An energy management system for hybrid microgrids in remote communities

The integration of renewable energy source (RES) and energy storage systems (ESS) in microgrids has provided potential benefit to end users and system operators. However, intermittent issues of RES and high cost of ESS need to be placed under scrutiny for economic operation of microgrids. This paper presents a two-layer predictive energy management system (EMS) for microgrids with hybrid ESS consisting of batteries and supercapacitors. Incorporating degradation costs of the hybrid ESS with respect to the depth of charge and lifetime, long-term costs of batteries and supercapacitors are modeled and transformed to short-term costs related to real-time operation. In order to maintain high system robustness at minimum operational cost, a hierarchical dispatch model is proposed to determine the scheduling of utilities in microgrids within a finite time horizon, in which the upper layer EMS minimizes the total operational cost and the lower layer EMS eliminates fluctuations induced by forecast errors. Simulation studies demonstrate that different types of energy storages can be utilized at two control layers for multiple decision-making objectives. Scenarios incorporating different pricing schemes, prediction horizon lengths, and forecast accuracies also verify the effectiveness of the proposed EMS structure.

With the new challenges brought by the high penetration of Renewable Energy Resources (RESs) into the modern grid, developing new solutions and concepts are necessary. Microgrid (MG) is one of the new concepts introduced to overcome upcoming issues in the modern electricity grids. MGs and Multi-Microgrids (MMGs) are defined as the building blocks of smart grids. MGs are the small units, where power generation and consumption happen at the same location and MG makes the decisions by itself. MGs can operate grid-connected or island mode depending on the functionality of the MG. Energy Management System (EMS) is the decision making centre of the MG. The data from the devices is received by the EMS and after processing, the commands are sent to the controllable components. Management of voltage, active and reactive power, neutral current, unit commitment and economic dispatch are of the tasks of EMS. In this PhD thesis, an optimal EMS for MGs and MMGs is developed. The main objective of this project by developing the EMS is to optimize the energy flow in the MGs and MMGs to obtain peak load shaving in a cost beneficial system. In order to achieve an efficient EMS, communication system, forecasting system, scheduling system, and optimization system are modelled and developed. Different types of EMS operation, centralized, decentralized and distributed, are investigated in this work to achieve the best combination for MMG EMS operation. The communication system is mainly utilizing Modbus TCP/IP protocol for data transmission at local level and Internet of Things (IoT) protocols (MQTT) for the global communication level. A communication operation algorithm is proposed to manage the MMG EMS under different communication operation modes and communication failure conditions. Furthermore, a monitoring system is developed to collect the data from different devices in the MG. The data is processed in the MG EMS and the commands are sent to components through the communication infrastructure. The link between MGs and MMGs is through the proposed two-level communication system, where the expansion of MGs to a MMG is investigated. In the MMG, MGs are functioning as a unit while having different priorities and operating under different policies. Each MG has its own MG EMS and the EMSs transfer information through the communication system between each other in either centralized, decentralized, distributed, or no communication modes under the MMG EMS. The forecasting system is required in the EMS to predict the future MG characteristics such as power generation and consumption. The forecasted data is the input to the optimization and scheduling system of EMS. Employing the forecasting system in the EMS would increase the accuracy of the optimization and scheduling systems. In this thesis, the timeseries-based forecasting algorithms are employed to predict next day’s active power using the load data, generation data, weather data and temperature data as the inputs. The heart of EMS is the scheduling and optimization system. The purpose of the scheduling system is to define the amount and the time of energy flow in the MG for different generation sources and consumption loads. Furthermore, scheduling system is responsible for peak load shaving and valley filling. On the other hand, the optimization system has the task of minimizing the operation costs of the MGs. The role of market in the scheduling and optimization is important. Time of Use (ToU) tariff is the pricing system, which determines the peak and off peak hours for energy usage pricing. In order to apply the optimization system, a model of the system, an objective function and systems constraints are defined, where aging of battery energy storage system (BESS), operational cost of components and MG cost benefits are considered. To operate the EMS scheduling and optimization system, IBM CPLEX Optimization Studio solver conducts the optimization while for the scheduling system, objective function and constraints are defined in MATLAB. In this thesis, a rule-based, MILP and MIQP optimization system for commercial MGs including electric vehicles (EVs) are proposed to investigate performance of MG EMS for different case studies. In this thesis, the literature for different scheduling and forecasting systems is investigated and different optimization algorithms are analysed. The communication protocols utilized in this research are described and compared to other protocols in the literature. In different chapters of this thesis, the modelling of MGs and MMG EMS, different modules of EMS, forecasting, optimization, scheduling and communication systems are described and analysed. A novel communication system for MMG EMS operation is proposed for commercial buildings. The performance of MG EMS and MMG EMS is examined for power and neutral current sharing, operation cost optimization, and demand peak shaving applications and results are compared to investigate the performance of proposed algorithms.

