An Optimization Procedure in a Sirocco Fan Using Experimental Data

1997 ◽  
Author(s):  
H. S. Min ◽  
H. S. Rew ◽  
C. O. Ahn
Keyword(s):  
Neural Network ◽  
Taguchi Method ◽  
Flow Rate ◽  
Noise Level ◽  
Optimization Procedure ◽  
Design Variables ◽  
The Neural Network ◽  
Sirocco Fan ◽  
Important Design ◽  
Good Agreement

Abstract The flow rate and the specific noise level of 18 sirocco fans were measured and these data were analyzed by the Taguchi method and the neural network. The optimal design value obtained by the neural network generally showed good agreement with that by the Taguchi method. The effects of eight important design variables on the flow rate and the specific noise level were discussed.

Neural Networks for Tubular Joint Optimization in Offshore Jacket Structures

2016 ◽  
Author(s):  
Niels Hørbye Christiansen ◽  
Benny Korsholm Tang
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Design Space ◽  
Joint Optimization ◽  
Joint Models ◽  
Tubular Joints ◽  
Design Variables ◽  
The Neural Network ◽  
Jacket Structures ◽  
Tubular Joint

The use of jacket structures to support offshore installations has for a long time been a popular choice in places with appropriate water depths. In recent years the use of jacket structures as offshore wind turbine foundations has also attracted increasing attention and is becoming a feasible alternative to traditional monopile foundations. One of the key challenges in jacket design is optimizing tubular joints in terms of fatigue resistance. As it is not practically possible to include detailed FEM joint models in global jacket models designers are forced to look for alternative methods to obtain realistic joint representations. This is done by calculating influence factors (INF) and stress concentration factors (SCF) to be applied to simplified models of relevant tubular joints in global models in order to achieve a realistic force flow in the structure. One simple and widely used method is to apply parametric formulas, e.g. those suggested by Efthymiou. However, these approximating formulas have a fairly limited validity range. Therefore, on complex joint the most reliable way to determine INF’s is by setting up refined FEM models of relevant joints and then subsequently using the calculated factors in the global model. This strategy is computationally demanding and hence, very time consuming, as a new detailed FEM analysis of the tubular joint must be conducted for each step in the optimization process. The present paper demonstrates how this time consuming procedure can be avoided by use of artificial neural networks (ANN) trained to estimate INF’s on tubular joints. The neural network is trained on a pre-generated library of detailed FEM joint models and is then able to predict INF’s on joints that are not part of the library — and thereby providing a significant reduction in calculation time during the jacket/joint optimization process. The analysis is conducted on a typical joint on a three legged jacket structure. The joint is located on a jacket leg and has two incoming braces. Such a joint has a finite number of free design variables, e.g. chord diameter/thickness, brace diameter/thickness, brace angle, gap etc. Each of these free variables can be considered as a dimension in the joint design space. Having a sufficient number of FEM joint models in the library the neural network can be trained to recognize and predict underlying patterns in this design space. The method is demonstrated on a limited number of design variables but should easily be extended to cover all variables as the joint library is expanded to include all dimensions.

Designing and Harnessing the Metastable States of a Modular Metastructure for Programmable Mechanical Properties Adaptation

2015 ◽  
Vol 138 (2) ◽  
Author(s):  
R. L. Harne ◽  
Z. Wu ◽  
K. W. Wang
Keyword(s):  
Mechanical Properties ◽  
Metastable States ◽  
System Level ◽  
Macroscopic System ◽  
Actin Networks ◽  
Module Design ◽  
Design Variables ◽  
Module Development ◽  
Important Design ◽  
Good Agreement

Recent studies on periodic metamaterial systems have shown that remarkable properties adaptivity and versatility are often the products of exploiting internal, coexisting metastable states. Motivated by this concept, this research develops and explores a local-global design framework wherein macroscopic system-level properties are sought according to a strategic periodic constituent composition and assembly. To this end and taking inspiration from recent insights in studies of multiphase composite materials and cytoskeletal actin networks, this study develops adaptable metastable modules that are assembled into modular metastructures, such that the latter are invested with synergistic features due to the strategic module development and integration. Using this approach, it is seen that modularity creates an accessible pathway to exploit metastable states for programmable metastructure adaptivity, including a near-continuous variation of mechanical properties or stable topologies and adjustable hysteresis. A model is developed to understand the source of the synergistic characteristics, and theoretical findings are found to be in good agreement with experimental results. Important design-based questions are raised regarding the modular metastructure concept, and a genetic algorithm (GA) routine is developed to elucidate the sensitivities of the properties variation with respect to the statistics amongst assembled module design variables. To obtain target multifunctionality and adaptivity, the routine discovers that particular degrees and types of modular heterogeneity are required. Future realizations of modular metastructures are discussed to illustrate the extensibility of the design concept and broad application base.

