Higher Root Mean Square Research Articles

Determining and modeling rheological characteristics of quince seed gum

In this research, rheological characteristics of quince seed gum were investigated. To describe shear thinning behavior, Cross model with high determination coefficient (R2) and low root mean square deviation (RMSE) was applied at shear rate of 0.01–1000 (s−1) and concentration range of 0.01–0.7 (%, w/w). Using Arrhenius model, activation energy was estimated to be 6988.74 J mol−1in the gum solution (0.3%) and temperature range of 25–55 °C. Linear viscoelastic range of the gum solution (0.3%, w/w) in the presence and absence of 0.2 M NaCl was measured as 11.4% and 7.7%, respectively. Loss and storage moduli of the gum solution (0.3%, w/w) at frequencies of 0.01–100 (Hz) showed that, at all of these frequencies, storage modulus was more than loss modulus and both of them increased with increase of frequency. Critical polymer concentration (C∗) was determined as 0.077 g 100 ml−1 through slopes of the lines for dilute and semi-dilute regions in concentration range of 0.01–0.7 (%, w/w). Capillary viscometer determined intrinsic viscosity and Huggins constant of quince seed gum as 1530 ml/g and 0.32, respectively.

Electromyographic assessment of trunk and shoulder muscles during a Pilates pull-up exercise

This study compares surface electromyographic activity of the internal oblique, rectus abdominis, multifidus, iliocostalis, anterior deltoids during the pull-up on a lower and on a higher difficulty level. We assessed nine adults with previous experience in Pilates. The root mean square (RMS) values were normalized by maximum isometric contraction for each participant. During the ascent phase, the low spring position showed a significantly higher RMS than the high spring position of 8.9% for deltoid, 17.2% for internal oblique, 22.3% for rectus abdominis, 4.1% for iliocostalis, and 5.6% for multifidus, and in the descent phase, the RMS in the lower spring exceeded significantly the high spring position in 1.6% for the deltoid, 10% for internal oblique, 31.4% for rectus abdominis and 11.4% for iliocostalis. There was no predominance of abdominal muscles over the shoulder muscle in any spring position. The pull-up exercise can be a useful choice for the core and anterior deltoid muscles strengthening.

Through-package fat determination in commercial samples of mayonnaise and salad dressing using time-domain nuclear magnetic resonance spectroscopy and chemometrics

Although fat is an essential component in the human diet, the consumption of food with high fat content has been the major cause of obesity. Here we are demonstrating that time-domain nuclear magnetic resonance (TD-NMR) decays, obtained by Carr–Purcell–Meiboom–Gill (CPMG) pulse sequence, can be used to predict the total fat content in sealed packages of commercial food emulsions, such as mayonnaise and salad dressing that may contain more than 50% of fat. The PLS model of the CPMG decays shows high linear correlation (>0.9) and low root mean square errors (RMSE) for cross-validation and validation data. Therefore, this procedure can be used to predict the total fat content in food emulsion after packaging, to detect problems in the manufacturing process or fraud. In addition, the TD-NMR instrument and the proposed methods can also be adapted for using at the end of the food chain (store) to measure the fat content in consumer packaged goods.

Reducing Adhesion Force by Means of Atomic Layer Deposition of ZnO Films with Nanoscale Surface Roughness

Adhesion is a big concern for the design of Si-based microelectromechanical devices. A ZnO film with nanoscale surface roughness is a promising candidate to decrease adhesion as the protective coating. In this study, the adhesion force of ZnO films prepared by atomic layer deposition (ALD) on a Si (100) substrate was studied. The root-mean-square (RMS) roughness of the ZnO films was in the range of 0.7-4.28 nm, and the contact angle of water was in the range of 85-88°. The adhesion force was measured by atomic force microscopy (AFM) at both low (12%) and high (60%) relative humidities. The results show that the adhesion force decreases as the surface roughness increases. A low adhesion force at high RMS roughness is attributed to the large asperities on the film, and a large adhesion force at high humidity is attributed to the large capillary force. The experimental adhesion force was compared to the force calculated using the Rabinovich model. Although the theoretical value underestimates the experimental value, the proportion of the two components of the adhesion force is clearly shown. At the low humidity, the van der Waals force component differs not greatly with the capillary force component, while at the high humidity, the capillary force component becomes dominant.

