Inferential estimation of polypropylene melt index using stacked neural networks based on absolute error criteria

2010 ◽  
Author(s):  
Luyue Xia ◽  
Haitian Pan
Keyword(s):  
Neural Networks ◽  
Absolute Error ◽  
Melt Index ◽  
Polypropylene Melt

Estimation of Polypropylene Melt Index through the Application of Stacked Neural Networks

2011 ◽  
Vol 187 ◽  
pp. 411-415
Author(s):  
Lu Yue Xia ◽  
Hai Tian Pan ◽  
Meng Fei Zhou ◽  
Yi Jun Cai ◽  
Xiao Fang Sun
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Model ◽  
Neural Network Model ◽  
Model Performance ◽  
Measurement Interval ◽  
Melt Index ◽  
Polypropylene Melt ◽  
Single Neural Network

Melt index is the most important parameter in determining the polypropylene grade. Since the lack of proper on-line instruments, its measurement interval and delay are both very long. This makes the quality control quite difficult. A modeling approach based on stacked neural networks is proposed to estimation the polypropylene melt index. Single neural network model generalization capability can be significantly improved by using stacked neural networks model. Proper determination of the stacking weights is essential for good stacked neural networks model performance, so determination of appropriate weights for combining individual networks using the criteria about minimization of sum of absolute prediction error is proposed. Application to real industrial data demonstrates that the polypropylene melt index can be successfully estimated using stacked neural networks. The results obtained demonstrate significant improvements in model accuracy, as a result of using stacked neural networks model, compared to using single neural network model.

Modeling Methodology Based on Stacked Neural Networks for Inferential Prediction of Polypropylene Melt Index

2011 ◽  
Vol 55-57 ◽  
pp. 670-674
Author(s):  
Lu Yue Xia ◽  
Hai Tian Pan ◽  
Meng Fei Zhou ◽  
Yi Jun Cai ◽  
Xiao Fang Sun
Keyword(s):  
Neural Networks ◽  
Model Performance ◽  
Propylene Polymerization ◽  
Melt Index ◽  
Modeling Methodology ◽  
Data Points ◽  
Robust Least Squares ◽  
Polypropylene Melt ◽  
Characteristic Process

A modeling methodology based on stacked neural networks by combining several individual networks in parallel is proposed. Stacked neural network as an effective method for modeling of inherently complex and nonlinear systems especially a system with a limited number of experimental data points is chosen for yield prediction. Proper determination of the stacking weights is essential for good stacked neural networks model performance, so determination of appropriate weights for combining individual networks using robust least squares estimation is proposed. Inferential prediction of melt index as the most important characteristic process of polypropylene polymerization has been carried out. The application of the proposed modeling method based on stacked neural networks to the development of melt index soft sensor in an industrial propylene polymerization plant demonstrates its effectiveness.

Nanosecond Photodynamics Simulations of a Cis-Trans Isomerization Are Enabled by Machine Learning

2020 ◽  
Author(s):  
Jingbai Li ◽  
Patrick Reiser ◽  
André Eberhard ◽  
Pascal Friederich ◽  
Steven Lopez
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Excited State ◽  
Adaptive Sampling ◽  
Computational Cost ◽  
Ground Truth ◽  
Absolute Error ◽  
Photochemical Reactions ◽  
Computational Techniques ◽  
Full Potential

<p>Photochemical reactions are being increasingly used to construct complex molecular architectures with mild and straightforward reaction conditions. Computational techniques are increasingly important to understand the reactivities and chemoselectivities of photochemical isomerization reactions because they offer molecular bonding information along the excited-state(s) of photodynamics. These photodynamics simulations are resource-intensive and are typically limited to 1–10 picoseconds and 1,000 trajectories due to high computational cost. Most organic photochemical reactions have excited-state lifetimes exceeding 1 picosecond, which places them outside possible computational studies. Westermeyr <i>et al.</i> demonstrated that a machine learning approach could significantly lengthen photodynamics simulation times for a model system, methylenimmonium cation (CH<sub>2</sub>NH<sub>2</sub><sup>+</sup>).</p><p>We have developed a Python-based code, Python Rapid Artificial Intelligence <i>Ab Initio</i> Molecular Dynamics (PyRAI<sup>2</sup>MD), to accomplish the unprecedented 10 ns <i>cis-trans</i> photodynamics of <i>trans</i>-hexafluoro-2-butene (CF<sub>3</sub>–CH=CH–CF<sub>3</sub>) in 3.5 days. The same simulation would take approximately 58 years with ground-truth multiconfigurational dynamics. We proposed an innovative scheme combining Wigner sampling, geometrical interpolations, and short-time quantum chemical trajectories to effectively sample the initial data, facilitating the adaptive sampling to generate an informative and data-efficient training set with 6,232 data points. Our neural networks achieved chemical accuracy (mean absolute error of 0.032 eV). Our 4,814 trajectories reproduced the S<sub>1</sub> half-life (60.5 fs), the photochemical product ratio (<i>trans</i>: <i>cis</i> = 2.3: 1), and autonomously discovered a pathway towards a carbene. The neural networks have also shown the capability of generalizing the full potential energy surface with chemically incomplete data (<i>trans</i> → <i>cis</i> but not <i>cis</i> → <i>trans</i> pathways) that may offer future automated photochemical reaction discoveries.</p>

scholarly journals Augmenting Around-Device Interaction by Geomagnetic Field Built-in Sensor Utilization