Prediction of Detonation Pressure of Aluminized Explosive by Artificial Neural Network

2013 ◽  
Vol 641-642 ◽  
pp. 460-463
Author(s):  
Yong Gang Liu ◽  
Xin Tian ◽  
Yue Qiang Jiang ◽  
Gong Bing Li ◽  
Yi Zhou Li
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Optimal Condition ◽  
Detonation Pressure ◽  
Ann Model ◽  
Training Set ◽  
The Neural Network ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Good Agreement

In this study, a three-layer artificial neural network(ANN) model was constructed to predict the detonation pressure of aluminized explosive. Elemental composition and loading density were employed as input descriptors and detonation pressure was used as output. The dataset of 41 aluminized explosives was randomly divided into a training set (30) and a prediction set (11). After optimized by adjusting various parameters, the optimal condition of the neural network was obtained. Simulated with the final optimum neural network [6–9–1], calculated detonation pressures show good agreement with experimental results. It is shown here that ANN is able to produce accurate predictions of the detonation pressure of aluminized explosive.

scholarly journals The new Passive microwave Neural network Precipitation Retrieval (PNPR) algorithm for the cross-track scanning ATMS radiometer: description and verification study over Europe and Africa using GPM and TRMM spaceborne radars

2016 ◽  
Author(s):  
Paolo Sanò ◽  
Giulia Panegrossi ◽  
Daniele Casella ◽  
Anna Cinzia Marra ◽  
Francesco Di Paola ◽  
...  
Keyword(s):  
Neural Network ◽  
Mean Squared Error ◽  
Passive Microwave ◽  
Precipitation Radar ◽  
Water Vapor Absorption ◽  
The Neural Network ◽  
Trmm Pr ◽  
The Cross ◽  
Cross Track ◽  
Good Agreement

Abstract. The objective of this paper is to describe the development and evaluate the performance of a totally new version of the Passive microwave Neural network Precipitation Retrieval (PNPR v2), an algorithm based on a neural network approach, designed to retrieve the instantaneous surface precipitation rate using the cross-track ATMS radiometer measurements. This algorithm, developed within the EUMETSAT H-SAF program, represents an evolution of the previous version (PNPR v1), developed for AMSU/MHS radiometers (and used and distributed operationally within H-SAF), with improvements aimed at exploiting the new precipitation sensing capabilities of ATMS with respect to AMSU/MHS. In the design of the neural network the new ATMS channels compared to AMSU/MHS, and their combinations, including the brightness temperature differences in the water vapor absorption band, around 183 GHz, are considered . The algorithm is based on a single neural network, for all types of surface background, trained using a large database based on 94 cloud-resolving model simulations over the European and the African areas. The performance of PNPR v2 has been evaluated through an intercomparison of the instantaneous precipitation estimates with co-located estimates from the TRMM Precipitation Radar (TRMM-PR) and from the GPM Core Observatory Ku-band Precipitation Radar (GPM-KuPR). In the comparison with TRMM-PR, over the African area, the statistical analysis was carried out for a two-year (2013-2014) dataset of coincident observations, over a regular grid at 0.5° × 0.5° resolution. The results have shown a good agreement between PNPR v2 and TRMM-PR for the different surface types. The correlation coefficient (CC) was equal to 0.69 over ocean and 0.71 over vegetated land (lower values were obtained over arid land and coast), and the root mean squared error (RMSE) was equal to 1.30 mm h−1 over ocean and 1.11 mm h−1 over vegetated land. The results showed a slight tendency to underestimate moderate to high precipitation, mostly over land, and overestimate moderate to light precipitation over ocean. Similar results were obtained for the comparison with GPM-KuPR over the European area (15 months, from March 2014 to May 2015 of coincident overpasses) with slightly lower CC (0.59 over vegetated land and 0.57 over ocean) and RMSE (0.82 mm h−1 over vegetated land and 0.71 mm h−1 over ocean), confirming a good agreement also between PNPR v2 and GPM-KuPR. The performance of PNPR v2 over the African area was also compared to that of PNPR v1. PNPR v2 has higher R over the different surfaces, with general better estimate of low precipitation, mostly over ocean, thanks to improvements in the design of the neural network and also to the improved capabilities of ATMS compared to AMSU/MHS. Both versions of PNPR algorithm have shown a general consistency with the TRMM-PR.

