Search

The need for solar photovoltaic (PV) power forecasting arises due to rapid fluctuations in solar PV output. This variation can cause an imbalance between the demand and generation in a microgrid. This paper attempts to analyze the impact of inaccuracy in predicting PV power generation on the frequency of a standalone microgrid system. For this purpose, two different forecasting models have been studied for two different types of weather conditions, characterized by varying degrees of cloud cover. Moreover, when addressing frequency fluctuations, restoring them to the nominal value becomes imperative, and this is where secondary controllers play a crucial role. This paper further evaluates and compares the performance of three proportional-integrative derivative (PID) controllers for smoothing frequency control. For the considered combinations of two forecasting models and three secondary controllers, the standard deviation (SD) of frequency is the lowest for the long short-term memory (LSTM) forecaster and Particle Swarm Optimization (PSO) – PID (PSO-PID) controller. The corresponding reduction of SD, after replacing persistence: PSO-PID with LSTM: PSO-PID, in combination with |clear: cloudy| day is |28.43%: 32.12%| for overshoot and |11.87%: 18.36%| for undershoot frequency deviation. [ABSTRACT FROM AUTHOR]

Renewable energy sources such as wind and solar power exhibit strong stochasticity and volatility, resulting in decreased power supply security and sustainability. A strategically optimized hybrid renewable energy system (HRES) is crucial for maintaining stable load operations and achieving sustainable energy development. This paper introduces an energy optimization management model for an independent HRES consisting of wind turbines, photovoltaic systems, diesel generators, and energy storage units. Operational strategies focus on energy storage-led loads following diesel generator-led load prioritizations. The model aims to optimize objectives to include economic, environmental, and power supply reliability indices. A dynamic adaptive parameter approach balances the parameters of the objective function at various instances. The optimal capacity allocation of the model is solved using the improved gray wolf optimization (IGWO) algorithm. This approach incorporates the golden sine strategy, the levy flight strategy, and the dynamic inverse learning strategy into the traditional GWO algorithm. Analyzing different test functions, evaluation metrics, and actual load data indicates that the proposed algorithm excels in global optimization capabilities and search speeds. The model significantly reduces the economic and environmental costs of the HRES microgrids and improves the sustainable development of renewable energy in various scenarios. [ABSTRACT FROM AUTHOR]

Virtual synchronous generators (VSGs) have considerable potential to be applied in renewable energy microgrid systems because they provide inertia and damping, thus improving system stability. However, the fault characteristics of renewable energy microgrids differ from those of conventional grids. Consequently, investigating the fault responses of VSGs and establishing the associated fault model would significantly enhance the fault calculation and protection mechanism design in renewable energy microgrids. This paper briefly introduces the VSG control principles and fault control strategy before establishing the corresponding transient- and steady-state VSG fault models and calculating the model parameters. For simplicity, the fault model is linearized at the steady-state estimation point and semi-decoupled for the computation of the voltage amplitude and power angle. Finally, the results from MATLAB/Simulink simulations are compared with those from the proposed fault models, and the influence of the system parameters on the model is discussed. The results demonstrate the validity, simplicity, intuitiveness, and broad applicability of our approach. [ABSTRACT FROM AUTHOR]

The demand response (DR) resources of multiple entities have different operation strategies, which makes it difficult to realize the co-optimization. This paper proposes a cooperative hierarchical scheduling optimization model of interconnected multi-microgrid and an active distribution network (ADN) considering DR with different operation strategies. First, the collaborative hierarchical scheduling framework is proposed and the interaction characteristics and cooperative scheduling mode of multi-interconnected microgrid and ADN are analyzed. Second, the two-layer cooperative hierarchical scheduling model is established, considering DR with different strategies. The upper layer model takes the minimum operating cost of the ADN as the objective to optimize the trading tariff between the ADN and microgrids, and its own DR, while the lower layer model takes the operating cost of multi-microgrid as the objective to optimize the purchasing and selling of electrical energy for the ADN, the DR, and the interaction power. Finally, the case studies with three microgrids and an ADN are used to demonstrate the validity and effectiveness of the proposed model. [ABSTRACT FROM AUTHOR]

