JWSS - Journal of Water and Soil Science

---

Rice husk, an agricultural waste, is produced about 100 million tons annually in the world and 0.5 million tons in Iran. Due to growing environmental concerns about disposal of these wastes, efforts are required to apply the wastes in industry. In this research, the mechanical properties of concrete incorporating rice husk ash (RHA) in sulfate environments (such as irrigation canals) were investigated and the increase in concrete strength was compared with control samples. In order to burn the husks at a controlled temperature to obtain a highly reactive pozzolanic RHA, a furnace was designed and built. The experiments included 405 samples of cubical (707070-mm) and cylindrical (50.8101.6-mm) concrete samples (105 samples for A, B and C treatments, respectively) which were stored in different ages (7, 28, 60, 180 days) under three different conditions (solutions of magnesium, calcium and sodium sulfates). The portions of RHA as cement replacement were 20 and 30 percents for B and C treatments. The results showed that the samples of concrete containing 20 percent RHA as cement replacement had higher compressive and tensile strengths in sulfate environments at 180 days compared with those of control concretes. The concrete samples containing RHA showed sharper gradients compared to control samples of up to 180 days under sulfate conditions. The best portion of RHA in concrete was determined to be 20 percent by weight.

---

Estimation of seepage is essential prior to lining of earth canals. In Iran such investigation has been achieved in some irrigation networks using empirical relationships derived in other countries. Estimation of water loss in canal is required in design, operation and management of water distribution systems. Water seepage may be determind by using empirical equations proposed by F.A.O. These equations are applicable for different soils and hydraulic parameters. However, the appropriate estimating equation should be determined for each region. Therefore, these equations should be calibrated for local usage and different canal vegetation conditions. In this investigation water losses in canals at the Rudast region of Isfahan were measured by inflow and outflow procedure. Different canals reaches were selected in soils of relatively heavy, medium and light textures. The density of vegetation population in canals were low, medium and high. The estimated seepage losses by different empirical equations were not corresponded to those of measured values. Therefore, by using the measured seepage at different soil textures and vegetation densities the empirical coefficients of six empirical equations of F.A.O. (Ingham, Davis and Wilson, Affengendon, Moritz, Molesworth and Yennidumia, Misra) were modified for the study region. The relationships between measured seepage and estimated seepage before and after modification of the empirical equations were determined by regression analysis. These equations estimated the seepage loss much smaller than the measured values. The regression parameters (selope, intercept, and coefficient of determination of regression equation) indicated that after modification, the Ingham and moritz equation with higher slopes (0.91, 1.01), lower intercepts (-0.096, -0.039) and higher coefficient of determination (0.96) estimated the closest seepage values to the measured values respectively. The misra equation was the next best equation for seepage estimation. The results of present investigation indicated that the modified Ingham and Moritz equations were the most appropriate ones for estimation of seepage losses at different soil textures and vegetation densities in the study region.

---

Poor performance of irrigation canals and its effect on decreasing of Agricultural water productivity require attention for their improvement. In this paper a new mathematical model is introduced which could present optimal operation considering downstream requirements of turnouts, canal inlet flow, actual constraints and real conditions of canal system. Four performance indicators of delivery including efficiency, adequacy, equity and stability are considered as an objective function in the process of optimization. Since this objective function is an implicit function of decision variables (regulation of turnouts and control structures) and hydraulic parameters, it is necessary to implement hydrodynamic model, using numerical optimization methods. SA (Simulated Annealing) technique is a numerical meta – heuristic intelligent search method which is used in combination with a hydrodynamic model (ICSS) (Irrigation Conveyance System Simulation.) for performance optimization of canal system. Theoretically it is proven that SA technique is capable of tending towards global optimum solution asymtotically. Taking short random steps in SA algorithm guarantees avoiding instability in hydrodynamic model. The developed model has been applied on E1R1 Distributary canal of Dez irrigation network for ten days. The results indicated that optimal performance improved very well in comparison with the present situation.In this model the weighting coefficients of indicators are determined using sensitivity analysis in optimization process. Consistency test on the derived coefficients shows that proposed method is appropriate. Applying weighting coefficients for performance indicators in the processes of optimization has resulted in 7 to 21 percent improvement compared to the case of equall weighting coefficients. Also, the results indicate that the developed model (ICSS-DOM) (ICSS-Delivery Optimization Model) is an efficient tool for the evaluation and optimization of irrigation canal performance, producing good and valid results in a relatively short and suitable time.

---

There are different automatic downstream control algorithms developed to increase the flexibility of the irrigation system. CARDD control logic is one of the heuristic and distance automatic downstream control systems. In this research, mathematical model of the CARDD control logic coordinated with ICSS hydrodynamic model was developed and the CARDD control logic was tested and evaluated under different situations. In order to evaluate the performance of this control algorithm, one of the canals suggested by ASCE (canal number 2) was used. The CARDD control algorithm was tested under the operational scenarios suggested by ASCE in which the control algorithm was evaluated by intense and gradual flow changes in a newly constructed (tuned) canal. Performance indicators were calculated and analysed. In the case of gradual flow changes, the maximum diversion of the water depth from the target was about 5% and was always in permitted range. In intense flow changes, the maximum diversion of the water depth from the target was about 8% and it was recovered in about 1 hour, which is a considerable time. The charts showing the variation of water depth at each turnout and the calculated performance indicators showed the satisfactory performance of CARDD control algorithm in gradual flow changes. In intense flow changes, although the maximum variation of the water depth was limited, the system response time was relatively long.

---

Modernization of irrigation canals as an operation improvement tool is essential to promote the performance of canal networks and indeed requires control systems. Proportional integral derivative (PID) algorithms have more applications than the other controllers in different places of the world, but tuning these controllers for different hydraulic conditions of canals is considered as a major problem for designing control algorithms. Since the bottom slope is one of the effective factors in the water flow dynamic behavior, in this research, the distant downstream Proportional Integral Derivative feedback control with decouplers was designed with a change in longitudinal slope in a reference canal and its performance was investigated. The canal characteristics were used to tune this controller and the system identification as a new method was applied for determining canal characteristics. SOBEK hydrodynamic model modulated with MATLAB software was used to design and run the control algorithms, and slope influence on water flow behavior, tuning controller, and coefficients of controller were investigated with different values of slope. Then, controller performance for hypothetical period of operation in various scenarios was evaluated with computation performance indices. The results showed less resonance behavior of water flow and less potential of controller in steep slope

---

---