Estimating soil wetting patterns for drip irrigation using genetic programming

The soil wetting pattern is an important factor to fulfil the adequacy of water requirements in a controlled drip irrigation system. The soil wetting pattern is a parameter that is considered in the installation of irrigation systems and control instruments in the field. This study aimed to developed estimate vertical and horizontal soil wetting in the application of control systems using soil moisture sensor in drip irrigation systems. This research conducted in laboratory experiments with surface drip irrigation involved soil sample from pepper plantation, three discharge rates, and 26 soil moisture sensors. The empirical model provides estimated the soil wetted as a parameter of application time, emitter discharge, soil bulk density, initial soil moisture content, and saturated hydraulic conductivity. The development of empirical formulas for predicting wetted radius at different soil depths is carried out by non-linear regression analysis to obtain empirical best-fit coefficients. The empirical model of this study can estimate the full wetting pattern with acceptable accuracy and performs that represent a good performance of a controlled drip irrigation system. The result of performance appraisal validation for the wetted radius and wetted depth on sandy loam soils was assessed based on the lowest mean error was -2.86 and -0.284 cm as well as the highest efficiency model was 0.959 and 0.857 respectively.

Knowledge of the soil wetting pattern characteristics of vertical moistube-irrigation is essential for the design of cost-effective and efficient irrigation systems. We conducted laboratory experiments to determine the specific discharge calculation formula and compare the accuracy of HYDRUS-2D simulation. The cumulative infiltrations, wetting pattern distances, and water content distributions predicted with HYDRUS-2D were found to align well with experimental data. The results provide support for using HYDRUS-2D as a tool for investigating and designing moistube-irrigation management practices. Numerical simulations were carried out with HYDRUS-2D to investigate the influence of soil texture, initial water content, pressure head, moistube length, and buried depth on wetting pattern characteristics. There are small differences in the shape of the soil wetting pattern, as well as significant differences in size. The wetting pattern and water content contour are approximately “ellipsoid” around the moistube. Soil texture has a significant effect on the wetting pattern characteristics, the vertical and horizontal wetting front distance, and the wetted soil volume decrease along with the increase of soil clay content. The initial water content, pressure head, and moistube length have great influence on the wetting front distance and the wetted soil volume. Both are positively correlated with the initial water content, pressure head, and length. Moistube buried depth affects the wetting pattern position. The soil wetting pattern decreases synchronously as the buried depth drops.

Brittleness of the rock is one of the most critical features for design of underground excavation project. Therefore, proper estimating of rock brittleness can be very useful for designers and evaluators of geotechnical applications. In this study, feasibility of genetic programming (GP) model in predicting brittleness of the rock samples is examined. In this regard, several laboratory tests including uniaxial compressive strength, Brazilian, density and punch penetration were conducted on 48 rock samples collected from various locations of India. Considering multiple-inputs, several GP models were constructed to estimate brittleness index (BI) of the rock and finally, the best GP model was selected. Note that, GP can make an equation for predicting output of the system using model inputs. To show applicability of the developed GP model, non-linear multiple regression (NLMR) was also applied and developed. Considering some model performance indices, performance prediction of the GP and NLMR models were evaluated and it was found that the GP model is superior to NLMR one. Based on coefficient of determination (R2) of testing datasets, by proposing GP model, it can be improved from 0.882 (which was obtained by NLMR model) to 0.904. It is worth mentioning that the proposed predictive models in this study should be planned and used for the same conditions and the established inputs ranges.

Accurate estimation of river flow can have a significant importance in water resources management. In this study, Genetic programming (GP) and Support Vector Machine (SVM) methods were used to forecast daily discharge of Barandoozchay River. The daily discharge data of Barandoozchay River measured at the Dizaj hydrometric station during 2007 to 2011 was used for modeling, which 80% of the data used for training and remaining 20% used for testing of models. The results showed that in the both of considered methods, the models including discharges of one, two and three days ago had higher accuracy in verification step and the accuracy of models decreased with increasing discharge values. Comparing the performance of GP and SVM methods indicated that, however the accuracy of the GP method with the R=0.978 and RMSE=1.66 (m3/s) was slightly more than SVM method with R=0.976 and RMSE=1.80 (m3/s), but the SVM is easier than GP method. Thus, the SVM method can be used as an alternative method in forecasting daily river discharge.

