Modernization of Electric Power Systems

Flexible Alternating Current Transmission System (FACTS) are power electronic-based devices connected to the electrical systems in series or shunt or combined series-shunt. FACTS can control the system parameters such as voltage magnitude and phase angles of the buses, the power flow through the transmission line, and the transmission lines’ impedance. FACTS can be categorized based on their power electronic devices into (1) variable impedance type (TCSC, TCPSC, SVC) and (2) Voltage Source Converter (VSC) type (SSSC, UPFC, IPFC, GUPFC, STATCOM). Modeling the FACTS devices into power flow solutions is challenging due to the required modifications into the power flow solution, including the alterations in the Y-Bus matrix, Jacobian and the corrections matrix, and the mismatches matrix. This chapter aims to introduce a comprehensive survey of modeling methods of the combined series shunt VSC-based FACTS devices into power flow solution, including the power injection model, the extended model, the developed model, the indirect model, the simplified model, the decoupled model, and the π injection model. In addition to that, a comparison is presented related to the merits and the shortcomings of these methods.

Short circuit fault currents are increasing due to growing demand for electricity and high complexity in power systems. Because the fault currents reach the highest value, which the breakers are unable to restrict, the electrical grid’s security is in jeopardy. By entering a limiting impedance into a transmission line in series, these impedances restrict the rising amounts of fault currents to levels that are acceptable. Saturated core fault current limiters (SCFCLs) are a pivotal tool for limiting fault currents rise in power networks that have good performance characteristics. In a normal condition, these limiters have slight effects on the system and can effectively limit short-circuit currents when they occur. In this chapter, various structures of SCFCLs with different arrangements of ac windings and dc windings are presented, and the currents passing through the FCLs under the normal and faulty system conditions are assessed and compared. The flux density in various regions of the core in different arrangements has been investigated as well, and the desired analyses have been performed. Simulation will be presented based on COMSOL Multiphysics 5.4, a finite element software package which can provide a precious assessment to compare these protective devices with different configurations.

Protection art is to detect and isolate the various faults by using the dedicated protection relay. Over Current Relays (OCRs) in radial networks and Directional OCRs (DOCRs) in meshed networks are implemented as the main protection against phase faults besides Earth Fault Relays (EFRs) against earth faults. The coordination task for these protection relays assurances, the correct consecutive operation for the minimum network outage due to a particular fault scenario. The coordination process is defined by two relays: the main one that detects and operates first, and the backup one that operates if the main fails after a specified value called Coordination Time Margin (CTM). Moreover, this process can be optimized by minimizing the protective relays operating time as the Objective Function (OF) besides attaining set of constraints. The criterion of this optimization approach relies on extracting the best settings of these relays (i.e., current pickup, time dial, and/or the characteristic type). Previously, this problem is solved using trial and error and curve fitting methodologies. Though, conventional optimization techniques have been proposed for feasible results like Linear Programming (LP) and Non-LP (NLP). However, and due to the amelioration in computer science, this problem is mitigated in terms of computational errors and time consumption by solving it using the artificial intelligence techniques. Various optimizers with diverse nature have been tackled for resolving the DOCRs coordination problem. In this context, the performance of swarm-based, physics-inspired-based, evolutionary-based, and nature-based algorithms is interrogated. Particle Swarm Optimizer (PSO), Genetic Algorithm (GA), Water Cycle Approach, and Slime Mould Optimizer have proven their efficacy for solving this highly constrained optimization dilemma. Afterwards, some researchers augment this model by developing novel optimization frameworks to define the optimal settings of the protection relays. In this regard, the hybrid coordination between DOCRs and impedance relays in transmission networks is investigated to enhance the protection scheme. In addition, the coordination problem is resolved due to the permeation of Distributed Generators (DGs) that results in variations in fault current level. Depending on network topology and operating conditions, adaptive protection scheme is entrenched for readjusting the relay settings. Eventually, some ideas, metaheuristic algorithms, and various formulations that have been dedicated for solving this problem are examined later in this book chapter in detail. It is worth mentioning that the progress in this optimization problem is not over yet and it is still a challenging task for adapting new novel OFs with additional constraints. Many test cases with various scenarios under different complexities are demonstrated and analyzed along the body of this proposed chapter.

