Search

The Dynamic Economic Dispatch (DED) problem consists in satisfying a certain demand for electric power among scheduled generating units over a certain interval of time while satisfying the operating constraints of these units. The consideration of the valve-point effect (VPE) makes the problem more practical but also more challenging due to the non-linear and non-smooth constraints that are required for representing the model. We present a method, based on a sequence of piecewise linear approximations, which produces a feasible solution along with a lower bound on the global solution. In this way, this deterministic approach can trade off the speed which characterizes certain heuristics that are usually used to solve the DED-VPE for a better solution and insights about the problem. The method is applied to a widely used case study and provides a lower solution objective than the best known solution to date.

The unprecedented growth of renewable energy has led to various challenges in power system operations. Demand response can provide additional flexibility to the system in order to balance the effects of the massive integration of renewable resources. This presentation focuses on the comparison of two approaches for enabling demand response, real-time pricing and priority service pricing. This analysis is centered on the impact to consumers by assessing the effects of these schemes on their comfort and bill. We use an example based on realistic electrical consumption of a household contained in the Pecan Street dataset. The example provides an end-to-end illustration of (i) how a household selects options from a priority service menu, (ii) how devices in the household are dispatched by a home energy router, and (iii) what consumer welfare losses are relative to the golden standard of real-time pricing with or without a battery.

The Dynamic Economic Dispatch (DED) problem consists in satisfying a certain demand for electric power among scheduled generating units over a certain interval of time while satisfying the operating constraints of these units. The consideration of the valve-point effect (VPE) makes the problem more practical but also more challenging due to the non-linear and non-smooth constraints that are required for representing the model. We present a method, based on a sequence of piecewise linear approximations, which produces a feasible solution along with a lower bound on the global solution. In this way, this deterministic approach can trade off the speed which characterizes certain heuristics that are usually used to solve the DED-VPE for a better solution and insights about the problem. The method is applied to a widely used case study and provides a lower solution objective than the best known solution to date.

Continuous intraday electricity market has become increasingly important in recent years, due to the increasing integration of renewable resources in power systems. Trading in this market is challenging due to the multistage nature of the problem, its high uncertainty, and the fact that decisions need to be made rapidly in order to lock in profitable trades. We cast the problem of trading in continuous intraday markets as a reinforcement learning problem, and tackle the problem using policy function approximation. We specifically parametrize the trading policy using price thresholds, and optimize the choice of these thresholds using the REINFORCE algorithm. We demonstrate the effectiveness of our proposed policy by showing that it outperforms the method, classically used in the industry, rolling intrinsic of 4:2% (out of sample) on the 165 last days of 2015 in the German continuous intraday market.

The operation of power systems faces great challenges as the penetration of variable renewable energy keeps increasing rapidly. Meanwhile, the electrification of transportation and heating systems can increase future peak demand. As a result, grid infrastructures will need to be reinforced, including construction of combustion turbines in order to meet peak demand and offer flexibility to the system, and investment in transmission and distribution lines. However, there exists massive flexibility potential on the demand side, which can be exploited in order to reduce or postpone the upgrade of current power systems. Meanwhile, investment in smart meters is necessary to enable demand response. In this paper, we build a capacity expansion model for the European system and carry out a cost-benefit analysis considering the scenario of year 2050. We identify substantial flexibility potential in the residential sector and we find that the investment cost in smart meters can be justified from the perspective of energy suppliers.

Low-voltage distribution networks are emerging as an increasingly important component of power system operations. From a computational standpoint, the proactive utilization of these resources places daunting challenges due to their vast number and the non-linear physics that govern power flow in low-voltage networks. For this reason, a hierarchical approach to the organization of distribution markets which can better cope with computational scalability may be desirable. This paper models various alternatives to the coordination of transmission and distribution system operations, and investigates their relative performance on a small-scale network in terms of allocative efficiency, consistency with physical constraints, and pricing.