The effect of workplace smoking bans on heart rate variability and pulse wave velocity of non-smoking hospitality workers

To investigate the effect of a change in second-hand smoke (SHS) exposure on heart rate variability (HRV) and pulse wave velocity (PWV), this study utilized a quasi-experimental setting when a smoking ban was introduced. HRV, a quantitative marker of autonomic activity of the nervous system, and PWV, a marker of arterial stiffness, were measured in 55 non-smoking hospitality workers before and 3-12 months after a smoking ban and compared to a control group that did not experience an exposure change. SHS exposure was determined with a nicotine-specific badge and expressed as inhaled cigarette equivalents per day (CE/d). PWV and HRV parameters significantly changed in a dose-dependent manner in the intervention group as compared to the control group. A one CE/d decrease was associated with a 2.3% (95% CI 0.2-4.4; p = 0.031) higher root mean square of successive differences (RMSSD), a 5.7% (95% CI 0.9-10.2; p = 0.02) higher high-frequency component and a 0.72% (95% CI 0.40-1.05; p < 0.001) lower PWV. PWV and HRV significantly improved after introducing smoke-free workplaces indicating a decreased cardiovascular risk.

In vitro estimation of pressure drop across tracheal tubes during high-frequency percussive ventilation

Tracheal tubes (TT) are used in clinical practice to connect an artificial ventilator to the patient's airways. It is important to know the pressure used to overcome tube impedance to avoid lung injury. Although high-frequency percussive ventilation (HFPV) has been increasingly used, the mechanical behavior of TT under HFPV has not yet been described. Thus, we aimed at characterizing in vitro the pressure drop across TT (ΔPTT) by identifying the model that best fits the measured pressure–flow (P–) relationships during HFPV under different working pressures (PWork), percussive frequencies and mechanical loads. Three simple models relating ΔPTT and flow were tested. Model 1 is characterized by linear resistive [Rtube ⋅ ] and inertial [I ⋅ ] terms. Model 2 takes into consideration Rohrer's approach [K1 ⋅ + K2 ⋅ ] and inertance [I ⋅ ]. In model 3 the pressure drop caused by friction is represented by the non-linear Blasius component [Kb ⋅ ] and the inertial term [I ⋅ (t)]. Model 1 presented a significantly higher root mean square error of approximation than models 2 and 3, which were similar. Thus, model 1 was not as accurate as the latter, possibly due to turbulence. Model 3 presented the most robust resistance-related coefficient. Estimated inertances did not vary among the models using the same tube. In conclusion, in HFPV ΔPTT can be easily calculated by the physician using model 3.

Determination of quality parameters of tomato paste using guided microwave spectroscopy

Guided microwave spectrometry was evaluated for in-line determination of quality attributes, soluble solids content, titratable acidity and Bostwick consistency of tomato pastes in continuous processing. The spectrometer was tested for two processing operations, dilution and evaporation. A total of 34 tomato paste samples were prepared by blending 9 different tomato pastes together in various proportions with final Bostwick consistency ranging from 2 to 11 cm. During the processing operation, spectra of 5 samples were obtained when different soluble solids contents were achieved. At the same time, standard reference methods were used to measure the parameters of interest on the same samples taken from the processing flow loop. The spectrum was recorded from 136 to 2690 MHz in 2 MHz steps. Partial Least Square (PLS) was employed to produce prediction models for each quality attribute from the collected microwave spectra. Cross-validation was performed to validate the calibration models. These models demonstrated good predictability of the soluble solids content, titratable acidity and Bostwick consistency in both dilution and evaporation conditions, as shown by their high correlation coefficient (r) and low root mean square error of cross-validation (RMSECV) values. Interval PLS (iPLS) was applied to optimize the models by selecting the most relevant and non-noisy regions in the spectra. The iPLS models showed improved accuracy for prediction of all three of the parameters, especially for Bostwick consistency (r > 0.92 and RMSECV < 0.7 cm). These observations suggest that guided microwave spectroscopy gave accurate estimation of soluble solids content, titratible acidity and Bostwick consistency of tomato paste. Different processing conditions required varying prediction models for the changes in the dielectric properties of tomato pastes.