Improving the performance of distribution networks is a primary target for power system operators. Besides, energy resource limitations and cost-effective distribution of electricity to the consumers encourage engineers, distribution system operators, and researchers to increase the efficiency of electric power distribution systems. Fortunately, many technologies can effectively make such improvements. Active and reactive power compensators such as distributed generators (DGs) and shunt capacitor banks (SCBs) are examples of compensators that can effectively make such improvements in modern radial distribution systems (RDSs), in addition to using recent techniques such as energy storage technologies. Voltage regulators (VRs) can also help these compensators function better in a much more effective techno-economic manner in RDSs, enhance voltage profiles and load stability, and reduce voltage deviations from acceptable values. Unfortunately, rising project investment may result if uneconomic facilities or expensive technologies are used to reduce electric losses significantly. Therefore, economic considerations related to the installed equipment in the networks should be considered. In this regard, the well-known whale optimization algorithm (WOA) is applied in this work to allocate DGs, SCBs, and VRs in a realistic 37-bus distribution system to minimize power losses while conforming with several linear and nonlinear constraints. A cost-benefit analysis of the optimization problem is made in terms of – investment and running costs of the compensators used; saving gained from the power loss reduction, and benefits from decreasing the power to be purchased from the grid; reducing voltage deviations and overloading; and enhancing voltage stability (VS). Three loading scenarios are considered in this work – light, shoulder, and peak levels of load demand. The numerical findings obtained show a noteworthy techno-economic improvement of the quality of power (QoP) performance level of the RDS and approve the efficiency and economic benefits of the proposed solutions compared to other solutions in the literature.

Significant research is being carried out on grid integration and issues of renewable energy sources (RES). The RES are extensively spread geographically and have limited capacity. The interfacing of RES and distribution static synchronous compensator (D-STATCOM) devices into radial distribution networks (RDN) requires a special handling on distribution power flow algorithms (DPFAs). Depending on the output power characteristics and its control, the modeling of RES is either PQ or PV bus, and the modeling of D-STATCOM is PQ bus for DPFAs. DPFA is a highly significant tool, particularly in the areas of distribution automation and optimization. Therefore, PFAs should be fast and accurate. In this chapter, the radial topological characteristic of the distribution networks is exploited to split the RDN into several segments. The information of these newly split segments are stored in the matrix of branch numbers (BRNM) and matrix of bus numbers (BNM). These matrices simplify the execution of DPFA. The proposed DPFA employ fundamentals in electrical circuits and is easy to understand. In this chapter, both PQ and PV models are efficiently incorporated into the proposed DPFA. The proposed DPFA can change its operation manner from PQ mode to PV mode and vice versa with the presence of several RES. The proposed DPFA’s precision is examined on 15-bus RDN and it is identified as correct against the work that has already been published. Various tests are performed on 15-bus and 69-bus RDN for impacts of multiple interfacing of RES and D-STATCOM devices. Test results shows that the interfacings of these devices can enhance the voltage profile and lessen the real power (P) loss and reactive power (Q) loss.

This chapter proposes a convex optimization model for stationary-state hosting capacity assessment in power distribution networks and microgrids. The proposed model is based on a linear formulation of the power flow equations on the complex domain that transforms nonlinear and non-convex constraints into affine forms, including exponential loads models. Two metrics are considered, namely proportional growth and maximum hypervolume. Numerical experiments applied to a low-power distribution system (both radial and meshed) demonstrate the use of the model. The proposed method also considers the effects of topology and reactive power management, resulting in a simple, fast, and accurate model. A comparison between the proposed approximation and conventional power flow for different scenarios complements the analysis and demonstrates the accuracy and robustness of the proposed approach.

