Lack of fresh water has been a major obstacle to development and flourishing in human history. Desalination provides a new vision toward fresh water production in the upcoming future. The study has proposed a simple mathematical equation and ANN models to simulate eight types of sea water RO membranes. The Artificial Neural Network (ANN) models have been developed to simulate TDS corresponding to the temperature (T,°C), flow rate (gpm) and recovery percentage. The feed data was generated by ROSA software. The model developed using a simple rational mathematical method. ANN models were trained using feed-forward back propagation algorithm with two hidden layers and various numbers of neurons in each layer. The model verification analysis proved both mathematical and ANN models to be highly accurate, reliable and practical for analyzing, designing, operating and optimizing of RO systems. The correlation coefficients (R) of 0.96 and 0.97, respectively, confirmed that the equation and ANN models resulted in this study are in good agreement with the measured data.
In this analysis, three input parameters temperature, pH and electrical conductivity were chosen due to their easy and less costly measurement technique, and a package of six models were presented for estimating the concentrations of dissolved oxygen, DO percentage, biological oxygen demand, chloride, alkalinity and total hardness. 3001 data sets (a 3001 9 8 data array) were used to training the models. The models have been tested in order to verify their prediction values, and the resulted R factor (the rate of precision) for each model equals to 0.93, 0.95, 0.77, 0.82, 0.85 and 0.92, respectively. This proves that the package can be used to estimate the concentrations of water quality parameters with accuracy close to the reality. The River data collected from 210 monitoring stations located in all over Ireland have been used. The data set covers different conditions and makes the model applicable in many different places and conditions. For development of all models, feed-forward algorithm used for training, as well as the LevenbergMarquardt and tansign(x) functions as learning and transfer functions.
The present study focuses on temperature variations during the past 21 years using data obtained from San Joaquin River (Old River Station), to calculate the rate of temperature variation. The rate of temperature change (R) is calculated by adding up the difference between each year's mean temperature and that of the previous years. According to our calculation R equals to 0.0354°C/year, which means that if the local conditions would exist, we will have 3.54°C temperature rise within the next 100 years. Using the resource we calculated mean temperature for the past 21 years, which was equal to 17.12°C, meaning that the mean temperature of the year 2100 will be around 20.5°C, which will be incredibly high. We also made an ANN model (and ran it using MATLAB) to regenerate the missing data. The model is a feed-forward network with back propagation neurons trained by the Levenberg-Marquardt algorithm, with 4 layers containing 25 neurons. After making the model and before using it, we tested the model with existing data and compared the results that showed unexpected high correlation of 99 %.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.