One of the major challenges in proteomics is peptide identification from mass spectra containing high noise ratio and small number of signal (b-/y-ions) peaks. However, the accuracy and reliability of peptide identification in such highly imbalanced MS/MS data can be improved by applying a preprocessing step prior to peptide identification aiming at discriminating b-/y-ions from noise peaks in the spectra. In this study, we report a genetic programming (GP)-based preprocessing method for de-noising highly imbalanced and noisy CID MS/MS spectra. GP now becomes a popular machine learning method via automatic programming. GP preprocesses the highly noisy MS/MS spectra by classifying peaks as noise peaks or signal peaks in a binary classification manner. Meanwhile, a set of spectral fragment features based on the MS/MS fragmentation rules is extracted from the dataset to investigate their discriminating abilities by GP. A MS/MS spectral dataset containing thousands of spectra are used to train the GP model. As the GP tree-based representation has the capability for implicit feature selection during the evolutionary process, the evolved GP model with the selected features is compared with the best threshold-based method. The results show that the GP method improved the reliability of peptide identification and increased the identification rate of a de novo sequencing tool, PEAKS, to 99.4% from 80.1% achieved by the best threshold-based method. Moreover, the result of peptide identification by a database search tool, SEQUEST, using the data preprocessed by the GP method was statistically significant compared to the other methods.

Simulation of soil wetting pattern with subsurface drip irrigation from line source

ContextBecause of the complexity of soil water distribution, accurate prediction of wetting pattern is not easily accessible and this has led to a inefficiency in some proposed models in the literature. These models do not consider the hydraulic characteristics of the irrigation system and are developed solely on the basis of the water volume or infiltration rate, soil hydrodynamic properties, and other conditions of the percolation environment. ObjectiveDue to the importance of the estimation of wetting front beneath Subsurface Drip Irrigation (SDI) laterals, a semi-empirical model using easily accessible data of the SDI system is proposed to predict the wetting patterns in both distribution and redistribution phases. The aforementioned model has been developed based on a novel approach that couples the governing equations of the lateral hydraulics with empirical equations derived by dimensional analysis. MethodsTo develop the model and evaluate its performance, three 16-mm drip line pipes with 0.2, 0.4, and 0.5 m emitter spacing, and 2–5 l/h discharge were placed at 0.20 m depth in a soil box filled with clay loam soil. Water was applied for 3 hours at 50, 100, and 150 kPa operating pressures, and the wetting patterns’ geometries were measured in each lateral after 1, 2, 3, and 24 h. The performance of the model was then assessed and compared with that of three other models. Results and conclusionThe result demonstrated that the proposed model provides the most accurate estimations of the wetting depths and widths. RMSE and MAE statistical indexes of the wetting depth were 0.001–0.002 m and 0.004–0.009 m, respectively, whereas those associated with the wetting width were 0.001–0.003 m and 0.005–0.016 m, respectively. These values resulted in the lowest error when compared with the corresponding obtained from other well-known models. Consequently, the model allows acceptable predictions of the wetting patterns using accessible hydraulic parameters of the SDI system. SignificanceUniquely, the results of the lateral hydraulic analysis were applied to determine the wetting front dimensions in this study. Also, the results demonstrated that the model was successful in wetting pattern prediction beneath the lateral as a line source. While other models usually are used to estimate the wetting front around a point source.

This chapter examines the capability of Genetic Programming (GP) and different Artificial Neural Network (ANN) (Backpropagation [BP] and Generalized Regression Neural Network [GRNN]) models for prediction of air entrainment rate (QA) of triangular sharp-crested weir. The basic principal of GP has been taken from the concept of Genetic Algorithm (GA). Discharge (Q), drop height (h), and angle in triangular sharp-crested weir (?) are considered as inputs of BP, GRNN, and GP. Coefficient of Correlation (R) has been used to assess the performance of developed GP, BP, and GRNN models. For a perfect model, the value of R should be close to one. A sensitivity analysis has been carried out to determine the effect of each input parameter. This chapter presents a comparative study between the developed BP, GRNN, and GP models.

