Design, implementation, control and optimization of single stage pilot scale reverse osmosis process

Abstract

In this paper, a single-stage pilot-scale RO (Reverse Osmosis) process is considered. The process is mainly used in various chemical industries such as dye, pharmaceutical, Beverage, and so on. Initially, mathematical modeling of the process is to be done followed by linearization of the system. Here a dual loop construction with a master and a slave is used. The slave uses the conventional PID (Proportional Integral Derivative) with a reference model of the RO process and the master uses the FOPID (Fractional Order Proportional Integral Derivative) with a real time RO process. The slave's output is compared with output of the real time RO process to obtain the error which is in turn used to tune the master. The slave controller is tuned using Ziegler Nicholas method and the error criterion such as IAE (Integral Absolute Error), ISE (Integral Squared Error), ITSE (Integral Time Squared Error), ITAE (Integral Time Absolute Error) are calculated and the minimum among them was chosen as the objective function for the master loop tuning. Hence the tuning of the controller becomes a whole. Therefore two optimization techniques such as PSO (Particle Swarm Optimization) and Bacterial Foraging Optimization Algorithm (BFO) are used for the tuning of the master loop. From the calculations the ITSE was having the minimum value among the performance indices hence it was used as the objective function for the BFO and PSO. The best-tuned values will be obtained with the use of these techniques and the best among all can be considered for various industrial applications. Finally, the performance of the process is compared with both techniques and BFO outperforms the PSO from the simulations.

Novel Deep Learning Neural Networks for Breast Cancer Malignancy Estimation

“…The use of those labels can then be better decided by machine learning algorithms. The encoding of the dataset is an essential pre-processing step in supervised learning [8]. Using this method, the prediction classes can be changed for the obtained dataset from MALIGNANT and BENIGN to 1,0 and 0, respectively.…”

“…Scaling is a normalization strategy that is particularly beneficial when working with a dataset that comprises continuous features at multiple scales and you're using a model that runs in some kind of linear space (like linear regression or K-nearest neighbours). Feature scaling changes the mean and variance of the features in your dataset to zero and one [9]. This will make it easier to compare features in a linear fashion.…”

PID Control Design Using AGPSO Technique and Its Application in TITO Reverse Osmosis Desalination Plant

“…They have also reported a better settling time (ts = 147 s and ts = 25 s) of the proposed design than the ZN PID (ts = 529000 s and ts = 82 s) for permeate flux and conductivity, respectively. Recently, Guna et al (2021) [48] presented PID control designs for a single-stage pilot-scale RO process using particle swarm optimization (PSO) and bacterial foraging optimization (BFO) algorithms. To examine controller designs, they used four objective functions: integral absolute error (IAE), integral squared error (ISE), integral time absolute error (ITAE), and integral time squared error (ITSE).…”

“…The tuning of the PID controller of the TITO RO desalination plant is more complex because of the interaction effect. Literature suggests numerous natureinspired tuning algorithms (GA [46], IGA [47], PSO, and BFO [48]) for the PID controller to tune easily. + -…”