instances are randomly generated to compare the performances of the MILP model and the deterministic scheduling system. Then experimental design is applied to understand the behavior of the DSS and identify the best configuration of the DSS under different demand scenarios. Product-machine qualification decisions have long-term and significant impact on production scheduling. A robust product-machine qualification matrix is critical for meeting demand when demand quantity or mix varies. In the second part of this research, a stochastic mixed integer programming model is proposed to balance the tradeoff between current machine qualification costs and future backorder costs with uncertain demand. The L-shaped method and acceleration techniques are proposed to solve the stochastic model. Computational results are provided to compare the performance of different solution methods.
ContributorsFu, Mengying (Author) / Askin, Ronald G. (Thesis advisor) / Zhang, Muhong (Thesis advisor) / Fowler, John W (Committee member) / Pan, Rong (Committee member) / Sen, Arunabha (Committee member) / Arizona State University (Publisher)
Created2011
Description
Due to the growing concerns on the depletion of petroleum based energy resources and climate change; fuel cell technologies have received much attention in recent years. Proton exchange membrane fuel cell (PEMFCs) features high energy conversion efficiency and nearly zero greenhouse gas emissions, because of its combination of the hydrogen oxidation reaction (HOR) at anode side and oxygen reduction reaction (ORR) at cathode side. Synthesis of Pt nanoparticles supported on multi walled carbon nanotubes (MWCNTs) possess a highly durable electrochemical surface area (ESA) and show good power output on proton exchange membrane (PEM) fuel cell performance. Platinum on multi-walled carbon nanotubes (MWCNTs) support were synthesized by two different processes to transfer PtCl62- from aqueous to organic phase. While the first method of Pt/MWCNTs synthesis involved dodecane thiol (DDT) and octadecane thiol (ODT) as anchoring agent, the second method used ammonium lauryl sulfate (ALS) as the dispersion/anchoring agent. The particle size and distribution of platinum were examined by high-resolution transmission electron microscope (HRTEM). The TEM images showed homogenous distribution and uniform particle size of platinum deposited on the surface of MWCNTs. The single cell fuel cell performance of the Pt/MWCNTs synthesized thiols and ALS based electrode containing 0.2 (anode) and 0.4 mg (cathode) Pt.cm-2 were evaluated using Nafion-212 electrolyte with H2 and O2 gases at 80 oC and ambient pressure. The catalyst synthesis with ALS is relatively simple compared to that with thiols and also showed higher performance (power density reaches about 1070 mW.cm-2). The Electrodes with Pt/MWCNTs nanocatalysts synthesized using ALS were characterized by cyclic voltammetry (CV) for durability evaluation using humidified H2 and N2 gases at room temperature (21 oC) along with commercial Pt/C for comparison. The ESA measured by cyclic voltammetry between 0.15 and 1.2 V showed significant less degradation after 1000 cycles for ALS based Pt/MWCNTs.
ContributorsLiu, Xuan (Author) / Madakannan, Arunachalanadar (Thesis advisor) / Munukutla, Lakshmi (Committee member) / Tamizhmani, Govindasamy (Committee member) / Arizona State University (Publisher)
Created2011
Description
Infant mortality rate of field deployed photovoltaic (PV) modules may be expected to be higher than that estimated by standard qualification tests. The reason for increased failure rates may be attributed to the high system voltages. High voltages (HV) in grid connected modules induce additional stress factors that cause new degradation mechanisms. These new degradation mechanisms are not recognized by qualification stress tests. To study and model the effect of high system voltages, recently, potential induced degradation (PID) test method has been introduced. Using PID studies, it has been reported that high voltage failure rates are essentially due to increased leakage currents from active semiconducting layer to the grounded module frame, through encapsulant and/or glass. This project involved designing and commissioning of a new PID test bed at Photovoltaic Reliability Laboratory (PRL) of Arizona State University (ASU) to study the mechanisms of HV induced degradation. In this study, PID stress tests have been performed on accelerated stress modules, in addition to fresh modules of crystalline silicon technology. Accelerated stressing includes thermal cycling (TC200 cycles) and damp heat (1000 hours) tests as per IEC 61215. Failure rates in field deployed modules that are exposed to long term weather conditions are better simulated by conducting HV tests on prior accelerated stress tested modules. The PID testing was performed in 3 phases on a set of 5 mono crystalline silicon modules. In Phase-I of PID test, a positive bias of +600 V was applied, between shorted leads and frame of each module, on 3 modules with conducting carbon coating on glass superstrate. The 3 module set was comprised of: 1 fresh control, TC200 and DH1000. The PID test was conducted in an environmental chamber by stressing the modules at 85°C, for 35 hours with an intermittent evaluation for Arrhenius effects. In the Phase-II, a negative bias of -600 V was applied on a set of 3 modules in the chamber as defined above. The 3 module set in phase-II was comprised of: control module from phase-I, TC200 and DH1000. In the Phase-III, the same set of 3 modules which were used in the phase-II again subjected to +600 V bias to observe the recovery of lost power during the Phase-II. Electrical performance, infrared (IR) and electroluminescence (EL) were done prior and post PID testing. It was observed that high voltage positive bias in the first phase resulted in little
o power loss, high voltage negative bias in the second phase caused significant power loss and the high voltage positive bias in the third phase resulted in major recovery of lost power.
