Wyniki wyszukiwania - BazTech

1

The assessment of artificial neural network rainfall-runoff models under different input meteorological parameters. Case study: Seybouse basin, Northeast Algeria

Aoulmi Yamina, Marouf Nadir, Amireche Mohamed

Journal of Water and Land Development

|

2021

|

no. 50

38--47

EN

Over the past two decades, artificial neural networks (ANN) have exhibited a significant progress in predicting and modeling non-linear hydrological applications, such as the rainfall-runoff process which can provide useful contribution to water resources planning and management. This research aims to test the practicability of using ANNs with various input configurations to model the rainfall-runoff relationship in the Seybouse basin located in a semi-arid region in Algeria. Initially, the ANNs were developed for six sub-basins, and then for the complete watershed, considering four different input configurations. The 1st (ANN IP) considers only precipitation as an input variable for the daily flow simulation. The 2nd (ANN II) considers the 2nd variable in the model input with precipitation; it is one of the meteorological parameters (evapotranspiration, temperature, humidity, or wind speed). The third (ANN IIIP,T,HUM) considers a combination of temperature, humidity, and precipitation. The last (ANN VP,ET,T,HUM,Vw) consists in collating different meteorological parameters with precipitation as an input variable. ANN models are made for the whole basin with the same configurations as specified above. Better flow simulations were provided by (ANN IIP,T) and (ANN IIP,Vw) for the two stations of Medjez-Amar II and Bordj-Sabath, respectively. However, the (ANN VP,ET,T,HUM,Vw)’s application for the other stations and also for the entire basin reflects a strategy for the flow simulation and shows enhancement in the prediction accuracy over the other models studied. This has shown and confirmed that the more input variables, as more efficient the ANN model is.

2

Machine learning for predicting discharge fuctuation of a karst spring in North China

Cheng Shu, Qiao Xiaojuan, Shi Yaolin, Wang Dawei

Acta Geophysica

|

2021

|

Vol. 69, no. 1

257--270

EN

The quantitative analyses of karst spring discharge typically rely on physical-based models, which are inherently uncertain. To improve the understanding of the mechanism of spring discharge fuctuation and the relationship between precipitation and spring discharge, three machine learning methods were developed to reduce the predictive errors of physical-based groundwater models, simulate the discharge of Longzici spring’s karst area, and predict changes in the spring on the basis of long time series precipitation monitoring and spring water fow data from 1987 to 2018. The three machine learning methods included two artifcial neural networks (ANNs), namely multilayer perceptron (MLP) and long short-term memory–recurrent neural network (LSTM–RNN), and support vector regression (SVR). A normalization method was introduced for data preprocessing to make the three methods robust and computationally efcient. To compare and evaluate the capability of the three machine learning methods, the mean squared error (MSE), mean absolute error (MAE), and root-mean-square error (RMSE) were selected as the performance metrics for these methods. Simulations showed that MLP reduced MSE, MAE, and RMSE to 0.0010, 0.0254, and 0.0318, respectively. Meanwhile, LSTM–RNN reduced MSE to 0.0010, MAE to 0.0272, and RMSE to 0.0329. Moreover, the decrease in MSE, MAE, and RMSE was 0.0397, 0.1694, and 0.1991, respectively, for SVR. Results indicated that MLP performed slightly better than LSTM–RNN, and MLP and LSTM–RNN performed considerably better than SVR. Furthermore, ANNs were demonstrated to be prior machine learning methods for simulating and predicting karst spring discharge.

3

Discrimination of earthquakes and quarries in the Edirne district (Turkey) and its vicinity by using a linear discriminate function method and artifcial neural networks

Tan Aylin, Horasan Gündüz, Kalafat Doğan, Gülbağ Ali

Acta Geophysica

|

2021

|

Vol. 69, no. 1

17--27

EN

In this study, seismic events in the Edirne district (Turkey) and its vicinity have been investigated in order to discriminate earthquakes from quarry blasts. A total of 150 seismic events with Md≤3.5 duration magnitude from a seismic activity catalog between 2009 and 2014 recorded by the Enez (ENEZ), Erikli (ERIK) and Gelibolu (GELI) broadband stations operated by Boğaziçi University, Kandilli Observatory and Earthquake Research Institute Regional Earthquake-Tsunami Monitoring Center were used in this study. The maximum S-wave and maximum P-wave amplitude ratio of vertical component velocity seismograms, power ratio (Complexity) and total signal duration of the waveform were calculated. Earthquakes and quarry blasts were discriminated using the linear discriminate function (LDF) and back propagation feed forward neural networks, an artifcial neural network (ANN) learning algorithm, taking the determination coefcient and variance account values between these parameters into consideration. Eighty-one (54%) of the total 150 seismic events studied were determined to be earthquakes, and sixty-nine (46%) of them were determined to be quarry blasts. The LDF and ANNs methods were applied to the data in Edirne and its vicinity using a pair of parameters and were compared to each other for the frst time. The accuracy of the methods are 95% and 99% for LDF and ANNs, respectively.