The integration of renewable energy sources into the power grid poses several challenges due to their variability in output power as they primarily depend on weather conditions. As a countermeasure, battery energy storage systems are introduced in order to mitigate output power fluctuations of the renewable energy sources. Moreover, demand response programs have been attracting huge attention as an effective mechanism for efficient energy management. It enables power suppliers to cope with the output uncertainty of renewable energy systems. First step, we propose an HRES design methodology that takes into account one year. In order to find the optimal power supply configuration, it is necessary to take into account the changes in load demand due to weather conditions and seasons for one year. However, the computational cost of simulating one year is huge and requires a rather long simulation time. Therefore, we propose a simplified simulation method that uses clustering to account for changes in load demand due to weather and seasonality in one year. Second, the first proposed method is used to propose a method for optimizing the HRES equipment capacity and operation schedule considering demand response. In this paper, mixed-integer linear programing is used as the optimization method. The proposed method considers two types of demand response and achieves cost reduction by controlling load demand in response to generated power fluctuations. [ABSTRACT FROM AUTHOR]

Given the pillar role of renewable energy in the low-carbon energy transition and the balancing role of energy storage, many supporting policies have been promulgated worldwide to promote their development. To achieve the ambitious goal of no less than 1200 GW of wind and solar by 2030, China has also introduced policies to encourage the deployment of energy storage for the grid integration of renewable energy. The national policy is conducive to enhancing system flexibility for renewable integration, but it will also add the costs of renewable energy, especially when counting the regional differences. This paper adopts an improved levelized cost of electricity model to examine the total costs of renewable power co-deployed with energy storage in different provinces of China. The results show that the nationally unified energy storage co-deployment requirement, namely, 15% capacity ratio of renewable installation and 4 h duration, will negatively affect the economics of renewable generation, leading to an average cost increase in 15% and 21% for wind and photovoltaic generation, respectively. The economics of co-deploying energy storage under current market mechanism is inferior, but it can be effectively improved when energy storage participates in ancillary services market. With the revenue of frequency regulation, the cost of renewable co-deployed with energy storage can be even less than that without co-deployment in most provinces, except for Hebei, Jiangxi, and Gansu. An independent market entity status is thus conducive to encourage the co-deployment. Our finding also implies for a province-specific energy storage co-deployment policy. [ABSTRACT FROM AUTHOR]

In this paper, offline adaptive control of a microgrid in an islanded operation mode is presented. The proposed control scheme consists of a power controller, voltage controller, and current controller, which are operating in a cascade structure. The power controller is designed based on the droop characteristics while the voltage and current loops are optimal Proportional-Integral (PI) controllers. Due to the fact that the performance of all PI controllers fundamentally depends on the proper choice of controller parameters, these parameters are optimized based on the offline utilization of the Model Reference Adaptive Control (MRAC) method and the Genetic Algorithm. The adaptation mechanism in the MRAC method is usually determined using the mathematical model of the system, while in this paper, an optimal adaptation mechanism is achieved by the Genetic Algorithm. The optimal parameters of the controller could be obtained based on the measurements of the Distributed Generation (DG) output voltage when the DG is operating with a test load without any knowledge of its model. To investigate the performance of the suggested controller, simulations are done by MATLAB/Simulink in three different scenarios including load perturbation, disconnection of a DG, and nonlinear load, where the results are compared with IEEE recommendation. In addition, the suggested controller is compared with the H2/H∞ robust controller. [ABSTRACT FROM AUTHOR]

A disconnect between real world financing and technical modeling remains one of the largest barriers to widespread adoption of microgrid technologies. Simultaneously, the optimal design of a microgrid is influenced by financial as well as technical considerations. This paper articulates the interplay between financial and technical assumptions for the optimal design of microgrids and introduces a design approach in which two financing structures drive an efficient design process. This approach is demonstrated on a descriptive test case, using well accepted financial indicators to convey project success. The major outcome of this paper is to provide a framework which can be adopted by the industry to relieve one of the largest hurdles to widespread adoption, while introducing multiple debt financing models to the literature on microgrid design and optimization. An equally important outcome from the test case, we provide several points of intuition on the impact of varying financing terms on the optimal solution. [ABSTRACT FROM AUTHOR]

