Fast-forward solver for inhomogeneous media using machine learning methods: artificial neural network,support vector machine and fuzzy logic

Encountering with a nonlinear second-order differential equation including ϵ _r and μ _r spatial distributions, while computing the fields inside inhomogeneous media, persuaded us to find their known distributions that give exact solutions. Similarities between random distributions of electric properties and known functions lead us to estimate them using three mathematical tools of artificial neural networks (ANNs), support vector machines (SVMs) and Fuzzy Logic (FL). Assigning known functions after fitting with minimum error to arbitrary inputs using results of machine learning networks leads to achieve an approximate solution for the field inside materials considering boundary conditions. A comparative study between the methods according to the complexity of the structures as well as the accuracy and the calculation time for testing of unforeseen inputs, including classification, prediction and regression is presented. We examined the extracted pairs of ϵ _r and μ _r with ANN, SVM networks and FL and got satisfactory outputs with detailed results. The application of the presented method in zero reflection subjects is exemplified.

     

Evaluating the Performance of Agricultural Water Distribution Systems Using FIS,ANN and ANFIS Intelligent Models

Increasing water use efficiency in the agricultural sector requires the use of appropriate methods for intelligent performance evaluation of surface water distribution systems in agriculture. Therefore, in this study a systematic approach was developed for operational performance appraisal of the agricultural water distribution systems. For this purpose, Fuzzy Inference System (FIS), Artificial Neural Network (ANN), and Adaptive Neuro-Fuzzy Inference System (ANFIS) were used to evaluate the technical performance of irrigation network, considering the uncertainties in the water exploitation process. The performance of the developed models was studied on the Roodasht irrigation canal, located in central Iran, which suffers from severe fluctuations in the inflow, by evaluating the adequacy, efficiency, and equity of surface water distribution. Hydraulic simulation of water distribution system, as well as providing the information required for training and validation of the intelligent models, were performed using the HEC-RAS model. The results showed that compared to the FIS model, ANN and ANFIS models similarly predicted the model outputs with lower errors at almost the same level. The adequacy, efficiency, and equity indicators were predicted by ANFIS model with MAPE of 0.16, 0.01 and 0.23, respectively. Also, FIS model was only able to predict the efficiency and could not predict the adequacy and equity with appropriate performance. The findings of this study reveal that since the ANFIS model uses both FIS and ANN models in its structure, it considers the model uncertainty reliably, and it can be used to evaluate the performance of agricultural water systems.

     

Risk Assessment of Agricultural Water Conveyance and Delivery Systems by Fuzzy Fault Tree Analysis Method

Improving the efficiency of main Agricultural Water Conveyance and Delivery Systems (AWCDS) has a significant impact on improving water productivity in agriculture. Therefore, risk assessment of mentioned systems is necessary to increase reliability of operational performance. Accordingly, this study for the first time presents a unique framework to assess the adequacy, equity, and efficiency of agricultural water distribution and delivery risk assessment within AWCDS. In this way, the Fault Tree Analysis (FTA) technique is employed for risk assessment of “undesirability of supply and delivery”. The west Dez main irrigation canal in Khuzestan province of Iran was determined as the case study of the research. A set of questionnaires filled up by managers and experts of this irrigation district, the failure probabilities of the basic events are gathered in the form of linguistic terms. Due to the uncertainty in these terms, the system’s risk assessment to determine the failure probability of the top event was performed based on Fuzzy Fault Tree Analysis method (FFTA). The results of the study showed that the failure probability in the fuzzy approach is 0.55 which is roughly 0.15 more than crisp approach. Also, the rating of the basic events based on their contribution to the occurrence of the top event was carried out using importance measures. Five major events were identified with an emphasis on operational and socio-economic issues related to distribution and delivery of water. Comparing the results of risk assessment with the mathematical model reveals that the latter’s failure probability will be less than the system’s FTA due to non-consideration of some important factors.     