4

Wykorzystanie sztucznej inteligencji do modelowania stanu geometrycznego powierzchni po obróbce EDM stali narzędziowych

Nieciąg H., Kudelski R., Zagórski K.

Autobusy : technika, eksploatacja, systemy transportowe

|

2018

|

R. 19, nr 12

570--574

PL

W artykule przedstawiono zastosowanie metod sztucznej inteligencji w postaci sztucznych sieci neuronowych (SSN) do modelowania stanu geometrycznego wyrobu kształtowanego w procesie EDM. Zastosowano SSN o różnej architekturze i różne algorytmy uczące. Zbadano jakość modeli i ich skuteczność w prognozowaniu wybranych cech geometrycznych wyrobów z stali narzędziowych.

EN

The article presents the application of artificial intelligence methods in the form of artificial neural networks (SSN) for modeling the geometrical state of a product shaped in the EDM process. The SSN with different architecture and different learning algorithms were implemented. The models' quality and their effectiveness in predicting some geometrical features of tool steel products were examined.

5

To predict military spending in China based on ARIMA and artificial neural networks models

Kuo K.-C., Chang C.-Y., Lin W.-C.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 3b

176--181

EN

This study takes the initiative to forecast China’s military spending based on autoregressive integrated moving average (ARIMA) models and artificial neural networks (ANNs) models. The mean absolute percentage error (MAPE) approach is applied to measure prediction accuracy. The results indicate that these single variable ARIMA models show higher accuracy and stability than those made by the single variable ANNs models across the four time periods, namely the short term (1 year), the medium term (3 years), the medium-long term (5 years), and the long term (10 years). As to multiple variable ANNs models, the prediction accuracy of each model with different variables has advantages in different time periods. The highest accuracy for the long term predictions among all of the multivariate models is made by ANN2 including China’s military spending and GDP. ANN3 including variables of China’s military spending, GDP, and inflation rates illustrates the most accurate prediction for the short term and medium-long term, while ANN4 including China’s military spending, GDP, inflation rates, and Taiwan’s military spending shows the highest accuracy for the medium term prediction. This concludes the contributions of this study.

PL

W artykule przedstawiono wyniki analizy dotyczącej przewidywanych wydatków Chin na militaria, opracowanej na podstawie modelu autoregresji (ang. ARIMA) oraz sztucznych sieci neuronowych (ANN). Dokładność predykcji oparta została na funkcji średniej wartości absolutnej procentowego uchybu. Badania wykazują, że model ARIMA ma wyższą dokładność i stabilność niż model oparty na ANN w odniesieniu do czterech, różnych okresów (1, 3, 5, 10 lat), przy czym dla ANN badanie wykonano dla czterech wartości dokładności predykcji.

6

Pulse shape discrimination of neutrons and gamma rays using kohonen artificial neural networks

Tambouratzis T., Chernikova D., Pzsit I.

Journal of Artificial Intelligence and Soft Computing Research

|

2013

|

Vol. 3, No. 2

77--88

EN

The potential of two Kohonen artificial neural networks (ANNs) - linear vector quantisation (LVQ) and the self organising map (SOM) - is explored for pulse shape discrimination (PSD), i.e. for distinguishing between neutrons (n’s) and gamma rays (’s). The effect that (a) the energy level, and (b) the relative size of the training and test sets, have on identification accuracy is also evaluated on the given PSD dataset. The two Kohonen ANNs demonstrate complementary discrimination ability on the training and test sets: while the LVQ is consistently more accurate on classifying the training set, the SOM exhibits higher n/ identification rates when classifying new patterns regardless of the proportion of training and test set patterns at the different energy levels; the average time for decision making equals ˜100 μs in the case of the LVQ and ˜450 μs in the case of the SOM.

7

Wykorzystanie SSN w analizie przekazywania się drgań wzbudzanych wstrząsami górniczymi z gruntu na budynek

Chudyba Ł.

