K. Srinivasa Raju

River Flow Forecasting using Recurrent Neural Networks

Forecasting a hydrologic time series has been one of the most complicated tasks owing to the wide range of data, the uncertainties in the parameters influencing the time series and also due to the non availability of adequate data. Recently, Artificial Neural Networks (ANNs) have become quite popular in time series forecasting in various fields. This paper demonstrates the use of ANNs to forecast monthly river flows. Two different networks, namely the feed forward network and the recurrent neural network, have been chosen. The feed forward network is trained using the conventional back propagation algorithm with many improvements and the recurrent neural network is trained using the method of ordered partial derivatives. The selection of architecture and the training procedure for both the networks are presented. The selected ANN models were used to train and forecast the monthly flows of a river in India, with a catchment area of 5189 km2 up to the gauging site. The trained networks are used for both single step ahead and multiple step ahead forecasting. A comparative study of both networks indicates that the recurrent neural networks performed better than the feed forward networks. In addition, the size of the architecture and the training time required were less for the recurrent neural networks. The recurrent neural network gave better results for both single step ahead and multiple step ahead forecasting. Hence recurrent neural networks are recommended as a tool for river flow forecasting.

Artificial neural networks and multicriterion analysis for sustainable irrigation planning

The objective of the present paper is to select the best compromise irrigation planning strategy for the case study of Jayakwadi irrigation project, Maharashtra, India. Four-phase methodology is employed. In phase 1, separate linear programming (LP) models are formulated for the three objectives, namely, net economic benefits, agricultural production and labour employment. In phase 2, nondominated (compromise) irrigation planning strategies are generated using the constraint method of multiobjective optimisation. In phase 3, Kohonen neural networks (KNN) based classification algorithm is employed to sort nondominated irrigation planning strategies into smaller groups. In phase 4, multicriterion analysis (MCA) technique, namely, Compromise Programming is applied to rank strategies obtained from phase 3. It is concluded that the above integrated methodology is effective for modeling multiobjective irrigation planning problems and the present approach can be extended to situations where number of irrigation planning strategies are even large in number.

Multiobjective fuzzy linear programming for sustainable irrigation planning: an Indian case study

Abstract Multiobjective fuzzy linear programming (MOFLP) irrigation planning model is formulated for the evaluation of management strategy for the case study of Jayakwadi irrigation project, Maharashtra, India. Three conflicting objectives net benefits, agricultural production and labour employment are considered in the irrigation planning scenario. All three criteria are to be maximised and the last two are sustainability related. All three objective functions are quantified by linear membership functions in a fuzzy multi objective framework. It is observed from MOFLP solution that net benefits, agricultural production and labour employment are 2.031×109 Rupees, 2.1186×106 tons, 3.5858×107 man-days respectively with degree of truth (λ) 0.5715. Analysis of results indicated that net benefits, agricultural production, labour employment have decreased by 4.13, 5.39 and 3.4% as compared to ideal values in the crisp linear programming (LP) model.

Integrated Sustainable Irrigation Planning with Multiobjective Fuzzy Optimization Approach

Multiobjective fuzzy methodology is applied to a case study of Khadakwasla complex irrigation project located near Pune city of Maharashtra State, India. Three objectives, namely, maximization of net benefits, crop production and labour employment are considered. Effect of reuse of wastewater on the planning scenario is also studied. Three membership functions, namely, nonlinear, hyperbolic and exponential are analyzed for multiobjective fuzzy optimization. In the present study, objective functions are considered as fuzzy in nature whereas inflows are considered as dependable. It is concluded that exponential and hyperbolic membership functions provided similar cropping pattern for most of the situations whereas nonlinear membership functions provided different cropping pattern. However, in all the three cases, irrigation intensities are more than the existing irrigation intensity.

Ranking of CMIP5-based global climate models for India using compromise programming

Thirty-six Coupled Model Intercomparison Project-5-based global climate models (GCMs) are explored to evaluate the performance of maximum (Tmax) and minimum (Tmin) temperature simulations for India covering 40 grid points. Three performance indicators used for evaluating GCMs are correlation coefficient (CC), normalised root mean square error (NRMSE) and skill score (SS). Entropy method is applied to compute the weights of the three indicators employed. However, equal weights are also considered as part of sensitivity analysis studies. Compromise programming (CP), a distance-based decision-making technique, is employed to rank the GCMs. Group decision-making approach is used to aggregate the ranking patterns obtained for individual grid points. A simple but effective ensemble approach is also suggested.