scholarly journals A Promising Wavelet Decomposition –NNARX Model to Predict Flood

2020 ◽  
Vol 11 (2) ◽  
pp. 89-99
Author(s):  
Mohd Azrol Syafiee Anuar ◽  
Ribhan Zafira Abdul Rahman ◽  
Azura Che Soh ◽  
Samsul Bahari Mohd Noor ◽  
Zed Diyana Zulkafli
Keyword(s):  
Neural Network ◽  
Flow Rate ◽  
Wavelet Decomposition ◽  
Performance Testing ◽  
Original Data ◽  
Flood Prediction ◽  
The Neural Network ◽  
Major Disaster ◽  
A New Technique

Flood is a major disaster that happens around the world. It has caused many casualties and massive destruction of property. Estimating the chance of a flood occurring depends on several factors, such as rainfall, the structure and the flow rate of the river. This research used the neural network autoregressive exogenous input (NNARX) model to predict floods. One of the research challenges was to develop accurate models and improve the forecasting model. This research aimed to improve the performance of the neural network model for flood prediction. A new technique was proposed for modelling nonlinear data of flood forecasting using the wavelet decomposition-NNARX approach. This paper discusses the process of identifying the parameters involved to make a forecast as the rainfall value requires the flow rate of the river and its water level. The original data were processed by wavelet decomposition and filtered to generate a new set of data for the NNARX prediction model where the process can be compared. This research compared the performance of the wavelet and the non-wavelet NNARX model. Experimental results showed that the proposed approach had better performance testing results in relation to its counterpart in terms of hourly forecast, with the mean square error (MSE) of 2.0491e-4 m2 compared to 6.1642e-4 m2, respectively. The proposed approach was also studied for long-term forecast up to 5 years, where the obtained MSE was higher, i.e., 0.0016 m2.

scholarly journals Polyp and Surgical Instrument Segmentation with Double Encoder-Decoder Networks

2021 ◽  
Vol 1 (1) ◽  
pp. 5-7
Author(s):  
Adrian Galdran
Keyword(s):  
Neural Network ◽  
Optimization Procedure ◽  
Experimental Results ◽  
Surgical Instruments ◽  
Surgical Instrument ◽  
Post Processing ◽  
Endoscopic Images ◽  
Good Agreement

This paper describes a solution for the MedAI competition, in which participants were required to segment both polyps and surgical instruments from endoscopic images. Our approach relies on a double encoder-decoder neural network which we have previously applied for polyp segmentation, but with a series of enhancements: a more powerful encoder architecture, an improved optimization procedure, and the post-processing of segmentations, based on tempered model ensembling. Experimental results show that our method produces segmentations that show a good agreement with manual delineations provided by medical experts.

Neural network potential for Zr-Rh system by machine learning

2021 ◽  
Author(s):  
Kun Xie ◽  
Chong Qiao ◽  
Hong Shen ◽  
Riyi Yang ◽  
Ming Xu ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Mechanical Property ◽  
Molecular Dynamics ◽  
Molecular Dynamics Simulations ◽  
Structural Features ◽  
Ab Initio Molecular Dynamics ◽  
The Neural Network ◽  
Dynamics Simulations ◽  
Good Agreement

Abstract Zr-Rh metallic glass has enabled its many applications in vehicle parts, sports equipment and so on due to its outstanding performance in mechanical property, but the knowledge of the microstructure determining the superb mechanical property remains yet insufficient. Here, we develop a deep neural network potential of Zr-Rh system by using machine learning, which breaks the dilemma between the accuracy and efficiency in molecular dynamics simulations, and greatly improves the simulation scale in both space and time. The results show that the structural features obtained from the neural network method are in good agreement with the cases in ab initio molecular dynamics simulations. Furthermore, we build a large model of 5400 atoms to explore the influences of simulated size and cooling rate on the melt-quenching process of Zr77Rh23. Our study lays a foundation for exploring the complex structures in amorphous Zr77Rh23, which is of great significance for the design and practical application.

scholarly journals A Neural Network-Based Four Phases Interleaved Boost Converter for Fuel Cell System Applications