Zeszyty Naukowe. Budownictwo / Politechnika Śląska

|

2008

|

z. 113

31-38

PL

Praca dotyczy analizy przekazywania się drgań podłoża gruntowego, spowodowanych wstrząsami górniczymi, na budynek o średniej wysokości. Wykorzystano jeden z parametrów używanych w Górniczej Skali Intensywności (GSI-2004) do oceny skutków oddziaływania drgań na obiekty powierzchniowe - maksymalną amplitudę wypadkowego przyspieszenia drgań poziomych. Porównywano wyniki jednoczesnych pomiarów drgań podłoża gruntowego i fundamentów badanego obiektu. Analizowano wpływ różnych parametrów wstrząsów i drgań. Wyniki badań doświadczalnych wykorzystano w próbie oceny dynamicznej interakcji podłoże-budynek z zastosowaniem sztucznych sieci neuronowych (SSN).

EN

The paper deals with the analysis of the transmission of ground vibrations induced by mining tremors to basement of medium-height building. The one of parameters used in scales of mine-induced dynamic influences (GSI-2004) to evaluate the harmfulness of ground vibrations to surface structures - maximal values (amplitudes) of resultant acceleration of horizontal vibrations were applied. Results of the same time vibrations measurements on the ground and on the foundation level were compared. The influence of mining tremors parameters was discussed. The experimental data were used in the soil-structure interaction evaluation with the application of artificial neural networks (ANNs).

8

Artificial Neural Networks (ANNs) as a tool of estimating relation between parameters of anti-oxidative ability and nutritional ingredients in the diet of pregnant women

Kolarzyk E., Mendyk A., Stępniewski M., Łyszczarz J., Pietrzycka A., Ostachowska-Gąsior A.

Bio-Algorithms and Med-Systems

|

2006

|

Vol. 2, no. 4

5--12

EN

The aim of this study was to evaluate if there is a relation between superoxide dismutase (SOD) activity and total anti-oxidative status measured as Ferric Reducing Abil-ity of Plasma (FRAP) concentration in saliva and intake of macro- and micro- nutritients in daily nutritional rations (DNR) in the first, second and third trimester of singleton uncompli-cated pregnancy by using artificial neural networks (ANNs). Materiał and methods: Eighty women: 40 pregnant women aged 27.1+5.4 years (examined group) and 40 healthy women (the control group) were subjects of this study. Artificial neural networks (ANN) and clustering analysis with Ward's group-ing method were used for evaluation of the relationship be¬tween FRAP and SOD in saliva and intake of macronutrients, minerals (Na, K, Ca, P, Mg, Fe, Zn, Cu, Mn), and vitamins (A, C, E, B1.B2, B6, PP)inDNR. Results: During pregnancy FRAP and SOD values were Iower than in the controls, but only for FRAP the differences were highly statistically significant (p<0.001). No functional relation-ships between FRAP and SOD and macro- and micronutrients were identified by ANN model. For whole pregnancy period cluster analysis identified 2 major clusters for which differenti-ating variables were SOD and retinoids intake but different patterns for each trimester of pregnancy were revealed. Conclusion: ANNs and cluster analysis showed that there were no statistical relationships between the intake of micro-and macronutrients in DNR and SOD or FRAP level in saliva of pregnant women.

PL

Celem pracy było określenie czy istnieje zależność między aktywnością dysmutazy ponadtlenkowej (SOD) i całkowitą wydolnością antyoksydacyjną (mierzoną wartością FRAP) w ślinie kobiet ciężarnych oraz podażą makro-i mikroelementów w diecie (w pierwszym, drugim i trzecim trymestrze ciąży). Materiał i metodyka: Badaniami objęto 80 kobiet, w tym 40 ciężarnych w wieku 27.1 +5.4 lat (grupa badana) oraz 40 zdrowych kobiet (grupa kontrolna). Sztuczne sieci neuronowe i analiza skupień (metodą Warda) były zastosowane dla oce¬ny związku między wartościami FRAP i SOD w ślinie i poda¬żą makroskładników, składników mineralnych (Na, K, Ca, P, Mg, Fe, Zn, Cu, Mn) i witamin (A, C, E, B1, B2, B6, PP) w diecie kobiet ciężarnych. Wyniki: Podczas ciąży stwierdzono niższe wartości FRAP i SOD niż w grupie kontrolnej, ale tylko wartości FRAP różniły się w sposób statystycznie znamienny (p<0.001). Przy użyciu ANNs nie wykazano istotnych zależności między FRAP oraz SOD i składnikami diety. Dla całego okresu ciąży uzyskano metodą Warda 2 główne skupienia, w których zmiennymi różnicującymi było SOD i retinoidy, nie mniej jednak dla każdego trymestru ciąży uzyskano różne wzorce. Wniosek: Zarówno metodą sztucznych sieci neuronowych jak i metodą analizy skupień wykazano, że nie ma statystycznie istotnego związku między podażą mikro- i makroelementów w diecie oraz stężeniem SOD i wartością FRAP w ślinie kobiet ciężarnych.