Power quality (PQ), defined as a problem with steady supply voltage, current, or frequency, has increased attention in power engineering in recent years. The main reason for the growing interest in this problem is caused by a higher penetration of renewable power sources as well as the usage of different power electronics components. This chapter describes the main PQ issues in current power systems. In pursuit of that goal, a case study is given, with experimental results of the PQ measurements performed on the first photovoltaic system in Bosnia and Herzegovina during various periods and weather conditions.

In its broadest sense, the particular topology of radial distribution networks (RDNs), high resistance-to-reactance (R/X) ratios, the misuse and exploitation of these systems, specifically with increased variable renewable energy sources (VRESs) that are added in these networks, increased load demands, and nontechnical losses are leading to a significant decrease in voltage magnitude of some buses in these networks with a rapid increase in energy losses and voltage drop values. Therefore, improving the power quality (PQ) and reliability performance levels of RDNs has become an authoritative goal for system operators. Accordingly, in this work, smart coordination of distributed generation units (DGUs) and shunt capacitors (SCs) in Egyptian RDN is presented to increase resiliency and enhance the reliability of this system. The sine-cosine optimization (SCO) algorithm was used to solve the formulated multi-objective (MO) constrained optimization problem. The various objectives presented in this work aim to reduce the operating costs of the system and investment expenses for the proposed coordinated DGUs and SCs in the system while maximizing the benefits (savings) of reducing energy loss and contracted power from the utility, in addition to improving the voltage profile and loading capacity of the system. Three different loading levels were investigated in this work. Various linear and nonlinear constraints were taken into account. A weighted sum approach using the well-known analytic hierarchy process (AHP) was used to transform the MO problem into a normalized single-objective (NSO) optimization problem. The results obtained show a significant techno-economic enhancement of the PQ performance level of the system. Besides, SCO’s results outperformed the results attained using other considered algorithms presented in the literature to solve the problem.

Power Quality (PQ) refers to the power consumption capacity of electrical devices. Many concerns about PQ, such as electrical harmonics content, poor power factor, voltage instability, and imbalance, affect electrical equipment efficiency. In addition, nonlinear loads affect the sinusoidal waveform of the current and result in the flow of harmonic currents in the distribution networks that may interfere with communication systems and other equipment. When some or all of these problems exist with electrical distribution networks, higher energy usage, maintenance costs, and equipment instability or failure will begin. This chapter represents distribution network optimization by optimal allocation of distributed generators to enhance voltage stability, reduce system losses, harmonic distortion, and voltage unbalance in distribution networks when nonlinear loads exist. In such an optimization process, the newly published Bald Eagle Search (BES) optimization algorithm is used and implemented to achieve each objective of the problem. Moreover, multi-objective optimization is considered to optimize different objective functions simultaneously. The methodology is applied to different standard small and extensive distribution networks, such as the unbalanced 13-bus and 37-bus systems and balanced 69-bus systems. The obtained results show the BES algorithm’s effectiveness in reducing the THD and losses of the systems and improving the voltage profile and PQ of distribution systems.

This chapter aims to contribute by presenting the Knowledge Discovery in Database (KDD) process as a technical tool to identify, at the Common Coupling Point (CCP) of a computer factory, the main sources of harmonic voltage distortion (THDv). In the methodology utilization, computational intelligence and data mining techniques were used in the analysis of the data collected by power quality analyzers, strategically installed in the PCC of the factory and in the main loads, with the purpose of identifying the contribution of each load in the harmonic distortion of voltage in the PCC. The proposed methodology begins with the analysis of the loads and the plant layout for the installation of the power quality analyzers, application of the KDD process that includes the procedures of data collection, selection, cleaning, integration, transformation, reduction, mining, interpretation, and evaluation. In the data mining stage, the Decision Tree and Naïve Bayes techniques were applied, with the testing of several algorithms to find the ones that presented the best results for this type of application and if this methodology has applicability to identify the main sources of harmonic distortion. As a contribution, different data balancing indexes, training, and testing in different scenarios were also tested.