Performance Evaluation of Agricultural Surface Water Distribution Systems Based on Water-food-energy Nexus and Using AHP-Entropy-WASPAS Technique

This study aimed to quantitatively evaluate the performance of practical alternatives in modernization projects of water distribution in irrigation networks based on the water-food-energy nexus using the AHP-Entropy-WASPAS technique. Three methods of improved manual operation, decentralized automatic operation, and centralized automatic operation were developed under normal and water shortage operation scenarios and modeling the current status of water distribution in the main canal of the Rudasht irrigation network as a case study. Water-based, energy-based and food-based indicators were used to develop the nexus evaluation framework. The results showed that the average values of the water-food-energy nexus index in the manual operation method were estimated at 0.49 and 0.16 under normal and water shortage operation scenarios, respectively. These average values were estimated at 0.53 and 0.17 under normal and water shortage operation scenarios, respectively, by improving the method to the improved manual operation method. The decentralized automatic operation method improved these average values to 0.82 and 0.39 under normal and water shortage operation scenarios. Finally, using the centralized automatic operation method, these average values were 0.94 and 0.35 under normal and water shortage operation scenarios. Since the downstream secondary off-takes of the irrigation network receive no water even by upgrading the surface water distribution system to the decentralized automatic operation method under the water shortage operation scenario, it can be said that the performance of the centralized automatic operation method is more efficient than the decentralized automatic operation method due to the fair and uniform distribution of water in both normal and water shortage scenarios.

     

Providing a Reliable Water Level Control in Main Canals under Significant Inflow Fluctuations at Drought Periods within Canal Automation

Supplying agricultural water has recently become more challenging due to increase in drought frequency in semi-arid regions. Unreliable supply water resources lead to inflow fluctuations in irrigation districts locating at the downstream part of the rivers. The conventional operation cannot handle the fluctuations effectively; therefore, water delivery to the off-takes is seriously compromised. In this study, the capability of decentralized Proportional-Integral (PI) and centralized Linear Quadratic Regulators (LQR) controllers were investigated to alleviate the effects of severe inflow fluctuations in the main irrigation canals operation. In this regard, the Roodasht North Branch (RNB) main canal, located in the central Iran, consisting of 24 reaches was selected as a test case. Mathematical operational model of the canal was coupled with the designed PI and LQR controllers. The performance of the controllers were evaluated under the normal and severe unpredictable inflow fluctuations scenarios. The results showed that the PI configuration was able to successfully deal with and handle the normal inflow fluctuations. Water delivery to the upstream and middle canal reaches was regulated reasonably well, while the downstream canal reaches were still slightly suffering from the inflow fluctuation. The LQR controller showed an outstanding performance in dealing with both normal and severe inflow fluctuation test scenarios. The controller optimally adjusted the gates accordingly to manage even the sever inflow fluctuation. The results of this study demonstrate capabilities of the control configurations in managing the inflow fluctuations in main canals which can help moving towards the urgent needs for improving the current canal operational performance.     

Minimization of Operational and Seepage Losses in Agricultural Water Distribution Systems Using the Ant Colony Optimization

The present study’s main objective is to simultaneously minimize operational and seepage losses in agricultural water distribution systems, relying on the Ant Colony Optimization (ACO). For this purpose, the following arrangements were made: i) Hydraulic flow simulation of the distribution systems was conducted by developing an Integrator Delay (ID) model using MATLAB and appraisal performance of the water distribution system, ii) The seepage simulation alongside the system employing a calibrated and validated estimation equation, iii) developing the ACO model to minimize operational and seepage losses within the agricultural water distribution Systems. Two single-objective and one multi-objective functions were considered to minimize seepage loss, operational loss, and both loss components simultaneously. The Moghan irrigation water distribution system, Iran, was selected as the case study. Optimization results revealed that the first through third objective functions managed to reduce the total losses in the Moghan water distribution systems by, respectively, 0.39, 3.1, and 4% compared to the existing conditions. A comparison with the optimization results from LINGO, a nonlinear optimization model, was suggestive of the advantages of the ACO in terms of the optimal result and optimization time. The proposed method can be used as a practical measure to improve water productivity within the scale of agricultural water distribution systems by improving the manual operating system’s performance in the status quo.