Effect of Annealing on Surface of Nickel (Ni)/Indium Tin Oxide (ITO) Nanostructures Measured by Atomic Force Microscopy (AFM)

Nickel (Ni) / indium tin oxide (ITO) nanostructures were deposited on silicon (111) substrate by RF magnetron sputtering using a nickel target and metallic alloy target (In-Sn, 90%-10%). The post-deposition annealing has been done for Ni/ITO films in air and the effect of annealing temperature on the surface morphology of ITO films was studied. It has been found that the annealing temperatures increase the film surface roughness in Ni/ITO structure. At annealing temperature of 600°C, AFM analysis reveals the highest root mean square roughness, peak to valley and thickness value of 2.598 nm, 59.115 nm, and 11.358 nm, respectively. Watershed analysis on AFM images show that the numbers of grain boundaries in Ni/ITO are reduced when annealing temperature is increased to higher temperatures.

Low dispersive modeling of Rayleigh waves on partly staggered grids

In elastic media, finite-difference (FD) implementations of free-surface (FS) boundary conditions on partly staggered grid (PSG) use the highly dispersive vacuum formulation (VPSG). The FS boundary is embedded into a “vacuum” grid layer (null Lame’s constants and negligible density values) where the discretized equations of motion allow computing surface displacements. We place a new set of compound (stress-displacement) nodes along a planar FS and use unilateral mimetic FD discretization of the zero-traction conditions for displacement computation (MPSG). At interior nodes, MPSG reduces to standard VPSG methods and applies fourth-order centered FD along cell diagonals for staggered differentiation combined with nodal second-order FD in time. We perform a dispersion analysis of these methods on a Lamb’s problem and estimate dispersion curves from the phase difference of windowed numerical Rayleigh pulses at two FS receivers. For a given grid sampling criterion (e.g., six or ten nodes per reference S wavelength λ S), MPSG dispersion errors are only a quarter of the VPSG method. We also quantify root-mean-square (RMS) misfits of numerical time series relative to analytical waveforms. MPSG RMS misfits barely exceed 10 % when nine nodes sample the minimum S wavelength \(\lambda _{\text {MIN}}^{\mathrm {S}}\) in transit (along distances \(\sim \)145\(\lambda _{\text {MIN}}^{\mathrm {S}}\)). In same tests, VPSG RMS misfits exceed 70 %. We additionally compare MPSG to a consistently fourth-order mimetic method designed on a standard staggered grid. The latter equates the former’s dispersion errors on grids twice denser and shows higher RMS precision only on grids with six or less nodes per \(\lambda _{\text {MIN}}^{\mathrm {S}}\).

Assessing the potential of support vector machine for estimating daily solar radiation using sunshine duration

Estimation of solar radiation from sunshine duration offers an important alternative in the absence of measured solar radiation. However, due to the dynamic nature of atmosphere, accurate estimation of daily solar radiation has been being a challenging task. This paper presents an application of Support vector machine (SVM) to estimation of daily solar radiation using sunshine duration. Seven SVM models using different input attributes and five empirical sunshine-based models are evaluated using meteorological data at three stations in Liaoning province in China. All the SVM models give good performances and significantly outperform the empirical models. The newly developed model, SVM1 using sunshine ratio as input attribute, is preferred due to its greater accuracy and simple input attribute. It performs better in winter, while highest root mean square error and relative root mean square error are obtained in summer. The season-dependent SVM model is superior to the fixed model in estimation of daily solar radiation for winter, while consideration of seasonal variation of the data sets cannot improve the results for spring, summer and autumn. Moreover, daily solar radiation could be well estimated by SVM1 using the data from nearby stations. The results indicate that the SVM method would be a promising alternative over the traditional approaches for estimation of daily solar radiation.

Analysing Seasonal Differences between a Soil Water Balance Model and in Situ Soil Moisture Measurementsat Nine Locations Across Europe

