On the use of IPA in performance optimization of continuous marked graphs: A case study

Author(s):  
Carla Seatzu ◽  
Yorai Wardi
2022 ◽  
Vol 15 (1) ◽  
pp. 1-32
Author(s):  
Lana Josipović ◽  
Shabnam Sheikhha ◽  
Andrea Guerrieri ◽  
Paolo Ienne ◽  
Jordi Cortadella

Commercial high-level synthesis tools typically produce statically scheduled circuits. Yet, effective C-to-circuit conversion of arbitrary software applications calls for dataflow circuits, as they can handle efficiently variable latencies (e.g., caches), unpredictable memory dependencies, and irregular control flow. Dataflow circuits exhibit an unconventional property: registers (usually referred to as “buffers”) can be placed anywhere in the circuit without changing its semantics, in strong contrast to what happens in traditional datapaths. Yet, although functionally irrelevant, this placement has a significant impact on the circuit’s timing and throughput. In this work, we show how to strategically place buffers into a dataflow circuit to optimize its performance. Our approach extracts a set of choice-free critical loops from arbitrary dataflow circuits and relies on the theory of marked graphs to optimize the buffer placement and sizing. Our performance optimization model supports important high-level synthesis features such as pipelined computational units, units with variable latency and throughput, and if-conversion. We demonstrate the performance benefits of our approach on a set of dataflow circuits obtained from imperative code.


2014 ◽  
Vol 29 ◽  
pp. 1113-1122 ◽  
Author(s):  
Vítor Schwambach ◽  
Sébastien Cleyet-Merle ◽  
Alain Issard ◽  
Stéphane Mancini

2017 ◽  
Vol 2017 ◽  
pp. 1-10 ◽  
Author(s):  
Ying Zhang ◽  
Daniel Jian Sun ◽  
Alexandra Kondyli

Operational performance optimization of signalized intersections is one of the most important tasks for traffic engineers and researchers. To compensate for the limitations of practical implementation, simulation software packages have been widely used to evaluate different optimization strategies and thus to improve the efficiency of the intersections as well as the entire network. However, for the existing optimization studies on signalized intersections, the relationships among various optimization measures and the combination of strategies have not been fully investigated. In this paper, uniform design experimentation was introduced to combine different optimization measures into strategies and achieve the minimum time cost in model construction. VISSIM software package was then calibrated and used to evaluate various optimization strategies and identify the one with the best measurement of performance, namely, control delay at the signalized intersection. By taking a representative congested intersection in Shanghai as a case study, the optimal strategy was identified to reduce the overall control delay by 27.3%, which further verified the modeling capability of the proposed method.


2013 ◽  
Vol 52 (02) ◽  
pp. 101-111 ◽  
Author(s):  
Kamran Jorshari ◽  
Brendan O'Hara ◽  
Ramsey Jones

Author(s):  
Stephen K. Storm ◽  
Richard F. Storm ◽  
Daniel S. Storm ◽  
Sammy Tuzenew ◽  
Adam McClellan

Pulverizer performance optimization is the first step to a successful combustion optimization program and the inter-relationships of the pulverizers must be considered when attempting to optimize combustion, overall unit performance, operability, reliability, and capacity. Pulverizer capacity seems to be an industry challenge while many units today are undergoing drastic fuel changes. Considering there seems to be a huge disconnect when correlating mill performance with such issues as fuel line distribution, heat rate, NOx and environmental control equipment performance, it is the intent of this technical paper to provide better understanding of how mechanical optimization & tuning of the pulverizers can yield overall improved plant performance. Low NOx firing and/or optimization of the burner belt combustion with a limited amount of furnace residence time is absolutely essential to optimizing plant performance. For example, when pulverizer performance is poor, it is also often related to not only high furnace exit gas temperatures, increased slagging and/or high LOI, but also degrading electrostatic precipitator (ESP) performance from the coarse particle ash. Furthermore, reliability of the boiler (ie. tube leaks, fouling, and slagging) can also be impacted negatively by secondary combustion and consequent super heater and re-heater tube metals overheating and/or wall wastage often occurs from non-optimized fuel distribution being delivered from the pulverizers. Whether the reason for improving mill performance is for the aforementioned items and/or perhaps simply to reduce power generation costs with improved fuels flexibility, the purpose of this case study is to review the basics of vertical spindle mill performance improvements. The data used to support this paper is from a compilation of actual field testing & tuning results. Furthermore, Storm Technologies, Inc. (STI) suggests the aforementioned steps as an effective approach to optimization.


2019 ◽  
Author(s):  
Ahmed Y. Abdelmaksoud ◽  
Hesham A. Hegazi ◽  
Mohamed S. El Morsi ◽  
Sayed M. Metwalli

Abstract The wide spread use of solar energy and photovoltaic solar cells attracts researchers to work hard for the objective of improving their performance to be a viable attractive alternative to fossil fuels. This work presents the optimization of the output power of a photovoltaic solar cell to harvest maximum power. The power is optimized based on PVP full tracking of the sun and the effect of the tilt, the azimuth, and the incidence angles on its performance. A meta-model is utilized as a tool in the optimization technique. A case study which develops 2nd, 3rd, and 4th order hyper-surface equations that depend on day and hour to get the optimum tilt angle and optimum azimuth angle for the city of Hurghada in Egypt. The power produced from the polynomial meta-model algorithm is higher by 25.5% over the fixed tilt and azimuth angles. The developed equations from the meta-model optimization technique can be used in tracking control systems to get optimum tilt and optimum azimuth angles at any hour and at any day. By comparing the metamodel and the lengthy performance optimization technique, the error is around 1.1%. Comparison has also been made between the cost of the fixed and full tracking photovoltaic systems.


2019 ◽  
Vol 231 ◽  
pp. 1526-1541 ◽  
Author(s):  
Shima Pashapour ◽  
Ali Bozorgi-Amiri ◽  
Ali Azadeh ◽  
Seyed Farid Ghaderi ◽  
Abbas Keramati

Sign in / Sign up

Export Citation Format

Share Document