<title>Abstract</title> Small-scale farmers operating in water-scarce areas are frequently deprived of access to affordable and dependable irrigation scheduling technologies. To overcome this obstacle, a sensor-based automated drip irrigation system employing inexpensive components was developed and tested for small-scale farming purposes. The system features a microcontroller connected to in-house-developed NE555- based soil moisture and water-level sensors to enable irrigation as per the preset soil moisture limits without the need for internet connectivity. The system was evaluated in a laboratory setting using clay, sandy, loam, and sandy soils that were compacted in PVC columns and irrigated through drip emitters. Volumetric soil moisture content, irrigation duration, and water application volumes were recorded and compared with those from a commercial reference sensor (MPM160). The results revealed that the newly developed system and the reference sensor were in strong agreement in clay soils, with no significant differences at irrigation start or termination. The performance in sandy loam soil was good at irrigation termination, but there was unreliability at low-moisture levels, and there were significant differences in sandy soils due to the limitations of compaction and probe-soil contact. The irrigation durations were 18.9, 25.8, and 19.4 minutes on average, and the corresponding water volumes were 1280, 1073, and 896 mL for clay, sandy loam, and sandy soils, respectively. In general, the system was capable of performing tasks reliably in fine-textured soils and is an inexpensive sensor-based technology that can be used to help the water sector.

A laboratory experiment was carried out to study the relationships of emitter discharge rate (EDR) and distribution of water and salt for two typical coastal saline soils. The objective was to find a simple method to determine a suitable EDR for reclamation of coastal saline soil using drip irrigation. The experiment included five EDR treatments in a coastal saline sandy-loam soil (SAS): 0.23, 0.55, 0.8, 1.5, and 2.9 L·h−1; another five EDR treatments were conducted in a coastal saline silt soil (SIS): 0.23, 0.55, 1.0, 1.5, and 2.0 L·h−1. The water distribution and the low-salinity zone (with electrical conductivity of the saturated soil extract ≤1, ≤2, and ≤4 dS·m−1) were clearly affected by EDR. The relationships of EDR and wetted front, wetted zone volume, and low-salinity zone were difficult to obtain because of the requirements of determining and calculating the vertical wetted front and the volume of wetted, saturated, and low-salinity zones in the field. For this reason, these relationships could not be used to directly calculate EDR. Steady-state ponding area radius and EDR were linearly related, and constant ponding area was related to EDR but not to time. A method was proposed to rapidly determine EDR from this linear relationship. Based on this method, a specific radius of the steady-state ponding area, proposed at 3–4 cm, was substituted into the linear relationship to calculate suitable EDRs of 1.3–1.9 and 0.16–0.40 L·h−1 for SAS and SIS soils, respectively. Two field experiments were used to test the practicability of the method; there was no runoff or erosion in the irrigation period, and plant survival rates were not affected by salt. The method proposed can rapidly determine EDR (within 300 min) with a small amount of water (12 L), which saves labor (one or two persons), and it can be implemented in multiple locations without destroying soil and plants.

Emitter clogging is a persistent problem for drip irrigation using treated effluent. To find an effective flushing regime, research was conducted using two water types (treated effluent and fresh groundwater) with two flushing frequencies (every irrigation event and every five irrigation events) and a control (no flushing). Thin-wall laterals with inline-labyrinth emitters were examined. The irrigation duration was 1 h daily for 91 days. Emitter discharge was monitored. At the end of the study, the mean emitter discharge decreased about 9% in the treated effluent treatment, 7–8% in the fresh water treatment and 13% in the control treatment. Flushing was effective in preventing sediment accumulation, and using fresh water reduced emitter clogging better than using treated effluent. But there were no significant differences between the two flushing frequencies. Flushing with fresh water reduced the calcium and magnesium deposits of the emitter, and the changeable water environment discouraged the bacteria growth. Therefore, when treated effluent is used for drip irrigation, emitter clogging should be controlled by flushing with fresh water every five irrigation events.