Energies ◽  
2018 ◽  
Vol 11 (12) ◽  
pp. 3423 ◽  
Author(s):  
El Barhoumi ◽  
Ikram Ben Belgacem ◽  
Abla Khiareddine ◽  
Manaf Zghaibeh ◽  
Iskander Tlili
Keyword(s):  
Neural Network ◽  
Fuel Cell ◽  
Flow Rate ◽  
Fuel Flow Rate ◽  
Supply Pressure ◽  
Fuel Flow ◽  
Neural Network Controller ◽  
The Neural Network ◽  
Network Controller ◽  
Interleaved Boost Converter

This paper presents a simple strategy for controlling an interleaved boost converter that is used to reduce the current fluctuations in proton exchange membrane fuel cells, with high impact on the fuel cell lifetime. To keep the output voltage at the desired reference value under the strong fluctuations of the fuel flow rate, fuel supply pressure, and temperature, a neural network controller is developed and implemented using Matlab-Simulink (R2012b, MathWorks limited, London, UK). The advantage of this controller resides in its simplicity, where limited number of tests are carried out using Matlab-Simulink to construct it. To investigate the robustness of the proposed converter and the neural network controller, strong variations of the fuel flow rate, fuel supply pressure, temperature and air supply pressure are applied to both the fuel cell and the neural network controller of the converter. The simulation results show the effectiveness and the robustness of the both the proposed controller and converter to control the load voltage and minimize the current and voltage ripples. As a result of that, fuel cell current oscillations are considerably reduced on the one hand, while on the other hand, the load voltage is stabilized during transient variations of the fuel cell inputs.

Prediction of Detonation Velocity of Aluminized Explosive by Artificial Neural Network

2013 ◽  
Vol 790 ◽  
pp. 673-676
Author(s):  
Yue Qiang Jiang ◽  
Yong Gang Liu ◽  
Xin Tian ◽  
Gong Bing Li
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Optimal Condition ◽  
Detonation Velocity ◽  
Ann Model ◽  
Training Set ◽  
The Neural Network ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Good Agreement

In this study, a three-layer artificial neural network (ANN) model was constructed to predict the detonation velocity of aluminized explosive. Elemental composition and loading density were employed as input descriptors and detonation velocity was used as output. The dataset of 61 aluminized explosives was randomly divided into a training set (49) and a prediction set (12). After optimized by adjusting various parameters, the optimal condition of the neural network was obtained. Simulated with the final optimum neural network [812, calculated detonation velocity show good agreement with experimental results. It is shown that ANN is able to produce accurate predictions of the detonation velocity of aluminized explosive.

scholarly journals Dynamic Estimation of Saturation Flow Rate at Information-Rich Signalized Intersections

Information ◽  
2020 ◽  
Vol 11 (4) ◽  
pp. 178 ◽  
Author(s):  
Yi Wang ◽  
Jian Rong ◽  
Chenjing Zhou ◽  
Yacong Gao
Keyword(s):  
Neural Network ◽  
Flow Rate ◽  
Network Models ◽  
Signalized Intersections ◽  
Neural Network Models ◽  
The Road ◽  
The Neural Network ◽  
Dynamic Estimation ◽  
Saturation Flow Rate ◽  
Saturation Flow

Intersections are the bottlenecks of the road network. The capacity of signalized intersections restricts the operation of the road network. Dynamic estimation of capacity is necessary for signalized intersections refined management. With the development of technology, more and more detectors were installed near the intersection. It had been the information-rich environment, which provided support for dynamic estimation of capacity. A dynamic estimation method for a saturation flow rate based on a neural network was developed. It would grasp the dynamic change of saturation flow rates and influencing factors. The measure data at three scenarios (through lanes, shared right-turn and through lanes, shared left-turn and through lanes) of signalized intersections in Beijing were taken as examples to validate the proposed method. Firstly, the traffic flow characteristics of the three scenarios and factors affecting the saturation flow rate were analyzed. Secondly, neural network models of the three scenarios were established. Then the hyperparameters of neural network models were determined. After training, the neural network structure and parameters were saved. Lastly, the test set data was validated by the training model. At the same time, the proposed method was compared with the Highway Capacity Manual (HCM) method and the statistical regression method. The results show that both regression models and neural network models have better accuracy than HCM models. In a simple scenario, the neural network models are not much different from the regression models. With the increase of complexity of scenarios, the advantages of neural network models are highlighted. In through-left lane and through-right lane scenarios, the estimated saturation flow rates used by the proposed method were 7.02%, 4.70%, respectively. In the complexity of traffic scenarios, the proposed method can estimate the saturation flow rate accurately and timely. The results could be used for signal control schemes optimizing and operation managing at signalized intersections subtly.

Sign in / Sign up

Export Citation Format

Share Document