We compared soil moisture from the soil water balance model for European Water Accounting (swbEWA) with in situ observations from nine locations in three European climatic zones (continental, Mediterranean and maritime temperate), for different periods between 2003 and 2011. Despite the simplicity of the swbEWA model, the patterns of temporal changes in soil moisture content are well represented at all locations. Annual averages show that the model overestimates the soil moisture content, and that overestimations are the smallest when measurements are obtained from more than one depth. These results suggest that the relationship between simulated and observed soil moisture also depends on the number of measurements and the depth over which they are taken. In the continental climate, where snow cover and frozen soil influence soil moisture, we observe higher root mean square error values in winter months. However, in the Mediterranean and maritime temperate climates, we do not observe clear common seasonal patterns in the soil moisture profile, which makes it difficult to relate the model’s accuracy to climate. With the percentage of correctness and probability of detection measures, we tested the model performance in simulating dry versus non-dry events. The percentage of the correctly classified dry and non-dry events is higher than 84 % at all locations, whereas the probability to detect dry events is significantly lower, exceeding 50 % at only four out of nine stations. The frequency distribution of consecutive days with dry soil (CDDS) confirms the model performance: higher number of short dry periods (with less than 20 days of soil moisture near wilting point) are reproduced and observed in continental climates, whereas long dry periods (longer than 50 days) are noted in the Mediterranean climate. Overall, the statistical measures suggest that the model produces the highest accuracy in summer months at the stations in continental climates, whereas in the Mediterranean climate, the accuracy is slightly higher in the colder seasons.

Real Time Implementation of Multi Carrier Based PWM Techniques for Single Phase Nine Level Trinary Source CMLI Using dSPACE

Multi carrier based Pulse Width Modulation techniques have been implemented for switching of prototype of single phase trinary source cascaded multilevel inverter in which 9- level output voltage can be achieved with fewest components. Different species of Carrier Overlapping Pulse Width Modulation (COPWM-A, COPWM-B and COPWM –C) strategies along with trapezoidal reference are used. The gating pulses are generated using dSPACE dS1103 Real Time Interface (RTI) hardware and control desk with MATLAB- SIMULINK environment. It is observed that COPWM-A provides relatively lower THD. Relatively higher RMS output voltage for modulation index ma=0.8 to 1 is obtained with COPWM – B. The performance indices used are Total Harmonic Distortion (THD), Root Mean Square (RMS) output voltage, Crest Factor (CF),Form Factor (FF) and Distortion Factor (DF).

Improving conductivity and texture in ZnO:Al sputtered thin films by sequential chemical and thermal treatments

Smooth, transparent and conductive ZnO:Al (AZO) thin films have been deposited by sputtering at room temperature on glass substrates. The conductivity and texture of the as-grown layers have been enhanced by the application of thermal heating at 350°C and chemical etching in HCl or HF solutions. The structure, morphology, optical and electrical properties of the layers have been analysed before and after the successive thermal and chemical treatments, which were performed in different sequences. The surface texture has been characterized optically by the haze parameter for transmittance, which has been related to the root mean square roughness and the autocorrelation length coefficients for each sample. The highest optical scattering has been achieved with the HF etching and subsequent heating sequence, which gave also the highest root mean square roughness. In this way, an average haze of 36%, together with total visible transmittance about 80% and sheet resistance of 8Ω/sq have been obtained. Otherwise, the application of heating before HCl or HF etching resulted in AZO films with average haze of 11%, visible transmittance about 85% and sheet resistance near 6Ω/sq.

Tully-Fisher analysis of the multiple cluster system Abell 901/902

Context.

Delay and Doppler Spreads of Nonstationary Vehicular Channels for Safety-Relevant Scenarios

Vehicular communication channels are characterized by a non-stationary time- and frequency-selective fading process due to rapid changes in the environment. The non-stationary fading process can be characterized by assuming local stationarity for a region with finite extent in time and frequency. For this finite region the wide-sense stationarity and uncorrelated-scattering (WSSUS) assumption holds approximately and we are able to calculate a time and frequency dependent local scattering function (LSF). In this paper, we estimate the LSF from a large set of measurements collected in the DRIVEWAY'09 measurement campaign, which focuses on scenarios for intelligent transportation systems. We then obtain the time-frequency-varying power delay profile (PDP) and the time-frequency-varying Doppler power spectral density (DSD) from the LSF. Based on the PDP and the DSD, we analyze the time-frequency-varying root mean square (RMS) delay spread and the RMS Doppler spread. We show that the distribution of these channel parameters follows a bi-modal Gaussian mixture distribution. High RMS delay spread values are observed in situations with rich scattering, while high RMS Doppler spreads are obtained in drive-by scenarios.