o power loss, high voltage negative bias in the second phase caused significant power loss and the high voltage positive bias in the third phase resulted in major recovery of lost power.
ContributorsGoranti, Sandhya (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Rogers, Bradley (Committee member) / Macia, Narciso (Committee member) / Arizona State University (Publisher)
Created2011
Description
Building Applied Photovoltaics (BAPV) form an essential part of today's solar economy. This thesis is an effort to compare and understand the effect of fan cooling on the temperature of rooftop photovoltaic (PV) modules by comparing two side-by-side arrays (test array and control array) under identical ambient conditions of irradiance, air temperature, wind speed and wind direction. The lower operating temperature of PV modules due to fan operation mitigates array non uniformity and improves on performance. A crystalline silicon (c-Si) PV module has a light to electrical conversion efficiency of 14-20%. So on a cool sunny day with incident solar irradiance of 1000 W/m2, a PV module with 15% efficiency, will produce about only 150 watts. The rest of the energy is primarily lost in the form of heat. Heat extraction methods for BAPV systems may become increasingly higher in demand as the hot stagnant air underneath the array can be extracted to improve the array efficiency and the extracted low-temperature heat can also be used for residential space heating and water heating. Poly c-Si modules experience a negative temperature coefficient of power at about -0.5% /o C. A typical poly c-Si module would experience power loss due to elevation in temperature, which may be in the range of 25 to 30% for desert conditions such as that of Mesa, Arizona. This thesis investigates the effect of fan cooling on the previously developed thermal models at Arizona State University and on the performance of PV modules/arrays. Ambient conditions are continuously monitored and collected to calculate module temperature using the thermal model and to compare with actually measured temperature of individual modules. Including baseline analysis, the thesis has also looked into the effect of fan on the test array in three stages of 14 continuous days each. Multiple Thermal models are developed in order to identify the effect of fan cooling on performance and temperature uniformity. Although the fan did not prove to have much significant cooling effect on the system, but when combined with wind blocks it helped improve the thermal mismatch both under low and high wind speed conditions.
ContributorsChatterjee, Saurabh (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Rogers, Bradley (Committee member) / Macia, Narciso (Committee member) / Arizona State University (Publisher)
Created2011
Description
Photovoltaic (PV) modules undergo performance degradation depending on climatic conditions, applications, and system configurations. The performance degradation prediction of PV modules is primarily based on Accelerated Life Testing (ALT) procedures. In order to further strengthen the ALT process, additional investigation of the power degradation of field aged PV modules in various configurations is required. A detailed investigation of 1,900 field aged (12-18 years) PV modules deployed in a power plant application was conducted for this study. Analysis was based on the current-voltage (I-V) measurement of all the 1,900 modules individually. I-V curve data of individual modules formed the basis for calculating the performance degradation of the modules. The percentage performance degradation and rates of degradation were compared to an earlier study done at the same plant. The current research was primarily focused on identifying the extent of potential induced degradation (PID) of individual modules with reference to the negative ground potential. To investigate this, the arrangement and connection of the individual modules/strings was examined in detail. The study also examined the extent of underperformance of every series string due to performance mismatch of individual modules in that string. The power loss due to individual module degradation and module mismatch at string level was then compared to the rated value.
