Generating weighted fuzzy rules from relational database systems for estimating values using genetic algorithms

In recent years, some methods have been proposed to estimate values in relational database systems. However, the estimated accuracy of the existing methods are not good enough. In this paper, we present a new method to generate weighted fuzzy rules from relational database systems for estimating values using genetic algorithms (GAs), where the attributes appearing in the antecedent part of generated fuzzy rules have different weights. After a predefined number of evolutions of the GA, the best chromosome contains the optimal weights of the attributes, and they can be translated into a set of rules to be used for estimating values. The proposed method can get a higher average estimated accuracy rate than the methods we presented in two previous papers.     

Ceramers: Hybrid materials incorporating polymeric/oligomeric species with inorganic glasses by a sol-gel process

A sol-gel process has been successfully utilized to produce hybrid materials incorporating polymeric/oligomeric components of polydimethyl siloxane(PDMS) with silicon glasses. All the samples made were transparent and flexible. Dynamic mechanical studies indicate that a portion of the siloxane species is phase-separated while the remainder is well dispersed. The effect of acid content were proven to be significant on the dispersion of the siloxane components and on the structure and properties of final products.     

On-line learning delivery decision support system for highlyproduct mixed semiconductor foundry

A production learning system (PLS) based on the tool model was constructed as a decision support and real-time information update system to forecast the cycle time. A tool model includes a waiting model and a processing model. Each of the waiting and processing models uses a backpropagation neural network to establish the relationship between the input and output (time) of the model. Hence, cycle time estimation, tool group move and confirm line item performance (CLIP) value can be obtained based on the memory stored in the neural network. The result shows that the forecasting ability of the PLS has an error rate below 8% on average     

Efficient design/simulation algorithms for simultaneous switching noise in leadframe packages

 Two new modeling and simulation approaches for Simultaneous Switching Noise (SSN) are described and compared to “brute force” simulation by SPICE. Both simulation accuracy and simulation run-time are considered. The two new approaches are: 1) the “effective inductance” method, in which an approximate, very efficient method of extracting an SSN L _eff  is utilized; and 2) the “macromodel” method, in which the complex inductance network responsible for SSN is represented by only a few dominant poles in the frequency domain and the time domain response is obtained by an efficient convolution algorithm. Both approaches are shown to be accurate and fast, but only the effective inductance algorithm is robust in numerical convergence. Received: 19 March 1997 / Accepted: 25 March 1997     

The dominance of the K I-field for a crack under stress wave loading

The extent of the dominant singular field is investigated for a finite crack under stress wave loading. Using a boundary integral equation method the complete solution as well as the near field solution is determined. A comparison of the two fields indicates that the singular field dominates within a small domain at the crack tip. The size of the dominance region in the dynamic case may be very different from that in the static case.     

A steering procedure for the fast vertex fitting method

The fast vertex fitting method [P. Billoir and S. Qian, Nucl. Instr. and Meth. A 311 (1992) 139] has been enhanced by a steering procedure which performs the track selection to insure the vertex fitting quality. Meanwhile, a sign error in a formula in the above publication has been corrected.     

Comparisons of PWM and one-cycle control for power amplifier with multilevel converter

Based on the concepts of cascade multilevel converters and one-cycle technique, comparisons are conducted to characterize the distortion of the pulsewidth modulation (PWM) and one-cycle control methods that were applied to converters. Simulation results for the different control schemes are obtained in PSIM software initially to see the effect of one-cycle control different from that of PWM control. Through the comparisons, the advantages and disadvantages are identified for each method. The one-cycle scheme is better than PWM control in reducing undesirable harmonics and tracing dynamic waveforms. Simulation and experimental results are also provided to verify the conclusions.     

Mechanism analysis of gate-induced drain leakage in off-state n-MOSFET

An analytical expression for both band-to-band and band-trap-band indirect tunnelings is used to study the gate-induced drain leakage (GIDL) current of MOSFETs measured before and after hot-carrier stress. The voltage and temperature dependence of GIDL are characterized. Both results show that interface traps situated near the midgap participate in the conduction of GIDL, and band-trap-band indirect tunneling could be the major mechanism. This is further supported by the fact that the percentage increase in GIDL induced by hot-carrier stress is about the same as the corresponding increase in interface-trap density. On the other hand, under low-field conditions, trap-assisted Poole–Frenkle emission dominates over tunneling for temperatures even well below room temperature.