Owing to the industrial development of the Amazon Region, mainly in the Industrial Pole of Manaus, and the consequent increase in the need for energy generation, of which in this region more than 90% is supplied by thermoelectric power plants (TPPs), it became necessary to use of artificial intelligence techniques that provide TPP managers with support in decision making in defining the optimal output power of each generator, contemplating the reduction of costs and the indices of pollution in the atmosphere. The economic dispatch, or optimal dispatch, is one of the oldest and most important tasks in the management of electric power plants, and, owing to the severe impacts caused to the environment, this problem was extended to the optimization of the economic and environmental dispatch (EED). In this chapter it is proposed to present a new solution to the old EED optimization problem implemented by heuristic methods (nondominated genetic algorithm of classification, NSGA-II and NSGA-III), considering the shutdown of the generators with the highest operating cost, with the consequent reduction in fuel costs. The incremental cost and transmission losses are weighted in determining output optimal power values of the generators, meeting the restrictions of balance between total power and demand, losses, and reducing total fuel consumption, emissions, and even improving the efficiency of the TPPs. The proposed solution includes the following contributions: the shutdown of engines with higher fuel costs, reducing total costs, allowing for predictive maintenance on these engines; to determine optimal values of rated power in several scenarios in the TPP, considering variations in power generation and in the reduction of NOx and CO₂ emissions. To analyze the results of the proposed method, a set of generators from a TPP in Amazonas was used as parameters for the case study. The results presented in the proposal, from the analysis of several practical examples, show significant reductions in fuel costs and in pollution indices.

This chapter introduces a stochastic multi-period transmission expansion planning (SMTEP) model that considers the power system’s uncertainties and reliability constraints. Renewable generation sources (RGSs) are widely used in power systems. RGSs have a stochastic behavior that menaces the power system’s reliability and may result in partial or complete blackouts. The inclusion of N-1 security in SMTEP is also essential to ensure the continuity of electricity supply to loads under the worst conditions. The problem is formulated as a mixed-integer non-linear optimization problem. The whale optimization algorithm (WOA) is applied to solve the SMTEP problem. A reduction technique and an acceleration scheme are incorporated with the WOA to accelerate the convergence to the optimal solution and decrease computation time. The results of testing the WOA on a benchmark system and a realistic network show its efficiency and superiority, compared to other well-established algorithms, in terms of convergence time and quality of solutions. Further, case studies demonstrate the effectiveness of the suggested model in improving the power system’s reliability.

In electrical grids, there is always a mismatch between generation and electrical load demand. It is a big challenge to mitigate this mismatch. There are many efforts that try to suppress the mismatch, among which supply/demand sides scheduling. In addition, this is done by using deterministic and probabilistic techniques and methodologies to decrease the estimation error in both sides. However, every electrical grid suffers from some sort of mismatch. The supply/demand mismatch can be positive type: supply exceeds the demand; or negative type: supply is deficit to cover the demand [1].

One of the significant aspects in microgrids (MG) optimization frameworks is to cope with stochastic parameters of renewable energy sources (RESs), market pricing, and electrical demand. Thus, the study of market price volatility, the intermittency of the RESs, and electrical demand alteration has received a great concentration in the recent period. This chapter introduces a stochastic optimization method for scheduling the energy of MG considering random demand, the intermittent nature of RES, and the volatile market price. The optimal daily energy scheduling is employed to attain the OFs, such as the minimum operating cost of the MG over a 24-h horizon or minimizing the pollutant emissions over a 24-h horizon. In addition, this chapter introduces multi-OFs to reduce the overall operating cost and the pollutant emission of MGs over a 24-h while taking into account different operational restrictions. In the proposed stochastic structure, uncertainty modeling of the RESs, electrical loads, and market pricing is modeled by fuzzy C-means. The general algebraic modeling system (GAMS) can be employed for modeling the optimization issue expressed in the chapter.