Sensors ◽  
2021 ◽  
Vol 21 (9) ◽  
pp. 3087
Author(s):  
Sandi Ljubic ◽  
Franko Hržić ◽  
Alen Salkanovic ◽  
Ivan Štajduhar
Keyword(s):  
Magnetic Field ◽  
Neural Networks ◽  
Curve Fitting ◽  
Absolute Error ◽  
Interaction Space ◽  
Text Entry ◽  
Proof Of Concept ◽  
Field Sensor ◽  
Device Interaction ◽  
Input Techniques

In this paper, we investigate the possibilities for augmenting interaction around the mobile device, with the aim of enabling input techniques that do not rely on typical touch-based gestures. The presented research focuses on utilizing a built-in magnetic field sensor, whose readouts are intentionally affected by moving a strong permanent magnet around a smartphone device. Different approaches for supporting magnet-based Around-Device Interaction are applied, including magnetic field fingerprinting, curve-fitting modeling, and machine learning. We implemented the corresponding proof-of-concept applications that incorporate magnet-based interaction. Namely, text entry is achieved by discrete positioning of the magnet within a keyboard mockup, and free-move pointing is enabled by monitoring the magnet’s continuous movement in real-time. The related solutions successfully expand both the interaction language and the interaction space in front of the device without altering its hardware or involving sophisticated peripherals. A controlled experiment was conducted to evaluate the provided text entry method initially. The obtained results were promising (text entry speed of nine words per minute) and served as a motivation for implementing new interaction modalities. The use of neural networks has shown to be a better approach than curve fitting to support free-move pointing. We demonstrate how neural networks with a very small number of input parameters can be used to provide highly usable pointing with an acceptable level of error (mean absolute error of 3 mm for pointer position on the smartphone display).

scholarly journals Prediction of Blended Yarn Evenness and Tensile Properties by Using Artificial Neural Network and Multiple Linear Regression

2016 ◽  
Vol 16 (2) ◽  
pp. 43-50 ◽  
Author(s):  
Samander Ali Malik ◽  
Assad Farooq ◽  
Thomas Gereke ◽  
Chokri Cherif
Keyword(s):  
Neural Networks ◽  
Linear Regression ◽  
Multiple Linear Regression ◽  
Tensile Properties ◽  
Regression Models ◽  
Prediction Models ◽  
Research Work ◽  
Absolute Error ◽  
Knowledge Domain ◽  
Artificial Neural

Abstract The present research work was carried out to develop the prediction models for blended ring spun yarn evenness and tensile parameters using artificial neural networks (ANNs) and multiple linear regression (MLR). Polyester/cotton blend ratio, twist multiplier, back roller hardness and break draft ratio were used as input parameters to predict yarn evenness in terms of CVm% and yarn tensile properties in terms of tenacity and elongation. Feed forward neural networks with Bayesian regularisation support were successfully trained and tested using the available experimental data. The coefficients of determination of ANN and regression models indicate that there is a strong correlation between the measured and predicted yarn characteristics with an acceptable mean absolute error values. The comparative analysis of two modelling techniques shows that the ANNs perform better than the MLR models. The relative importance of input variables was determined using rank analysis through input saliency test on optimised ANN models and standardised coefficients of regression models. These models are suitable for yarn manufacturers and can be used within the investigated knowledge domain.

scholarly journals Modeling of Heavy Metal (Ni, Mn, Co, Zn, Cu, Pb, and Fe) and PAH Content in Stormwater Sediments Based on Weather and Physico-Geographical Characteristics of the Catchment-Data-Mining Approach

Water ◽  
2019 ◽  
Vol 11 (3) ◽  
pp. 626 ◽  
Author(s):  
Łukasz Bąk ◽  
Bartosz Szeląg ◽  
Aleksandra Sałata ◽  
Jan Studziński
Keyword(s):  
Heavy Metals ◽  
Neural Networks ◽  
Multilayer Perceptron ◽  
Sediment Quality ◽  
Drainage System ◽  
Chemical Properties ◽  
Absolute Error ◽  
Atmospheric Conditions ◽  
Drainage Systems ◽  
Catchment Characteristics

The processes that affect sediment quality in drainage systems show high dynamics and complexity. However, relatively little information is available on the influence of both catchment characteristics and meteorological conditions on sediment chemical properties, as those issues have not been widely explored in research studies. This paper reports the results of investigations into the content of selected heavy metals (Ni, Mn, Co, Zn, Cu, Pb, and Fe) and polycyclic aromatic hydrocarbons (PAHs) in sediments from the stormwater drainage systems of four catchments located in the city of Kielce, Poland. The influence of selected physico-geographical catchment characteristics and atmospheric conditions on pollutant concentrations in the sediments was also analyzed. Based on the results obtained, statistical models for forecasting the quality of stormwater sediments were developed using artificial neural networks (multilayer perceptron neural networks). The analyses showed varied impacts of catchment characteristics and atmospheric conditions on the chemical composition of sediments. The concentration of heavy metals in sediments was far more affected by catchment characteristics (land use, length of the drainage system) than atmospheric conditions. Conversely, the content of PAHs in sediments was predominantly affected by atmospheric conditions prevailing in the catchment. The multilayer perceptron models developed for this study had satisfactory predictive abilities; the mean absolute error of the forecast (Ni, Mn, Zn, Cu, and Pb) did not exceed 21%. Hence, the models show great potential, as they could be applied to, for example, spatial planning for which environmental aspects (i.e., sediment quality in the stormwater drainage systems) are accounted.