ContributorsJaspreet Singh (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Srinivasan, Devarajan (Committee member) / Rogers, Bradley (Committee member) / Arizona State University (Publisher)
Created2011
Description
The theme for this work is the development of fast numerical algorithms for sparse optimization as well as their applications in medical imaging and source localization using sensor array processing. Due to the recently proposed theory of Compressive Sensing (CS), the $\ell_1$ minimization problem attracts more attention for its ability to exploit sparsity. Traditional interior point methods encounter difficulties in computation for solving the CS applications. In the first part of this work, a fast algorithm based on the augmented Lagrangian method for solving the large-scale TV-$\ell_1$ regularized inverse problem is proposed. Specifically, by taking advantage of the separable structure, the original problem can be approximated via the sum of a series of simple functions with closed form solutions. A preconditioner for solving the block Toeplitz with Toeplitz block (BTTB) linear system is proposed to accelerate the computation. An in-depth discussion on the rate of convergence and the optimal parameter selection criteria is given. Numerical experiments are used to test the performance and the robustness of the proposed algorithm to a wide range of parameter values. Applications of the algorithm in magnetic resonance (MR) imaging and a comparison with other existing methods are included. The second part of this work is the application of the TV-$\ell_1$ model in source localization using sensor arrays. The array output is reformulated into a sparse waveform via an over-complete basis and study the $\ell_p$-norm properties in detecting the sparsity. An algorithm is proposed for minimizing a non-convex problem. According to the results of numerical experiments, the proposed algorithm with the aid of the $\ell_p$-norm can resolve closely distributed sources with higher accuracy than other existing methods.
ContributorsShen, Wei (Author) / Mittlemann, Hans D (Thesis advisor) / Renaut, Rosemary A. (Committee member) / Jackiewicz, Zdzislaw (Committee member) / Gelb, Anne (Committee member) / Ringhofer, Christian (Committee member) / Arizona State University (Publisher)
Created2011
Description
This dissertation transforms a set of system complexity reduction problems to feature selection problems. Three systems are considered: classification based on association rules, network structure learning, and time series classification. Furthermore, two variable importance measures are proposed to reduce the feature selection bias in tree models. Associative classifiers can achieve high accuracy, but the combination of many rules is difficult to interpret. Rule condition subset selection (RCSS) methods for associative classification are considered. RCSS aims to prune the rule conditions into a subset via feature selection. The subset then can be summarized into rule-based classifiers. Experiments show that classifiers after RCSS can substantially improve the classification interpretability without loss of accuracy. An ensemble feature selection method is proposed to learn Markov blankets for either discrete or continuous networks (without linear, Gaussian assumptions). The method is compared to a Bayesian local structure learning algorithm and to alternative feature selection methods in the causal structure learning problem. Feature selection is also used to enhance the interpretability of time series classification. Existing time series classification algorithms (such as nearest-neighbor with dynamic time warping measures) are accurate but difficult to interpret. This research leverages the time-ordering of the data to extract features, and generates an effective and efficient classifier referred to as a time series forest (TSF). The computational complexity of TSF is only linear in the length of time series, and interpretable features can be extracted. These features can be further reduced, and summarized for even better interpretability. Lastly, two variable importance measures are proposed to reduce the feature selection bias in tree-based ensemble models. It is well known that bias can occur when predictor attributes have different numbers of values. Two methods are proposed to solve the bias problem. One uses an out-of-bag sampling method called OOBForest, and the other, based on the new concept of a partial permutation test, is called a pForest. Experimental results show the existing methods are not always reliable for multi-valued predictors, while the proposed methods have advantages.
ContributorsDeng, Houtao (Author) / Runger, George C. (Thesis advisor) / Lohr, Sharon L (Committee member) / Pan, Rong (Committee member) / Zhang, Muhong (Committee member) / Arizona State University (Publisher)
Created2011
Description
Hydropower generation is one of the clean renewable energies which has received great attention in the power industry. Hydropower has been the leading source of renewable energy. It provides more than 86% of all electricity generated by renewable sources worldwide. Generally, the life span of a hydropower plant is considered as 30 to 50 years. Power plants over 30 years old usually conduct a feasibility study of rehabilitation on their entire facilities including infrastructure. By age 35, the forced outage rate increases by 10 percentage points compared to the previous year. Much longer outages occur in power plants older than 20 years. Consequently, the forced outage rate increases exponentially due to these longer outages. Although these long forced outages are not frequent, their impact is immense. If reasonable timing of rehabilitation is missed, an abrupt long-term outage could occur and additional unnecessary repairs and inefficiencies would follow. On the contrary, too early replacement might cause the waste of revenue. The hydropower plants of Korea Water Resources Corporation (hereafter K-water) are utilized for this study. Twenty-four K-water generators comprise the population for quantifying the reliability of each equipment. A facility in a hydropower plant is a repairable system because most failures can be fixed without replacing the entire facility. The fault data of each power plant are collected, within which only forced outage faults are considered as raw data for reliability analyses. The mean cumulative repair functions (MCF) of each facility are determined with the failure data tables, using Nelson's graph method. The power law model, a popular model for a repairable system, can also be obtained to represent representative equipment and system availability. The criterion-based analysis of HydroAmp is used to provide more accurate reliability of each power plant. Two case studies are presented to enhance the understanding of the availability of each power plant and represent economic evaluations for modernization. Also, equipment in a hydropower plant is categorized into two groups based on their reliability for determining modernization timing and their suitable replacement periods are obtained using simulation.