The energy demand is rising significantly with increase in population and human development index of the country. The reduction of the energy extracted from the conventional energy sources provides more opportunities towards the renewable energy sources (RES). Current trends suggest the power generation at the end user using RES like solar photovoltaic energy system (PVES), wind energy conversion system (WECS), energy storage (ES) serves a microgrid to cater to the load demand. The incorporation of energy storage will make the renewable system reliable, and it helps the voltage maintenance at DC bus. The proposed work focuses on hybrid renewable power generating system. The photovoltaic energy system and battery energy storage system (BESS) are used to form a hybrid renewable power generation system fed to standalone DC load. The DC load is taken into the consideration to avoid the conversion stage from DC to AC which improves the conversion efficiency and eliminates the cost of inverter. The hybrid renewable energy system is used to form a DC microgrid that caters to the load demand of a small community.

The proposed work comprises of simulation analysis for the PVES and the BESS which forms DC microgrid. The DC loads are connected at DC bus. The PVES and the BESS are interfaced with the DC bus via quasi-double boost converter and bidirectional DC to DC converter, respectively. The quasi-double boost converter is connected with PVES and performs DC to DC conversion. To enhance the effectiveness of the PVES, the Perturb and Observed (P&O)-based MPPT algorithm is used for the extraction of the maximum power from Solar PV array. To make the system reliable of the system, the battery energy storage is interfaced at DC bus via DC to DC bidirectional converter. Generally, fixed gain proportional integral (PI) controllers are incorporated with RES and battery energy systems to regulate the voltage. The optimal tuning of these controllers is essential for the explicit conditions. The change in climatic conditions is major concerned in renewable energy sources in present case in PVES. The voltage regulation should be maintained and that too with change in solar insolation and temperature over a wide range of operation. In order to keep the voltage regulation within limits, the gains of the controller must be scheduled dynamically. The proposed work typifies a hybrid PVES and BESS where gain scheduled proportional integral (GSPI) controllers are accommodated for the synchronized control for PVES and BESS for DC microgrid applications. The simulation results are compared for fixed gain controller and GSPI controller. The mathematical model of PVES and BESS is presented along with the control system of GSPI controller is presented. The performance of PVES and BESS hybrid system is investigated for permutations and combinations of change in climatic conditions and change in DC load at DC bus.

High-impedance faults (HIFs) occur in distribution networks due to contact between an electrified conductor and objects such as trees or conductor falling to the ground. In these cases, a small current flows in the conductor due to the low voltage in the network and the high impedance between the ground and the conductor. According to the conducted tests, the fault current amplitude can take a value between less than 1 A and 100 A. Protective devices are generally used to protect grid equipment, such as transformers or overhead and underground power lines, against fault currents exceeding the permissible values for this equipment. However, an HIF falls within these values; hence, conventional protections in distribution networks, such as overcurrent protection or ground fault protection, cannot detect this fault and do not damage this equipment. The main reason for detecting HIFs in traditional power grids is to prevent harm to persons as a result of electrocution. Moreover, the electric arc occurring due to the fault, especially when the conductor breaks and falls onto a tree, can result in fire. Early detection of this fault can prevent potential outages and reduce outage duration in distribution networks. Since the nature of an HIF depends on numerous parameters, such as feeder structure, ground structure, and humidity, the disturbance in the current and voltage waves of the feeder are diverse and unpredictable. This has prompted many researchers to propose various solutions for detecting HIFs and has resulted in the production of relays for this purpose. The present chapter book involves a thorough investigation of HIF detection schemes along with their advantages and disadvantages.