We present a stochastic unit commitment model for assessing the reverse requirements resulting from the large-scale integration of renewable energy sources and deferrable demand in power systems. We present three alternative demand response paradigms for assessing the benefits of demand flexibility in absorbing the uncertainty and variability associated with renewable supply: centralized co-optimization of generation and demand by the system operator, demand bids and coupling renewable resources with deferrable loads. We present simulation results for a model of the Western Interconnection.

This task force paper presents multi-domain tools and their interfacing issues to perform simulations that require expertise in multiple areas of physics, such as electrothermal, electromechanical, electrochemical, and electromagnetic simulations, etc. Features of single multi-domain simulation tools, which are capable of simulating more than one system such as electrical chemical, mechanical, etc., are discussed. In addition, programs that are not capable of performing multi-domain simulation independently but can be interfaced with other programs to perform multi-domain simulation are also covered. Two case studies are reported where multi-domain simulations are performed in offline and real-time environments through the interfacing of more than one simulation tool.

International audience; The transformations of conventional Internal Combustion Engine (ICE)-powered vehicles into electrified vehicles are of growing interest. The retrofit hybridization of a Peugeot 308 SW passenger car is studied. For this purpose, a complete hybridization system is added to the rear axle of the vehicle. As a result, the control of the initial ICE-powered vehicle has to be updated. The aim of this paper is to deduce the potential fuel saving thanks to this hybridization. The simulation results show an energy saving of at least 9% and potential consump-tion reduction on a Worldwide harmonized Light-duty Vehi-cles Test Cycle (WLTC).

Grid-connected converters (GCCs) are showing potential in providing virtual inertia and has attracted wide attention recently. For the virtual inertia emulated by GCCs, it is proportional with the DC-link capacitance, and thus, the DC-link capacitance can directly affect the dynamic performance for the GCC emulating inertia through modifying the inertia constant. However, the impacts of the DC-link capacitance have never been discussed in any literature before. Considering this issue, the influence of the DC-link capacitance on the dynamic performance for the GCC providing virtual inertia is analyzed in this paper. In addition to that, the selection approach of the DC-link capacitance is presented by tuning the system damping ratio within its optimal range. Simulations verify the correctness. Accepted version

SNCF is the largest electricity consumer in France. The rolling stock has the highest share of France railway system electricity consumption at 84%. The traction power of High-Speed Trains (HST) can reach up to 50% of the total rolling stock electricity consumption during the rush hours of weekdays. Given the new policies regarding energy regulation in peak power demand periods, load modulation by the traction power of HST is one of the flexibilities that are investigated by SNCF in the Smart Grid project. This paper presents the study related to the feasibility of a Targeted Traction Power Modulation (TTPM) of HST. Experimental tests have been achieved as proof of concept on the SNCF South-Europe-Atlantic HS line. Experimentation results showed between 20 and 30% of energy savings during the peak power demand period. Refereed/Peer-reviewed

This paper proposes a simple method of determining the size of excitation capacitance required to buildup a predetermined no load voltage for off-grid operation of a three-phase self-excited induction generator. The method is based on the combination of the principle of L-C resonance and non-linear magnetization characteristic of the generator. For given values of voltage and speed, the magnetizing reactance of the generator is first determined from the solution of a polynomial. The size of the excitation capacitance is then obtained from the principle of L-C resonance. The effectiveness of the proposed method is checked through a more sophisticated generator model where the electrical characteristics are represented by four first-order differential equations. The voltage build-up process in time domain is also investigated through the above model. The results obtained by the proposed method are also verified through an experimental setup consisting of a 1.5-kW induction machine and are found to be in very good agreement. Refereed/Peer-reviewed

