News Release on Neural Network Research: August-2018

Memristors power quick-learning neural network

A new sort of neural network created with memristors will dramatically improve the potency of teaching machines to suppose like humans. The network, known as a reservoir computer system, may predict words before they’re aforementioned throughout the speech, and facilitate predict future outcomes supported this. [1]

Prediction of compressive and flexural strengths of a modified zeolite additive mortar using artificial neural network

Artificial neural network (ANN) has been wont to predict the compressive and flexural strength of a mortar created with a changed water softener additive (MZA). The ANN had 3 layers, including the input, hidden and output layer. The input layer had six parameters: cement amount, silicon oxide sand amount, changed water softener additive (MZA) amount, water amount, action amount, and cargo weights. The output layer consisted of either the compressive or the flexural strength. whereas developing the ANN model, thirty samples were used for coaching and testing. 2 assessments were applied, initial to work out the effective range of neurons within the hidden layer in predicting the compressive strength. The second assessment evaluated the accuracy with that the neural network would predict the compressive or flexural strength beneath totally different load weights. [2]

Estimation of hydrogen flow rate in atmospheric Ar:H2 plasma by using artificial neural network

Atmospheric Ar:H2 plasma is Associate in Nursing eco-friendly choice for the reduction of metal oxides. For higher reduction performance and safety concern, the element gas injected into the reactor ought to be monitored. An element rate estimation system is conferred during this paper by mistreatment a man-made neural network (ANN) model fed with options of optical emission spectra of the plasma. ANN models area unit studied with 2 completely different sets of input, i.e. for the first case the inputs to the model are the three features of Hα line such as the peak intensity count, full-width half-maximum and area under Hα line, whereas for the second case, the peak intensity count of a group of emission lines like Hα, Ar I, O I, K I, metallic element D lines area unit thought-about because the inputs. [3]

Developing artificial neural network models to predict allowable bearing capacity and elastic settlement of shallow foundation in Sharjah, United Arab Emirates

This analysis proposes the utilization of artificial neural network to predict the allowable bearing capability and elastic settlement of shallow foundation on granular soils in Sharjah, United Arab Emirates. knowledge obtained from existing soil reports of 600 boreholes were wont to train and validate the model. 3 parameters (footing breadth, effective unit weight, and SPT blow count) ar thought-about to possess the foremost important impact on the magnitude of allowable bearing capability and elastic settlement of shallow foundations, and therefore were used because of the model inputs. Throughout the study, depth of footing was restricted to one.5 m below the existing ground level and groundwater level depth taken at the extent of the footing. [4]

PLANTING PATTERN MANAGEMENT BASED ON RBF NEURAL NETWORK AND OPTIMIZE PROFIT TO DETERMINANE THE TIME PLANTING SEASON ON LOMBOK ISLAND

Cropping pattern area unit determination concerning planting schedule, sort of planting, and planting space that applied on irrigation space. to get ideal planting pattern, we must always be leveling between water accessibility and water required for every plant. exploitation prediction of hydrological information reminiscent of rain and climatological information reminiscent of temperature, wind speed, radiation, and wetness will verify water accessibility on irrigation space. This prediction methodology referred to as Radial Basis operate Artificial Neural Network (RBF). the information that used for this analysis reminiscent of rain, temperature, wind speed, radiation, and wetness were taken from the Water Resources info Center (WRIC) West Nusa Tenggara (NTB) throughout the last thirty-one years, that is from 1983 till 2013. This information wont to predict hydrological and climatological information on 2014. The result information will verify the water consumption wants for a plant (evapotranspiration), the effectiveness of rain, and preparation of water consumption would like for land, then connected with irrigation water accessibility volume and the way a long period of cropping plant to induce cropping pattern style. [5]

Reference

[1] Memristors power quick-learning neural network

Date: December 22, 2017, Source: University of Michigan (web link)

[2] Prediction of the compressive and flexural strengths of a modified zeolite additive mortar using an artificial neural network

Onyari EK, Ikotun BD. Prediction of compressive and flexural strengths of a modified zeolite additive mortar using artificial neural network. Construction and Building Materials. 2018 Oct 30;187:1232-41. (web link)

[3] Estimation of hydrogen flow rate in atmospheric Ar:H2 plasma by using the artificial neural network

Das S, Das DP, Sarangi CK, Bhoi B. Estimation of hydrogen flow rate in atmospheric Ar: H 2 plasma by using artificial neural network. Neural Computing and Applications. 2018:1-9.(web link)

[4] Developing artificial neural network models to predict allowable bearing capacity and elastic settlement of shallow foundation in Sharjah, United Arab Emirates

Omar M, Hamad K, Al Suwaidi M, Shanableh A. Developing artificial neural network models to predict allowable bearing capacity and elastic settlement of shallow foundation in Sharjah, United Arab Emirates. Arabian Journal of Geosciences. 2018 Aug 1;11(16):464. (web link)

[5] PLANTING PATTERN MANAGEMENT BASED ON RBF NEURAL NETWORK AND OPTIMIZE PROFIT TO DETERMINE THE TIME PLANTING SEASON ON LOMBOK ISLAND

AFIK RITONGA1* AND MOHAMMAD ISA IRAWAN1

1Department of Mathematics, Institut Teknologi Sepuluh Nopember, Faculty of Mathematics and Natural Sciences, Surabaya, Indonesia. (web link)

Be the first to comment

Leave a Reply

Your email address will not be published.


*