Lighting is a prerequisite of any facility, and it has an effect on people’s regular lives. This accounts for a significant portion of overall energy consumption in residential, commercial, and industrial sectors. Lighting accounts for 19% of worldwide power use and 25–30% of household energy consumption. Light quality and quantity have an effect on health, leisure, security, life style, performance, as well as the economy. Lighting has accounted for a considerable amount of many nations’’ energy budgets. As a result, especially in the household sector, it becomes an important area for energy saving. Lighting management systems are crucial for maximizing energy savings.

One of the simplest ways to reduce energy bills is to switch to energy-efficient lighting. The term “efficient” refers to how much light is emitted for a given amount of power input. The energy efficiency of the lighting loads provide the adequate illumination output of the lighting devices for the purpose for which the product was developed while using the least amount of energy possible. Lighting that is energy efficient conserves energy while providing a high degree of illumination quality and quantity. Traditional lights not only consume a lot of electricity, but they also waste a lot of it by producing heat rather than light (for instance 90% of consumed energy in case of incandescent lamps). Energy-efficient lighting is required to minimize energy consumption, which lowers electricity costs, saves power rather than wasting it via losses, reduces carbon emissions (since typical lights release CO₂), and reduces peak load demand.

For lighting systems, conventional lights like incandescent lamps are substituted with energy-saving lamps such as fluorescent lamps, CFL lamps, and LED lamps. Lighting controls such as timers, motion sensors, and ultrasonic sensors are included in an energy efficient lighting system. These sensors detect the presence of humans and other living animals, as well as instructions for remote operation. It employs electronic circuitry to achieve light dimming when needed. GSM/SCADA/GPS-based centralized systems often easily and accurately track and automate the lighting system to save electricity. These energy-saving measures will help outdoor lighting, residential building’s in-house lighting, and indoor lighting for industrial structures. Aforesaid strategies not only minimize energy usage, but they also improve lighting efficiency, improve safety including employee well-being, and tackle climate change.

The unavailability of equipment in thermoelectric plants for any reason becomes a risk for the entrepreneur, who as a consequence bears even greater losses with the high cost of machines stopped, in addition to the penalties sanctioned and provided by law. Based on this assumption, the maintenance programs are methodologies that aim to contribute with techniques and tools to mitigate this problem. However, only the use of maintenance programs is not enough; thus, this research aims to develop a computational model capable of predicting the Key Performance Indicator Reliability, with the purpose of indicating the probability of the equipment to operate in a pre-defined space of time, having as object of study a group of internal combustion machines of Thermoelectric Plants. Thus, the research meets the objectives of cataloging the significant variables for the prediction model; analyze two ANN training algorithms (Levenberg-Marquardt and Bayesian Regularization) considering the supervised learning approach, where the amount of neurons, hidden layers and activation functions are requirements for the network performance; Developing the prediction model for the reliability of the motor group, where the training algorithms are validated using the best model stopping criterion, finding the best network performance based on Mean Squared Error (MSE), Root Mean Square Error (RMSE), Linear Regression, and Best Model stopping criterion; and finally, simulating the cataloged failure data in order to analyze the technical state of the motor group with the best model. The innovation of the research is characterized by the computational methods of data processing with optimization methods such as Levenberg-Marquardt for the search of the local minimum and faster convergence of the ANN and Bayesian Regularization which is a vector machine focused on machine learning and pattern recognition, characterizing the use of artificial intelligence techniques to predict the reliability in days and months. In addition, it is used for predictive maintenance indicators such as: Mean Time Between Failures (MTBF), Mean Time To Repair (MTTR), Availability, and Reliability. To analyze the results of this research, a set of 20 load generation units was used as parameters for investigating the frequency of failures, the two optimization algorithms were applied, with a combination between the activation functions: in Sigmoid, Linear, and Hyperbolic Tangent, the research results show that the two techniques present 100% correlation between the output and simulated variables, characterizing the efficiency in predicting in days.