Optimal Generation Scheduling of Pumped Storage Hydrothermal System with Solar and Wind Energy Sources

Abstract

Electricity has been playing an important role in the people living standards and the development of various economic sectors like agriculture, industry, commerce, etc. In the past decade, thermal, oil, gas, and nuclear plants, have been the primary sources of energy generation. The emissions produced by the thermal plants have been one of the main reasons of renewable energy sources (RESs) in the power systems energy generation. From the past decade, large hydropower plants have been constructed to fulfil the power demand with reducing cost and emissions. The outcome of this development is the alteration of the river's natural flow, degradation of water bodies, and pristine conditions. The pumped-storage unit (PSU) is one of the small hydroelectric units with fast response during demand variations and cost minimization of energy generation. The integration of RESs with pumped storage hydrothermal system (PSHTS) becomes vital with respect to the present energy scenario of the country. The importance lies in the acknowledgment of the RESs, which are abundantly available in the country. There are several RESs, such as wind, solar, hydro, etc. have been integrated with the conventional power plants to reduce the emission pollutants and the power generation cost. However, the power generation of RESs are uncertain due to weather conditions. One of the most sustainable ways to mitigate variations in the generation and demand has been the incorporation of the energy storage systems (ESS). The ESS may improve power management and facilitates a robust energy balance within the integrated energy system (IES). The larger involvement of RESs into the existing power system creates a challenge for optimum generation schedule.The optimal generation schedule of an IES is a nonlinear, mixed-variable, constrained optimization problem. The various conventional search techniques have been applied to resolve the optimal generation scheduling problem with number of presumptions to make optimization problem more tractable and simple. In order to overcome the limitations of conventional search techniques, researchers have global optimization techniques, which do not require any derivative information. The conventional methods are more exploitative in nature and global search techniques are explorative. Therefore, a space for an optimization technique exists that maintains a fine balance between exploration and exploitation phase to reduce the convergence speed of finding the optimum solution. In multi-objective optimization problems (MOOP), the objective functions are conflicting in nature and multi-objective optimization techniques are applied to find most satisfying non-dominated solution in the presence of trade-offs between two or more conflicting objectives. The MOOP solution methodologies have been categories in two ways: the method employed to obtain Pareto optimal solutions, the ways and mean to interact with decision maker(DM) and the information available to the DM. The classical and multi-objective global optimization techniques have been implemented to get a trade-off solution to theinteractive MOOP. The multi-objective global optimization algorithms have ability to search and obtain the set of Pareto optimal solutions in a single run. Further, decision-making is one of the vital aspects of MOOP. Among the decision-making methodologies, the interactive methods are recognized to achieve most satisfactory results because few non-dominated solutions have to be evaluated. In this work, an optimization technique is proposed combining modified crisscross search particle swarm optimization (MCS-PSO) technique and modified binary PSO (MB-PSO) technique. The proposed optimization technique (POT) has been applied to solve economic, emission and multi-objective scheduling problems for standard hydrothermal system (HTS), PSHTS, and IES. Further, the results have been compared with state-of-art optimization techniques. For MOOP, fuzzy based surrogate worth trade-off (SWT) method has been applied to club different objectives and cardinal priority ranking method is used to select the best non-dominated solution. The brief discussion regarding each chapter of the thesis has been presented. In chapter 1, the requisite background review of hydro, thermal, pumped storage units and RES have been presented. The significant contributions of several researchers related to optimum generation scheduling of HTS, PSHTS, IES have been reviewed. In Chapter 2, the crisscross search PSO (CSPSO) and improved binary PSO (IBPSO) technique have been implemented for the economic, emission, and the multi-objective generation scheduling problem of the PSHTS. Numerical analysis has been conducted on two test systems (PSHTS-I, and PSHTS-II) with two cases under each test system. The results reveal that the implemented technique is superior as compared to other state-of-art techniques. Further, the impact of the PSU has been analysed in terms cost and emission pollutant. In order to check the statistical performance of the implemented technique, two-sample student t-test has been applied and the performance of the CSPSO-IBPSO technique found satisfactory. In the successive chapter, an optimization technique has been proposed for resolving the economic, emission and multi-objective PSHTS scheduling problems. The proposed optimization technique (POT) combines MCS-PSO, and MB-PSO for continuous and binary decision variables. Two test systems (HTS-I, and PSHTS-III) have been undertaken and results reveal that the POT is superior as compared to other state-of-art techniques. To check the statistical performance of the POT, two-sample student t-test has been applied and the performance of the POT found satisfactory. In addition, parameter sensitivity has been implemented by perturbing each parameter successively. It has been concluded that the parameter perturbation has very little effect on the result obtained from the POT. In the ensuing chapter, the PSHTS problem formulation has been extended to the IES scheduling problem considering uncertainties of RESs. The economic, emission and multi-objective scheduling problem of IES has been addressed, in which cost, and emission has been optimized. The POT has been applied to search the optimal results. Two test systems (IES-I, and IES-II) with two cases under each test system have been undertaken to examine the effectiveness of POT. Further, the impact of RES and its uncertainties on the optimum generation schedule and its impact on cost, and emission have been analysed. It is verify from results that the POT has a significant impact on the economic, emission and multi-objective scheduling of the IES. In the succeeding chapter, the multi-objective generation scheduling problem of IES incorporating RES has been resolved by implementing the POT-surrogate worth trade-off approach. Four different test systems (HTS-I, PSHTS-III, IES-I, and IES-II) have been undertaken to examine the effectiveness of POT-SWT approach. It has been concluded that the SWT approach achieves better non-dominated solutions as compared to its counterparts. Finally, the main conclusions and contributions of the thesis have been presented and suggestions for the future research in this field have been specified.