Power sharing enhancement through a decentralized droop-based control strategy in an islanded microgrid

During system faulty condition, the output of inverter-type distributed generation is different from the traditional generation but according to the different distributed generation (DG) control methods. In this paper, a new fault analyzing method which considers the negative effect of DG control methods is proposed. Firstly, an equivalent fault analysis model of inverter type DG is developed by considering the constraints of different control methods, in terms of active and reactive power (PQ) control and V/F control. Based on the equivalent model, the fault analysis of island microgrid under the scenarios of PQ and V/F control is carried out by establishing equivalent island microgrid model. The fault voltage and current are calculated by solving the constrain equations based on different DG controls. Besides, both symmetrical and asymmetrical fault conditions are analyzed. Finally, the well-developed power grid simulation software DIgSILENT is applied to build a typical micro-grid model containing inverter-type DGs which is functioned with PQ control and V/F control. The results of the proposed analysis method are compared with the well-developed simulation results and typical conventional analyzing method. The comparison results verify that the proposed fault analysis method is effective and has great accuracy on the computation of fault data.

Economic–environmental hierarchical frequency management of a droop-controlled islanded microgrid

Experimental validation of a real time energy management system for microgrids in islanded mode using a local day-ahead electricity market and MINLP

The increasing interest in renewable energy-based power systems globally is driven by their abundance and environmentally friendly attributes. Islanded hybrid microgrid systems (IHMS) are a relatively new development in this field and involve the integration of two or more sustainable sources, such as wind turbines, solar photovoltaic (PV) systems, and other forms of renewable energy such as the ocean, wave, and geothermal energy. In order to ensure an uninterrupted power supply for the growing community and industrial sector of Perhentian Island, Malaysia, alternative power sources must be properly synchronized and managed through an energy management system. To this end, the main contribution of this study is the comprehensive analysis of various optimization methods in terms of net present cost (NPC) and convergence rate. The results of the analysis indicate that HOMER proved to be relatively faster in terms of convergence rate, with the NPC recorded as 387,185$ and the Levelized Cost of Energy (LCOE) recorded as 0.64$/kWh, which are the least among the other techniques evaluated. The hybrid energy system was designed to acquire the optimal quantity and size of power-generating modules, including PV systems, wind turbines, batteries, and diesel generators, while also meeting the load requirements. The optimization problem incorporated the LCOE and NPC into the cost function. Various optimization techniques were developed and tested. In addition, an advanced control method, which includes the use of Proportional–integral–derivative (PID) control and Fuzzy Logic Controller (FLC) with automated tuning, was applied to manage voltage and frequency. The control strategy was implemented in MATLAB Simulink, along with a full model of the islanded hybrid microgrid system. The simulation results demonstrate the effectiveness of the proposed FLC in maintaining the voltage and frequency within the acceptable range during various operating conditions. In conclusion, this manuscript provides a comprehensive study on the optimization and control of a solar-wind islanded hybrid microgrid. The proposed approach can be used as a valuable tool for the design and operation of solar-wind islanded hybrid microgrids in remote and islanded communities.

Harmonics are regarded as one of the main challenges in a microgrid. This issue may even get worse when different distributed generators (DGs) work together to solve the load sharing problems due to mismatched feeder impedances and diversified DG ratings. Even though load sharing can be achieved, the microgrid suffers from voltage unbalance and total harmonic distortion (THD) issues at the output of DG terminals as well as at the point of common coupling (PCC). Thus, in this paper, the power quality improving method is discussed, with a target of load sharing under the hierarchical control of different DG units and an active power filter (APF) in microgrids. To achieve this objective, we propose integrating a direct harmonic voltage controlled APF with DGs to improve their harmonic compensation performance. This proposed control scheme has many advantages over conventional control using a shunt resistive active power filter (R-APF) with voltage controlled DGs. First, based on the existing THD level of the PCC voltage, the proposed scheme provides improved voltage compensation and reduction in THD in the islanded microgrid. Secondly, equal load sharing can be achieved simultaneously. Thus, the proposed scheme provides better performance and a seamless interface as the proposed study mainly contains both the voltage controlled DGs and the local based voltage detection APF.