International audience; With the increasingly rapid penetration of power electronics into AC grid, there is an urgent need in enhancing stability and performance of grid-connected converters. An advanced $H_{\infty}$ multi-input multi-output (MIMO) control using linear matrix inequality (LMI) techniques is introduced and applied for electromagnetic transient (EMT) three-phase voltage source converter (VSC) grid-connected system (simple STATCOM). The proposed control is compared with classical cascade vector control through a validation test of short-circuit event.The comparison results prove that the advanced control approach provides much better performance and transient behaviour compared to classic vector controls, as a result of using $H_{\infty}$ optimization technique to minimize the effect of frequency's and grid voltage's variations. The validation test is done using MATLAB Simscape in EMT simulation, and also gives a step-by-step path of implementation for applying advanced control in industrial-level application of power converter.These results will be further used to control the units of the new concept of \textit{Dynamic Virtual Power Plant} introduced in the H2020 POSYTYF project.

This paper presents the design process, detailed analysis, and a comparative study on the advantages of axial-flux induction machines (AFIM). A radial-flux induction machine (RFIM) is also studied as benchmark motor. Both motors are designed with six poles, 36 stator slots, and 32 rotor bars. Finite-element analysis is used to predict the performance of the two motors under both transient and steady state conditions. A complete comparative study along with the simulation results are reported. It is found that the studied AFIM has higher efficiency and power density. Refereed/Peer-reviewed

Replacement of traditional synchronous generators by renewable energy sources in modern power systems reduces the rotating inertia and causes stability and damping problems. Large scale battery energy storage systems (BESS) are now becoming an integral part of modern power systems to address the intermittency issues of renewables. This paper proposes a method of utilizing the BESS for damping rotor angle oscillations by modulating the machine output power through controlling the BESS current. First, an analytical expression for extra damping provided by the BESS is derived for a simple system. The effectiveness of the proposed method is then tested on a single-machine infinite bus (SMIB) system for both classical and higher order machine models using Matlab Simulink and DIgSILENT PowerFactory software. The control scheme is also tested in a multi-machine system. The additional damping provided by the BESS is compared with that obtained by using a static synchronous compensator (STATCOM). It is demonstrated that BESS is more effective than STATCOM in damping rotor angle oscillations. Refereed/Peer-reviewed

In this work, a novel distribution grid level peer-to peer (P2P) power market framework is proposed. The proposed market framework enables the satisfaction of distribution grid constraints such as the voltage limits and transfer capacity limits, for all P2P power transactions between each producer-consumer pair. An Alternating direction method of multipliers (ADMM)- based coordinating algorithm is proposed in which the P2P power trading is facilitated at the first step, and the second step involves ancillary services market clearing considering the grid constraints and the power injections of the P2P participants, in a distributed and interactive manner. The distribution locational marginal prices are iteratively computed in the process so that the costs associated with P2P energy transactions are allocated to the respective producer-consumer pairs. As such, the proposed algorithm enables optimal P2P power trading while supporting optimal voltage and congestion management of the distribution grid. The case studies illustrate the effectiveness of the proposed ADMM-based coordination algorithm. Ministry of Education (MOE) Accepted version This work was supported by the Singapore Ministry of Education Academic Research Fund Tier 1 Grant (MOE2018- T1-002-093) This work was supported by the Agency for Science, Technology and Research (A*STAR) through its Singapore-Germany Academic-Industry (2+2) International Collaboration under Grant A1990b0060.

Development of multi-energy integration for the commercial/Industrial sector introduces a high level of interdependency between different energy carriers which requires integrated modeling of these multi-energy systems (MESs). Transportation electrification such as the integration of plug-in electric vehicles (PEVs) has the potential to have a significant effect on the operation of commercial MESs which are equipped with large scale PEVs' parking lots. Although the development of energy management models to optimally integrate PEVs into MES leads to higher benefits, the associated operational uncertainties can bring additional costs for system operators. Ignoring these uncertainties can result in nonoptimal or even infeasible operational solutions for MES facilities. In this paper, a two-stage robust energy management model has been proposed for MESs equipped with PEV parking facilities to optimally schedule MES facilities and its interaction with upstream network, while, supplying the required PEV charging patterns, considering the uncertainties of MES loads and energy market prices. The proposed model is developed as a min-max-min optimization problem which is solved through a column-and-constraint methodology, recasting the tri-level problem into a first-stage min problem and a second-stage maxmin problem. The second-stage problem is solved through duality theory and Big-M linearization technique. A comprehensive case study has been conducted to evaluate the effectiveness of the proposed model. According to the obtained results, MESs are able to supply PEVs and multi-energy loads, while participating in upstream network energy market with operational solutions that are robust against uncertainties, resulting in extra benefits and lower operational costs. Refereed/Peer-reviewed