ContributorsKwon, Ogeuk (Author) / Holbert, Keith E. (Thesis advisor) / Heydt, Gerald T (Committee member) / Pan, Rong (Committee member) / Arizona State University (Publisher)
Created2011
Description
Photovoltaic (PV) systems are one of the next generation's renewable energy sources for our world energy demand. PV modules are highly reliable. However, in polluted environments, over time, they will collect grime and dust. There are also limited field data studies about soiling losses on PV modules. The study showed how important it is to investigate the effect of tilt angle on soiling. The study includes two sets of mini-modules. Each set has 9 PV modules tilted at 0, 5, 10, 15, 20, 23, 30, 33 and 40°. The first set called "Cleaned" was cleaned every other day. The second set called "Soiled" was never cleaned after the first day. The short circuit current, a measure of irradiance, and module temperature was monitored and recorded every two minutes over three months (January-March 2011). The data were analyzed to investigate the effect of tilt angle on daily and monthly soiling, and hence transmitted solar insolation and energy production by PV modules. The study shows that during the period of January through March 2011 there was an average loss due to soiling of approximately 2.02% for 0° tilt angle. Modules at tilt anlges 23° and 33° also have some insolation losses but do not come close to the module at 0° tilt angle. Tilt anlge 23° has approximately 1.05% monthly insolation loss, and 33° tilt angle has an insolation loss of approximately 0.96%. The soiling effect is present at any tilt angle, but the magnitude is evident: the flatter the solar module is placed the more energy it will lose.
ContributorsCano Valero, José (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Madakannan, Arunachalanadar (Committee member) / Macia, Narciso (Committee member) / Arizona State University (Publisher)
Created2011
Description
The photovoltaic (PV) modules are primarily characterized for their performance with respect to incident irradiance and operating temperature. This work deals with data collection and automation of data processing for the performance and thermal characterizations of PV modules. This is a two-part thesis: The primary part (part-1) deals with the software automation to generate performance matrix as per IEC 61853-1 standard using MPPT (maximum power point tracking) data at the module or system level; the secondary part (part-2) deals with the software automation to predict temperature of rooftop PV modules using the thermal model coefficients generated in the previous studies of the Photovoltaic Reliability Laboratory (PRL). Part 1: The IEC 61853-1 standard published in January 2011 specifies the generation of a target performance matrix of photovoltaic (PV) modules at various temperatures and irradiance levels. In a conventional method, this target matrix is generated using all the data points of several measured I-V curves and the translation procedures defined in IEC 60891 standard. In the proposed method, the target matrix is generated using only three commonly field measured parameters: Module temperature, Incident irradiance and MPPT (Maximum Peak Power Tracking) value. These parameters are loaded into the programmed Excel file and with a click of a button, IEC 61853-1 specified Pmppt matrix is displayed on the screen in about thirty seconds. Part 2: In a previous study at PRL, an extensive thermal model to predict operating temperature of rooftop PV modules was developed with a large number of empirical monthly coefficients for ambient temperature, irradiance and wind speed. Considering that there is large number of coefficients for each air gap of rooftop modules, it became necessary to automate the entire data processing to predict the temperature of rooftop PV modules at different air gaps. This part of the work was dedicated to automatically predict the temperature of rooftop modules at different air gaps for any month in a year just using only four input parameters: Month, Irradiance, Ambient temperature and Wind speed.
ContributorsKoka, Kartheek (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Rogers, Bradley (Committee member) / Macia, Narciso F. (Committee member) / Arizona State University (Publisher)
Created2011