An Indian Case Study

Selection of the best irrigation subsystem is evaluated in a multiobjective framework for a case study of Sri Ram Sagar Project, Andhra Pradesh, India. Two Multicriterion Decision Making (MCDM) methods, namely, PROMETHEE-2 and EXPROM-2 are applied for evaluation. Eight performance criteria, namely, farm development works, environmental conservation, timely supply of inputs, conjunctive use of water resources, productivity, farmers’ participation, economic impact and social impact are evaluated for the five irrigation subsystems. Weighting of the performance criteria is calculated by Analytic Hierarchy Process (AHP). Group decision making methodology is incorporated by additive ranking analysis. Results indicated that same irrigation subsystem is found to be the best by both methods.

Ranking Irrigation Management Alternatives by Multicriterion Analysis

Selection of the best alternative plan in irrigation development strategies is examined in the multiobjective context. Three conflicting objectives-net benefits, agricultural production and labour employment-are considered. The procedure combines multiobjective optimization, cluster analysis, and multicriterion decision making (MCDM) methods. Five MCDM methods, ELECTRE-2, PROMETHEE-2, Analytic Hierarchy Process (AHP), Compromise Programming (CP) and Multicriterion QAnalysis (MCQA-2) are used in the evaluation. Spearman rank correlation test is employed to assess the correlation between them. The methodology resulted in the selection of the best alternative plan when applied to a case study of the Sri Ram Sagar Project, Andhra Pradesh, India.

Fuzzy Multiobjective Irrigation Planning Using Particle Swarm Optimization

AbstractParticle swarm optimization (PSO) technique is applied in multiobjective irrigation planning environment, to the case study of Khadakwasla complex, India. The project consists of Khadakwasla irrigation project, Janai Sirsai lift irrigation scheme, and Purandar lift irrigation scheme. Objectives considered are net benefits, crop production, and labor employment on an annual basis. Uncertainty in the three objectives is tackled by a fuzzy approach and through hyperbolic and exponential membership functions. An irrigation planning scenario of 75% dependable inflow with groundwater and treated wastewater is analyzed (termed as 75WGWHM) and is the basis for formulating the multiobjective problem. Two additional scenarios, S1 with 75% dependable inflow without groundwater using a hyperbolic membership function and S2 with 75% dependable inflow with groundwater using an exponential membership function, are also explored and compared with 75WGWHM. It is observed from the results that hyperbolic membership...

Multi-objective differential evolution application to irrigation planning

In the present study, applicability of Multi-objective Differential Evolution (MODE) in irrigation planning perspective is demonstrated through a case study of Mahi Bajaj Sagar Project, Rajasthan, India. Three objectives, namely, net benefits, agricultural production and labour employment are analysed in the multi-objective environment. Non-dominated alternatives generated by MODE are reduced to a manageable subset with the help of K-means cluster analysis for effective decision making. Optimal number of groups is determined based on the cluster validation indices, namely, Davies–Bouldin and Dunns. It is concluded that selection of suitable parameters is necessary for effective implementation of above methodologies in real-world planning situations.

Multiobjective Differential Evolution and Differential Evolution for Irrigation Planning

The present paper discusses the applicability of Multiobjective Differential Evolution (MODE) and single objective Differential Evolution (DE) to a case study of Mahi Bajaj Sagar Project, Rajasthan, India. Three objectives, namely, net benefits, agricultural production and labour employment are analyzed in the multiobjective framework using MODE. Four variations (strategies) of Differential Evolution, namely, DE/rand/1/bin, DE/rand/1/exp, DE/best/1/bin and DE/best/1/exp are explored. Population size, crossover and mutation probabilities and number of generations are the parameters that are required as input to MODE. In order to have a better insight on the performance of the strategies, DE in single objective framework is also employed with the same four strategies for all the three objectives with the above settings. In addition, a comparative analysis is also made for all the four strategies in both single and multiobjective framework.