This article proposes a new energy management system (EMS) for hybrid AC/DC microgrids (MGs) in remote communities. The major goal of the proposed EMS is to minimize the daily operational costs while supplying sufficient clean water to the remote community, meeting customers’ preferences, and maintaining the technical operational limits of the MG. To achieve this goal, the proposed EMS performs real-time supervisory control on all local controllers of all system assets in both generation and demand sides. On the generation side, distributed renewable and nonrenewable generation units are integrated into the system. Renewable sources of energy are controlled through a maximum power point tracking algorithm for increased efficiency, while nonrenewable DG units are controlled through droop parameters to satisfy the demand. On the demand side, optimal operational schedules for residential appliances and water desalination units are developed to satisfy customers’ requirements and achieve minimum operating costs. Moreover, the EMS controls the AC/DC interlinking converters (ICs) allowing bidirectional flow of power to and from both AC and DC sections of the MG. For practical real-time operation, the proposed EMS utilizes an adaptive rolling-time horizon when responding to changes/disturbances in the system or in user inputs. The assets scheduling problem is modeled as a mixed integer nonlinear programing. Simulation studies were conducted on a hybrid AC/DC MG to demonstrate the effectiveness of the proposed microgrid EMS.

This paper proposes energy management systems for micro-grids. In recent years, the use of renewable energy sources in micro-grids has become an effective means of power decentralization especially in remote areas where the extension of the main power grid is an impediment. But a mixture of renewable energy sources and conventional generation poses serious challenges in the operation and control of micro-grids as a result of the uncertainty associated with renewable sources. Therefore, excellent energy management with regards to the power production, control, reliability, and consumption is needed in the power system. The main objective of the energy management system is to minimize the system cost as well as meeting the demand. In order to take into account the uncertainties of distributed generators (DGs) and load consumption, demand response participation and load shedding is taken into account. Two different types of algorithms are used to solve the smart energy management system of the micro-grid. To minimize system costs, which includes unit commitment and demand response, a genetic algorithm is used. To further balance the supply and demand during extreme cases, load shedding is employed through the use of the artificial neural network. The simulation results displayed corroborate the merit of the proposed method.

The integration and control of Microgrid (MG) systems remain critical challenges in the widespread adoption of renewable energy sources, especially photovoltaic (PV). An adaptive control approach is proposed in this work to improve the MG stability in the presence of PV and battery energy storage systems (BESSs). The proposed approach incorporates adaptive centralized secondary control, primary control, and local PV/BESS control. The primary control based on the droop control approach is applied to regulate voltage and frequency in a decentralized manner while ensuring balanced power-sharing among different distributed generators (DGs) in the MG. Besides that, an adaptive coordinated secondary control is implemented to alleviate the deviations of frequency and voltage caused by PV intermittent generation and load variation, which has a central controller that restores nominal setpoints for all DGs. The BESS type used in this study is a lithium-ion battery which is applied to preserve the DC bus voltage approximately constant during various events, enhance system resilience against PV power intermittency, and balance load power demand. The biggest advantage of the proposed control approach is that it dynamically regulates battery charging and discharging to compensate for variations in PV generation and load demand, ensuring stable system operation. In contrast to conventional studies that assume an ideal DC source to represent DGs, this study models PV generation with real-time fluctuations and maximum power point tracking, providing a practical and realistic simulation environment. The robustness and effectiveness of the proposed technique are validated using MATLAB Software. The results obtained signify highly efficient voltage and frequency stability, improved system resilience under dynamic conditions, and optimal power-sharing among DGs. Finally, a comparative analysis with conventional models highlights the superior adaptability and reliability of the proposed approach, making it a viable solution for real-MG applications.

As a cutting-edge technology, microgrids feature intelligent EMSs and\nsophisticated control, which will dramatically change our energy\ninfrastructure. The modern microgrids are a relatively recent development with\nhigh potential to bring distributed generation, DES devices, controllable\nloads, communication infrastructure, and many new technologies into the\nmainstream. As a more controllable and intelligent entity, a microgrid has more\ngrowth potential than ever before. However, there are still many open\nquestions, such as the future business models and economics. What is the\ncost-benefit to the end-user? How should we systematically evaluate the\npotential benefits and costs of control and energy management in a microgrid?\n

10 - DC microgrids for electric vehicle wireless charging

Role of optimization techniques in microgrid energy management systems—A review

The increasing use of distributed and renewable energy resources presents a challenge for traditional control methods due to the higher complexity and uncertainty brought by these new technologies. To address these challenges, rein-forcement learning algorithms are used to design and implement an energy management system (EMS) for different microgrids configurations. Reinforcement Learning (RL) approach seeks to train an agent from their interaction with the environment rather than from direct data such as in supervised learning. With this in mind, the problem of energy management is be posed as a Markov decision process and it is solved using different state-of-the-art Deep Reinforcement Learning (DRL) algorithms, such as Deep Q-Networks (DQN), Proximal Policy Optimization (PPO) and Twin Delayed Deep Deterministic Policy Gradient (TD3). Additionally, these results are compared with traditional EMS implementations such as Rule-Based and Model Predictive Control (MPC) used as benchmarks. Simulations are run with the novel Pymgrid module build for this purpose. Preliminary results show that RL agents have comparable results to the classical implementations with some possible benefits for generic and specific use cases.