To address the instability issues of LCL-type wind power inverter's operation under weak grid conditions, stability control methods are investigated in this paper. Based on an inverter impedance model that considers an inner current control loop, a precise impedance model that considers dynamic characteristics of dc bus voltage and a reactive power outer loop is created in this paper. The phasor diagram method and an analytical method are used to interpret frequency characteristics of the inverter output impedance. The impedance analysis results show that the instability of the inverter in a weak grid has a weak correlation with the LCL filter parameter and a strong correlation with the inner and outer control loops. Phase-locked loop bandwidth reduction improves inverter's operation stability to some extent. However, it also reduces the inverter's tracking and anti-interference performance. Therefore, a capacitor current-feedback-based stability control method is proposed in this paper. The feasibility of the proposed control method is proven via general Nyquist-criteria-based impedance analysis. Finally, an RT-LAB hardware loop test platform is created to verify the validity of impedance modeling proposed in this paper and the effectiveness of the stability control strategy. [ABSTRACT FROM AUTHOR]

The intermittency of single renewable energy sources affects significantly their reliability, and hence, hybrid renewable energy systems (HRESs) are introduced. HRESs are notably used to meet the needs of residential loads. Nevertheless, power surplus and shortages still represent issues that must be handled by energy management systems. In this paper, a novel method for optimal energy management of a grid connected micro-grid composed of five smart Moroccan prosumers is presented. The proposed energy management strategy is organized using three levels, namely, HRES level, load-scheduling level, and communication-sharing level between HRESs. The first two levels of energy management are processed at the local level, i.e., inside each smart home. For the third level, our paper demonstrates the feasibility of interactions and communication between the five different houses interconnected using an optimized infrastructure topology as a smart neighborhood grid. Each of these houses is empowered by their own hybrid renewable energy system made of possible system hybridization based on PV (Photovoltaic), wind turbine, and battery storage system. This helps each house to manage its surplus and to deal with energy shortage through the proposed multi-level system's energy management algorithm. The application of information and communication technologies enables the set of smart homes connected in a micro-grid framework to contribute to the emergence of the concept of the internet of sustainable green things. Simulations conducted by the MATLAB platform validate the proposed methodology, which relies on micro-grid surplus injection and neighborhood storage during extreme surplus, else load dynamics thereafter micro-grid-injected energy recovering during hybrid energy deficit. The proposed system manages the energy produced by HRESs and reduces the energy demanded by the utility grid using the proposed topology that connects houses and ensures interactions between the micro-grid and the utility. The results show that the frequency of micro-grid deficiency was significantly reduced due to the established micro-grid with the five houses according to the proposed topology and the developed multi-level management system. [ABSTRACT FROM AUTHOR]

The grid-connected control technology of ac/dc hybrid microgrids affects power quality, stability, robustness, and other indicators. In order to achieve high-quality and reliability of ac/dc hybrid microgrids connected to a distributed network, this paper proposes a robust control method for grid-connected ac/dc hybrid microgrids based on mixed sensitivity. First, the grid-connected model of ac/dc hybrid microgrids is equivalent to the superposition theorem; then, the problem of how to control grid-connection can be transformed to the problem of how to deal with mixed sensitivity by considering the influence of the distributed network on ac/dc hybrid microgrids as a disturbance. Second, the grid-connected robust controller can be obtained by using the Riccati method repeatedly, which can ensure good nominal performance of the system and enhance the robustness of the system. Finally, the validity and feasibility of the proposed control method can be verified with three simulation conditions. [ABSTRACT FROM AUTHOR]