Ionic liquids (ILs) are more widely used within the industry than ever before, and accurate models of their physicochemical characteristics are becoming increasingly important during the process optimization. It is especially challenging to simulate the viscosity of ILs since there is no widely agreed explanation of how viscosity is determined in liquids. In this research, genetic programming (GP) and group method of data handling (GMDH) models were used as white-box machine learning approaches to predict the viscosity of pure ILs. These methods were developed based on a large open literature database of 2813 experimental viscosity values from 45 various ILs at different pressures (0.06–298.9 MPa) and temperatures (253.15–573 K). The models were developed based on five, six, and seven inputs, and it was found that all the models with seven inputs provided more accurate results, while the models with five and six inputs had acceptable accuracy and simpler formulas. Based on GMDH and GP proposed approaches, the suggested GMDH model with seven inputs gave the most exact results with an average absolute relative deviation (AARD) of 8.14% and a coefficient of determination (R2) of 0.98. The proposed techniques were compared with theoretical and empirical models available in the literature, and it was displayed that the GMDH model with seven inputs strongly outperforms the existing approaches. The leverage statistical analysis revealed that most of the experimental data were located within the applicability domains of both GMDH and GP models and were of high quality. Trend analysis also illustrated that the GMDH and GP models could follow the expected trends of viscosity with variations in pressure and temperature. In addition, the relevancy factor portrayed that the temperature had the greatest impact on the ILs viscosity. The findings of this study illustrated that the proposed models represented strong alternatives to time-consuming and costly experimental methods of ILs viscosity measurement.

A context‐free genetic programming (GP) method is presented that simulated local scale daily extreme (maximum and minimum) temperatures based on large scale atmospheric variables. The method evolves simple and optimal models for downscaling daily temperature at a station. The advantage of the context‐free GP method is that both the variables and constants of the candidate models are optimized and consequently the selection of the optimal model. The method is applied to the Chute‐du‐Diable weather station in Northeastern Canada along with the National Center for Environmental Prediction (NCEP) reanalysis datasets. The performance of the GP based downscaling models is compared to benchmarks from a commonly used statistical downscaling model. The experiment results show that the models evolved by the GP are simpler and more efficient for downscaling daily extreme temperature than the common statistical method. The different model test results indicate that the GP approach significantly outperforms the statistical method for the downscaling of daily minimum temperature, while for the maximum temperature the two methods are almost equivalent. However, the GP method remains slightly more effective for maximum temperature downscaling than the statistical method.

ISHS X International Symposium on Integrating Canopy, Rootstock and Environmental Physiology in Orchard Systems ROOTSTOCK AND ET-BASED IRRIGATION EFFECTS ON MINERAL NUTRITION, YIELD AND HARVEST QUALITY OVER DIFFERENT SEASONS IN 'PACIFIC GALA' APPLE

A new prediction model is derived for the uplift capacity of suction caissons using a hybrid method coupling genetic programming (GP) and simulated annealing (SA), called GP/SA. The predictor variables included in the analysis are the aspect ratio of caisson, shear strength of clayey soil, load point of application, load inclination angle, soil permeability, and loading rate. The proposed model is developed based on well established and widely dispersed experimental results gathered from the literature. To verify the applicability of the proposed model, it is employed to estimate the uplift capacity of parts of the test results that are not included in the modeling process. Traditional GP and multiple regression analyses are performed to benchmark the derived model. The external validation of the GP/SA and GP models was further verified using several statistical criteria recommended by researchers. Contributions of the parameters affecting the uplift capacity are evaluated through a sensitivity analysis. A subsequent parametric analysis is carried out and the obtained trends are confirmed with some previous studies. Based on the results, the GP/SA-based solution is effectively capable of estimating the horizontal, vertical and inclined uplift capacity of suction caissons. Furthermore, the GP/SA model provides a better prediction performance than the GP, regression and different models found in the literature. The proposed simplified formulation can reliably be employed for the pre-design of suction caissons. It may be also used as a quick check on solutions developed by more time consuming and in-depth deterministic analyses.