A Modified Genetic Algorithm for the Selection of Decoupling Capacitors in PDN Design
26 Jul 2021-
TL;DR: In this article, a new genetic algorithm (GA) is proposed for the selection and placement of capacitors to meet a target impedance using as few capacitors as possible, which is centered around controlling the number of unused port locations in the GA population solutions, with the result of smoothing out the GA convergence and speeding up the convergence rate.
Abstract: Decoupling capacitors are used to provide adequate and stable power for integrated circuits in printed circuit boards (PCB). For complicated and large designs, it is difficult to select capacitors to meet voltage ripple limits while also minimizing cost because the search space is too large. In this work, a new genetic algorithm (GA) is proposed for the selection and placement of capacitors to meet a target impedance using as few capacitors as possible. The GA is centered around controlling the number of unused port locations in the GA population solutions, with the result of smoothing out the GA convergence and speeding up the convergence rate. A result comparison is made of the proposed GA against other algorithms and found the GA competitive if not better for the select test cases.
Citations
More filters
TL;DR: In this paper , five machine learning models, including decision tree, random forest, extreme gradient boosting (XGBoost), categorical boosting (CatBoost), and light gradient boosting machine (LightGBM), were used to forecast target impedance values.
Abstract: The transmission characteristics of the printed circuit board (PCB) ensure signal integrity and support the entire circuit system, with impedance matching being critical in the design of high-speed PCB circuits. Because the factors affecting impedance are closely related to the PCB production process, circuit designers and manufacturers must work together to adjust the target impedance to maintain signal integrity. Five machine learning models, including decision tree (DT), random forest (RF), extreme gradient boosting (XGBoost), categorical boosting (CatBoost), and light gradient boosting machine (LightGBM), were used to forecast target impedance values. Furthermore, the Optuna algorithm is used to determine forecasting model hyperparameters. This study applied tree-based machine learning techniques with Optuna to predict impedance. The results revealed that five tree-based machine learning models with Optuna can generate satisfying forecasting accuracy in terms of three measurements, including mean absolute percentage error (MAPE), root mean square error (RMSE), and coefficient of determination (R2). Meanwhile, the LightGBM model with Optuna outperformed the other models. In addition, by using Optuna to tune the parameters of machine learning models, the accuracy of impedance matching can be increased. Thus, the results of this study suggest that the tree-based machine learning techniques with Optuna are a viable and promising alternative for predicting impedance values for circuit analysis.
01 Aug 2022
TL;DR: In this article , an optimization routine is applied for the decoupling capacitor placement on power distribution networks to identify the limit beyond which the placement of additional decaps is no longer effective, thus leading to wasting layout area and components, and to a cost increase.
Abstract: An optimization routine is applied for the decoupling capacitor placement on Power Distribution Networks to identify the limit beyond which the placement of additional decaps is no longer effective, thus leading to wasting layout area and components, and to a cost increase. A specific test example from a real design is used together with the required target impedance and frequency band of interest for the PDN design. The effectiveness of the decap placement while selecting different layers of the stack-up, and while moving the upper limit of the PDN design band is analyzed. Such analysis leads to helpful insights based on the progression of the input impedance during the optimization process, and to develop useful guidelines for avoiding over-design of the PDN.
References
More filters
TL;DR: In this article, the authors introduce genetic algorithms (GA) as a complete entity, in which knowledge of this emerging technology can be integrated together to form the framework of a design tool for industrial engineers.
Abstract: This paper introduces genetic algorithms (GA) as a complete entity, in which knowledge of this emerging technology can be integrated together to form the framework of a design tool for industrial engineers. An attempt has also been made to explain "why" and "when" GA should be used as an optimization tool.
893 citations
13 Jul 2015
TL;DR: This study has reported the significant work conducted on various selection techniques and the comparison of selection techniques used in Genetic Algorithm.
Abstract: This paper compares various selection techniques used in Genetic Algorithm. Genetic algorithms are optimization search algorithms that maximize or minimizes given functions. Indentifying the appropriate selection technique is a critical step in genetic algorithm. The process of selection plays an important role in resolving premature convergence because it occurs due to lack of diversity in the population. Therefore selection of population in each generation is very important. In this study, we have reported the significant work conducted on various selection techniques and the comparison of selection techniques.
143 citations
TL;DR: The method allows for an easy comparison of different selection schemes and exact analytical results are derived for traditional generational selection, steady-state selection with varying generation gap, a simple model of Eshelman's CHC algorithm (1991), and (/spl mu/+/spl lambda/) evolution strategies.
Abstract: A method for calculating genetic drift in terms of changing population fitness variance is presented. The method allows for an easy comparison of different selection schemes and exact analytical results are derived for traditional generational selection, steady-state selection with varying generation gap, a simple model of Eshelman's CHC algorithm (1991), and (/spl mu/+/spl lambda/) evolution strategies. The effects of changing genetic drift on the convergence of a GA are demonstrated empirically.
127 citations
TL;DR: In this article, a genetic algorithm is used for the selection and placement of decoupling capacitors in a power distribution network (PDN) to reduce the effort expended by the complex task of capacitance placement.
Abstract: The impedance of the power distribution network (PDN) needs to be minimized in order to prevent unwanted voltage fluctuations at frequencies where current transients occur. To reduce PDN impedance, one can place decoupling capacitors that act as local current sources. However, selecting and placing the right capacitors at the right locations are problematic because of the complexity of modern package and board structures. In addition, decoupling capacitors are not effective at higher frequencies, requiring more complicated techniques such as embedded decoupling. This paper introduces a method of reducing the effort expended by the complex task of decoupling capacitor placement: a genetic algorithm that is customized for the selection and placement of decoupling capacitors. The core engine of this optimizing algorithm is a recently developed technique, the multilayer finite element method (MFEM), which solves for PDN impedances. This paper also highlights a method of incorporating vertical circuit elements into MFEM. Using several test cases, it proves the validity of the inclusion of vertical elements in MFEM and the effectiveness of the optimizer.
43 citations
15 Oct 2018
TL;DR: The proposed reinforcement learning-based optimal on-board decoupling capacitor (decap) design method has successfully provided 37 optimal decap designs with 4 decaps assigned each and satisfied the required target impedance while minimizing the number of assigned decaps.
Abstract: In this paper, for the first time, we propose a reinforcement learning-based optimal on-board decoupling capacitor (decap) design method. The proposed method can provide optimal decap designs for a given on-board power distribution network (PDN). An optimal decap design refers to the optimized combination of decaps at proper positions to satisfy a required target impedance. Moreover, a minimum number of decaps should be assigned for optimal decap designs. The proposed method is applied to the test on-board PDN and successfully provided 37 optimal decap designs with 4 decaps assigned each. Self impedance of PDN with the provided design satisfied the required target impedance while minimizing the number of assigned decaps.
22 citations