With the wide application of high proportion of distributed clean energy in regional microgrids, the issue of maximizing the utilization of renewable energy among multi-microgrids has become a research hotspot. Based on the proposed multi-microgrids' energy collaborative optimization and complementation model, a multi-microgrids' energy real-time optimization management and dispatch strategy is proposed that fully considers the real-time complementarity of renewable energy between multi-microgrids and achieves the best coordinated dispatch of energy. Two typical scenarios were set up in the IEEE 33-bus network model for verification, and the synergistic effects of different schemes were compared and set up. The data obtained demonstrate that the dispatch and management strategy proposed in this paper can achieve the maximum integration of renewable energy and the lowest operating cost among multi-microgrids, and it also validates the real-time, feasibility, and effectiveness of the proposed strategy. [ABSTRACT FROM AUTHOR]

Micro energy grids (MEGs) play a vital role in realizing carbon neutrality and efficient utilization of renewable energy resources. This research focuses on optimizing the synergy of MEG interconnections. Given the diverse development paths of different operating entities within the system, information barriers emerge among MEGs, creating great difficulties for the collaborative system management. In response, this paper proposes a decentralized coordinated dispatch model targeting multiple stakeholders within the system. This model accounts for energy interactions between MEGs and the inherent uncertainty associated with renewable energy sources. Specifically, stochastic optimization approach was applied to characterize the uncertainty of renewable energy output by generating stochastic scenarios. Furthermore, it incorporates the analytical target cascading (ATC) method to decouple objective functions and constraints, creating autonomous scheduling sub-models for individual MEGs. This decentralized approach ensures independent modeling and coordinated problem-solving. Simulations verify that (1) the ATC-based inter-MEG energy interaction strategy effectively achieves decentralized coordinated scheduling of multiple MEGs and (2) the decentralized coordinated scheduling solution closely approximates the global optimum while considering the interest of various system entities. [ABSTRACT FROM AUTHOR]

The fluctuation of renewable energy resources and the uncertainty of demand-side loads affect the accuracy of the configuration of energy storage (ES) in microgrids. High peak-to-valley differences on the load side also affect the stable operation of the microgrid. To improve the accuracy of capacity configuration of ES and the stability of microgrids, this study proposes a capacity configuration optimization model of ES for the microgrid, considering source–load prediction uncertainty and demand response (DR). First, a microgrid, including electric vehicles, is constructed. Second, the uncertainty of renewable energy resources and electric demand is handled by Monte Carlo scenario generation techniques and K-means-based scenario reduction techniques. Then, a DR model combining price-based demand response and incentive-based demand response is constructed to achieve a better match between electricity demand and supply. Finally, the results of the ES capacity configuration are determined with the objective of minimizing the total daily cost of the microgrid. The simulation results show that the optimal configuration of ES capacity and DR promotes renewable energy consumption and achieves peak shaving and valley filling, which reduces the total daily cost of the microgrid by 22%. Meanwhile, the DR model proposed in this paper has the best optimization results compared with a single type of the DR model. It is verified through comparative analysis that under a certain proportion of flexible loads, the total daily cost of the microgrid is the lowest when the time-shiftable loads and the curtailable loads are both 10%. [ABSTRACT FROM AUTHOR]

Reducing the duration and frequency of blackouts in remote communities poses an engineering challenge for grid operators. Outage effects can also be mitigated locally through microgrids. This paper develops a systematic procedure to account for these challenges by creating microgrids prioritizing high value assets within vulnerable communities. Nighttime satellite imagery is used to identify vulnerable communities. Using an asset classification and rating system, multi-asset clusters within these communities are prioritized. Infrastructure data, geographic information systems, satellite imagery, and spectral clustering are used to form and rank microgrid candidates. A microgrid sizing algorithm is included to guide through the microgrid design process. An application of the methodology is presented using real event, location, and asset data. [ABSTRACT FROM AUTHOR]