scholarly journals Enhancing protein backbone angle prediction by using simpler models of deep neural networks

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Fereshteh Mataeimoghadam ◽  
M. A. Hakim Newton ◽  
Abdollah Dehzangi ◽  
Abdul Karim ◽  
B. Jayaram ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Structure Prediction ◽  
Protein Structures ◽  
Absolute Error ◽  
Grand Challenge ◽  
Protein Backbone ◽  
The Neural Network ◽  
Benchmark Datasets ◽  
The Neural Networks

Abstract Protein structure prediction is a grand challenge. Prediction of protein structures via the representations using backbone dihedral angles has recently achieved significant progress along with the on-going surge of deep neural network (DNN) research in general. However, we observe that in the protein backbone angle prediction research, there is an overall trend to employ more and more complex neural networks and then to throw more and more features to the neural networks. While more features might add more predictive power to the neural network, we argue that redundant features could rather clutter the scenario and more complex neural networks then just could counterbalance the noise. From artificial intelligence and machine learning perspectives, problem representations and solution approaches do mutually interact and thus affect performance. We also argue that comparatively simpler predictors can more easily be reconstructed than the more complex ones. With these arguments in mind, we present a deep learning method named Simpler Angle Predictor (SAP) to train simpler DNN models that enhance protein backbone angle prediction. We then empirically show that SAP can significantly outperform existing state-of-the-art methods on well-known benchmark datasets: for some types of angles, the differences are 6–8 in terms of mean absolute error (MAE). The SAP program along with its data is available from the website https://gitlab.com/mahnewton/sap.

Continuous Drilling Sensor Data Reconstruction and Prediction via Recurrent Neural Networks

2020 ◽  
Author(s):  
Andrzej T. Tunkiel ◽  
Tomasz Wiktorski ◽  
Dan Sui
Keyword(s):  
Neural Networks ◽  
Absolute Error ◽  
Sensor Data ◽  
Neural Network Approach ◽  
Hard Drives ◽  
Exhaust Temperature ◽  
Novel Approach ◽  
Weight On Bit ◽  
Curved Section ◽  
Oilfield Operations

Abstract There is an ever-increasing amount of data being recorded in oilfield operations. During drilling a well a large number of parameters is being monitored and saved, often reaching several hundreds. We are seemingly monitoring everything, from basic parameters such as Weight on Bit, Torque, and Rate of Penetration (ROP), to the exhaust temperature of engine no. 3. Unfortunately, the quality of collected data does not match the quantity. Critical sensors, such as gamma and inclination, are often lagging many meters behind the bit. Despite best efforts, sensors stop working, hard drives corrupt files, and data mud pulse telemetry uplinks fail. Methods of infilling data spanning many meters or minutes are necessary. We present a novel approach that enables reliable prediction of data lagging behind the bit through deep neural networks by merging trend-based prediction with traditional neural network approach. We were able to predict continuous inclination data in a curved section of a well with an average absolute error of only 0.4 degrees up to 20 meters from last known value.

scholarly journals Neural Network Based Model Comparison for Intraday Electricity Price Forecasting

Energies ◽  
2019 ◽  
Vol 12 (23) ◽  
pp. 4557 ◽  
Author(s):  
Ilkay Oksuz ◽  
Umut Ugurlu
Keyword(s):  
Neural Networks ◽  
Electricity Markets ◽  
Electricity Market ◽  
Network Architecture ◽  
Model Comparison ◽  
Market Price ◽  
Absolute Error ◽  
Electricity Price ◽  
Price Forecasting ◽  
Electricity Price Forecasting

The intraday electricity markets are continuous trade platforms for each hour of the day and have specific characteristics. These markets have shown an increasing number of transactions due to the requirement of close to delivery electricity trade. Recently, intraday electricity price market research has seen a rapid increase in a number of works for price prediction. However, most of these works focus on the features and descriptive statistics of the intraday electricity markets and overlook the comparison of different available models. In this paper, we compare a variety of methods including neural networks to predict intraday electricity market prices in Turkish intraday market. The recurrent neural networks methods outperform the classical methods. Furthermore, gated recurrent unit network architecture achieves the best results with a mean absolute error of 0.978 and a root mean square error of 1.302. Moreover, our results indicate that day-ahead market price of the corresponding hour is a key feature for intraday price forecasting and estimating spread values with day-ahead prices proves to be a more efficient method for prediction.

Sign in / Sign up

Export Citation Format

Share Document