Development and verification of a non-linear disaggregation method (NL-DisTrad) to downscale MODIS land surface temperature to the spatial scale of Landsat thermal data to estimate evapotranspiration

A nonlinear method (NL-DisTrad) was developed and tested to disaggregate satellite-derived estimates of land surface temperature of MODIS (Moderate Resolution Imaging Spectrometer) with a resolution of 960m to the scale of Landsat 7 ETM+ (Enhanced Thematic Mapper Plus) at 60m. This method uses the relationship that is captured at the hot edge pixels in the feature space between the Normalised Difference Vegetation Index (NDVI) and the land surface temperature (LST) at a coarse resolution to disaggregate the LST to a finer resolution. The residuals that are generated at the coarse resolution are modelled using an Artificial Neural Network model (ANN), and the resulting residuals are added to the disaggregated LST at a fine resolution. The ANN model was built using the NDVI from the neighbourhood pixels. The hypothesis is that the LST of a pixel will not only be affected by the vegetation within the pixel but also by the vegetation of surrounding pixels. The performance of this hybrid model NL-DisTrad (Hot edge model+ANN model) is assessed by comparing the results to the existing disaggregation method, TsHARP, and the observed Landsat LST. The NL-DisTrad disaggregation results were comparable to the observed Landsat LST even for pixels with non-uniform vegetation. The statistical analysis indicated that the proposed model disaggregates the LST better than TsHARP, based on the high Nash Sutcliffe Efficiency (NSE>0.80) and low root mean square error value (RMSE<0.96K). Furthermore, using SEBAL (Surface Energy Balance Algorithm for Land), it was found that the estimates of daily evapotranspiration (ET) from the LST that were disaggregated using NL-DisTrad were comparable to the ET estimates from the observed Landsat LST data. As the disaggregation method NL-DisTrad only needs the LST–NDVI relationship at the coarse resolution, the model could be used to disaggregate the coarse resolution MODIS temperature data to the fine resolution of satellites such as IRS-P6 or SPOT-5 that do not carry any thermal sensors.

Potential soil moisture products from the aquarius radiometer and scatterometer using an observing system simulation experiment

Abstract. Using an observing system simulation experiment (OSSE), we investigate the potential soil moisture retrieval capability of the National Aeronautics and Space Administration (NASA) Aquarius radiometer (L-band 1.413 GHz) and scatterometer (L-band, 1.260 GHz). We estimate potential errors in soil moisture retrievals and identify the sources that could cause those errors. The OSSE system includes (i) a land surface model in the NASA Land Information System, (ii) a radiative transfer and backscatter model, (iii) a realistic orbital sampling model, and (iv) an inverse soil moisture retrieval model. We execute the OSSE over a 1000 × 2200 km2 region in the central United States, including the Red and Arkansas river basins. Spatial distributions of soil moisture retrieved from the radiometer and scatterometer are close to the synthetic truth. High root mean square errors (RMSEs) of radiometer retrievals are found over the heavily vegetated regions, while large RMSEs of scatterometer retrievals are scattered over the entire domain. The temporal variations of soil moisture are realistically captured over a sparely vegetated region with correlations 0.98 and 0.63, and RMSEs 1.28% and 8.23% vol/vol for radiometer and scatterometer, respectively. Over the densely vegetated region, soil moisture exhibits larger temporal variation than the truth, leading to correlation 0.70 and 0.67, respectively, and RMSEs 9.49% and 6.09% vol/vol respectively. The domain-averaged correlations and RMSEs suggest that radiometer is more accurate than scatterometer in retrieving soil moisture. The analysis also demonstrates that the accuracy of the retrieved soil moisture is affected by vegetation coverage and spatial aggregation.

THAI DIALECTS AND STRUCTURAL MODELING WITH NOISES

Fundamental Frequency (F0) conveys the prosodic information of the human speech. The modeling of the dialects’ F0 in a particular language is vital issue that should be taken into account. Four main dialects are spoken in different regions of Thailand including central, north, northeast and south regions. Another important issue is the environmental noises which is often be perceived in the daily life and causing the degradation in speech quality. The robustness of the F0 modeling techniques can be evaluated by studying the effects of noises for Thai dialects. The structural model has been chosen in this study. The four-type background noises with five different levels of power are applied in this study. The synthesized F0 from the structural model has been compared with the F0 from natural speech with different scenarios including noise types, noise levels speech dialects and speech genders. From the experimental results, the root mean square errors between the synthesized F0 and the natural F0 are calculated. When increasing the noise level, the root mean square error decreases. As for the different noise types, air-conditioner noise gives the highest level of root mean square error, while the train noise brings the lowest level of root mean square error. As for the different male speech dialects, center and northeast dialects are rather higher than those of north and south dialects. As for the different female speech dialects, north dialect has the smallest deviation among all dialects. As for the different genders, female speech give higher root mean square error than male speech for all types of noises and all power levels of noises. By using the structural model, the results confirm that all Thai dialects response the proposed model differently. Moreover, all four types of simulated noises deteriorate the F0 contours of all dialects differently.