This paper optimally determines the configuration of the system, as well as each component capacity, in order to design a cost-effective and environment-friendly AC mini-grid hybrid power system for a remote area in South Australia. To this purpose, a modified Multi-Objective Particle Swarm Optimization (MOPSO) algorithm is applied. In each iteration of the modified proposed MOPSO, first, renewable power sources, including photovoltaic arrays and wind turbines along with energy storage components, are selected to form a hybrid power system along with the diesel generator. Then, the capacities of the components are optimized based on two objective functions, including Levelized Cost of Energy (LCOE) and CO2 emission, by analyzing the system using real data in hourly time-step for the load, solar insolation, ambient temperature, and wind speed over a 20-year timespan. Afterward, positions for system configuration and component capacities are updated according to the objective functions. Refereed/Peer-reviewed

Robustness, low cost and high torque density are the main characteristics of switched reluctance machines (SRM) that have made them promising for a variety of applications. In this paper a small diameter, long stack length (small aspect ratio) 6kW SRM is designed. In the first step, the dimensions of the studied SRM are derived based on conventional sizing equations. The significance of the rotor and stator pole arcs and also the effective torque zone as a constraint of the design are discussed. A finite-element study is carried out to validate the analytical model of the proposed SRM. The performance of the machine is comprehensively studied in different operating speeds. Simulation results demonstrate the capability of the machine for operation over a wide speed range.pa Refereed/Peer-reviewed

Load flow problems in electricity distribution systems (EDS) are generally solved through iterative methods such as Newton-Raphson, Gauss-Seidel, Forward-backward Sweep-based methods, etc. These methods are subject to problems such as poor convergence for EDSs with high R/X ratio branches and huge mathematical/programing burden in some cases. Moreover, in EDS studies such as power loss minimization, energy management, EDS planning/expansion, etc., load flow calculations need to be conducted alongside the required objective function and the related constraints of the study. To solve such models, optimization techniques such as metaheuristic methods, are used in which load flow calculation is conducted in each iteration, bringing extra mathematical burden. To address these limitations, this paper presents a new directly solvable and non-iterative load flow model which is developed using a new bus indexing procedure. A connectivity matrix is introduced to characterize the configuration of EDS and provide a feasible general representation of load flow equations. This enables the proposed modified load flow equations to be mergeable in any type of EDS study as model constraints. Moreover, unlike previous iterative models, it can be solved directly through standard optimization packages in a single shot with no need to further iterative optimization procedures, such as metaheuristic methods. This results in a moderate mathematical and programing burden. The 33-bus system is employed to evaluate the effectiveness of the proposed load flow model. The obtained results are reported and compared with forward-backward sweep-based load flow as a benchmark. Accordingly, it is shown that the proposed model maintains the same outcome without using iterative optimization techniques. Refereed/Peer-reviewed

With the predicted proliferation of Electric Vehicles (EV) worldwide over the next decades, effective EV charging strategies and the associated costs are of utmost interest. In addition to EV charging at home, various countries are introducing public charging facilities for fast charging of EVs. While there are implications of the various charging strategies on the load demand profile of the grid, the cost of charging EVs as well as the overall energy cost for households remain important drivers for the uptake of EVs in countries around the world. This paper investigates these critical aspects in the South Australian context, by examining various EV charging strategies and tariffs. The paper also investigates the impact of the charging strategies on the South Australian load demand profile. Refereed/Peer-reviewed