Comparison of Reference Evapotranspiration Calculations for Southeastern North Dakota

Potential water consumption for irrigation scheduling in North Dakota was typically calculated from a reference Evapotranspiration (ETref) using the Jensen-Haise method and its associated crop coefficient (Kc) curves developed in the 1970’s and 1980’s. The ETref method proposed by the American Society of Civil Engineers, Environmental and Water Research Institute (ASCE-EWRI) reference evapotranspiration task force has shown to be more accurate and therefore more widely used than any other methods. However, to apply the ASCE-EWRI method for irrigation scheduling requires a corresponding change of the Kc curves associated with the Jensen-Haise method. In this paper, a comparison of ETref estimates for 11 methods, including the ASCE-EWRI and the Jensen-Haise methods, was conducted using 18 years of data collected in southeastern North Dakota. The results show that the annual ETref by the Jensen-Haise method was nearly the same as the ASCE-EWRI grass ETref, but with a higher Root Mean Square Deviation (RMSD), 0.903 mm d-1, and a lower coefficient of determination (R2) 0.8659. The ETref comparison for the growing season only shows an RMSD of 1.007 mm d-1, R2 of 0.7996 and 8.13% overestimation. The ETref by the Jensen-Haisemethod has a higher monthly ETref than the ASCE-EWRI in June, July, and August, and a lower monthly ETref for all other months in an 18 year period. The ETref comparisons also show that the modified Penman method used by the High Plains Regional Climate Center (HPRCC Penman) has the best accuracy and correlation with the ASCE-EWRI ETref method. Indeed, all alfalfa based ETref methods, including Kimberly Penman and HPRCC Penman, show better performance than the grass based ETref methods, including FAO24 Penman, FAO24 Radiation, FAO24 Blaney-Criddle, Priestley-Taylor, Hargreaves, and the Jensen-Haise methods.

Prediction of penetration rate of rotary-percussive drilling using artificial neural networks – a case study / Prognozowanie postępu wiercenia przy użyciu wiertła udarowo-obrotowego przy wykorzystaniu sztucznych sieci neuronowych – studium przypadku

Abstract Penetration rate in rocks is one of the most important parameters of determination of drilling economics. Total drilling costs can be determined by predicting the penetration rate and utilized for mine planning. The factors which affect penetration rate are exceedingly numerous and certainly are not completely understood. For the prediction of penetration rate in rotary-percussive drilling, four types of rocks in Sangan mine have been chosen. Sangan is situated in Khorasan-Razavi province in Northeastern Iran. The selected parameters affect penetration rate is divided in three categories: rock properties, drilling condition and drilling pattern. The rock properties are: density, rock quality designation (RQD), uni-axial compressive strength, Brazilian tensile strength, porosity, Mohs hardness, Young modulus, P-wave velocity. Drilling condition parameters are: percussion, rotation, feed (thrust load) and flushing pressure; and parameters for drilling pattern are: blasthole diameter and length. Rock properties were determined in the laboratory, and drilling condition and drilling pattern were determined in the field. For create a correlation between penetration rate and rock properties, drilling condition and drilling pattern, artificial neural networks (ANN) were used. For this purpose, 102 blastholes were observed and drilling condition, drilling pattern and time of drilling in each blasthole were recorded. To obtain a correlation between this data and prediction of penetration rate, MATLAB software was used. To train the pattern of ANN, 77 data has been used and 25 of them found for testing the pattern. Performance of ANN models was assessed through the root mean square error (RMSE) and correlation coefficient (R2). For optimized model (14-14-10-1) RMSE and R2 is 0.1865 and 86%, respectively, and its sensitivity analysis showed that there is a strong correlation between penetration rate and RQD, rotation and blasthole diameter. High correlation coefficient and low root mean square error of these models showed that the ANN is a suitable tool for penetration rate prediction.