This paper explores cost-optimal pathways to 100% renewable power systems for the U.S. building stock. We show that long-duration misalignments of supply and demand, spanning from multi-day to seasonal timescales, present a dominant challenge that must be addressed to meet real-time 100% renewable targets. While long-duration misalignments can be addressed through energy storage, we show that alternative and readily available solutions that are more cost-effective should be considered first. Through a techno-economic analysis, we identify cost-optimal, region-dependent, supply-side, and demand-side strategies that reduce, and in some U.S. regions eliminate, the otherwise substantial capacities and associated costs of long-duration energy storage. Investigated supply-side strategies include optimal mixes of renewable portfolios and oversized generation capacities. Considered demand-side strategies include building load flexibility and building energy efficiency investments. Our results reveal that building energy efficiency measures can reduce long-duration storage requirements at minimum total investment costs. In addition, oversizing and diversifying renewable generation can play a critical role in reducing storage requirements, remaining cost effective even when accounting for curtailed generation. We identify regionally dependent storage cost targets and show that for emerging long-duration energy storage innovations to achieve broad adoption, their costs will need to compete with the decreasing cost of renewables. The findings of this research are particularly important given that most long-duration storage technologies are currently either uneconomical, geologically constrained, or still underdeveloped. [ABSTRACT FROM AUTHOR]

This paper proposes a stochastic parallel optimization approach for an islanded microgrid. The proposed approach solves two of the problems faced by such microgrids, namely, how to mitigate the adverse effects of multiple source–load uncertainties and how to accelerate the computation speed, by decomposing an entire microgrid into several entities, i.e., the utility and the end users. A chance constrained programming-based optimization model is established to address the former problem. By discretizing the continuous variables, the chance constraints of multiple uncertainties can be transformed into deterministic constraints. The second problem is solved by incorporating a new linear penalty function into the analytical target cascading method. With this proposed penalty function, the algorithm converges rapidly and the scheduling results for both the utility and the end users are obtained with low computational cost. A simulation is conducted to illustrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Interconnecting multiple combined heat and power (CHP) microgrids with the distribution network to form a CHP multi-microgrid system can promote the complementarity of various energy forms effectively and improve energy utilization efficiency. Therefore, a dual-layer stochastic optimal scheduling strategy of a multi-microgrid system interconnected with multiple CHP microgrids and multi-node distribution networks is proposed in this paper. In this strategy, the whole system is divided into two layers: (1) The lower layer adopts a stochastic simulation method based on versatile distribution to randomly generate a large number of scenarios and then uses a backward scenario reduction method to reduce the generated scenarios to a target number. Furthermore, with the goal of minimizing the sum of the weighted costs of the microgrid in each scenario, an economically optimal mixed integer programming model for a CHP microgrid with multiple energy storage devices is established. The output power of the electro-thermal units is optimized in each microgrid, and then the power to buy or sell power for each microgrid is determined. (2) The upper layer takes the minimum loss of the distribution system connected to multiple CHP microgrids as the goal and considers the power of each microgrid optimized by the lower layer as constraints. The optimal power flow model of the multi-node power distribution system is formulated based on the second-order cone relaxation technology, and the global optimal solution of the power node is obtained. Finally, the daily dispatch in which the IEEE-30 node power system connected to multiple CHP microgrids is used as an example for analysis. The results show that considering the uncertain factors of wind and solar output can effectively improve the reliability and robustness of the system when multiple CHP microgrids are connected to distribution network. Meanwhile, the coordinated use of multiple flexible coupling devices can reduce economic costs and promote energy consumption. [ABSTRACT FROM AUTHOR]

First, an extended fully distributed algorithm (FDA) is proposed, and then, on the basis of the extended FDA, a centralized-distributed algorithm (CDA) is modified to solve multi-step economic dispatch problems in microgrids. The FDA could be realized in fully distributed agents, and by contrast, CDA is presented in both locally centralized and globally distributed ways to reduce the amount of distributed agents for applying it to large-scale microgrids and eliminating the necessity for a central energy management system. Information exchange between agents, such as power mismatch and Karush-Kuhn-Tucker multiplier, is realized by consensus networks, and the optimal solutions can be found through iterative communication between each agent with its neighbours. Besides, different from the most existing single-step solver, this paper deals with multi-step decision-making problems incorporating storage optimization and generator ramp-rate constraints. Moreover, incremental costs of local generators are avoided to participate in global information exchange, and so, local privacy can be protected. Meanwhile, CDA is a robust method against communication failures. Several case studies implemented on a twelve-bus microgrid are tested, and comparison results among CDA, FDA, and the centralized quadratic programming algorithm (CA) are discussed to validate the proposed methods. Simulation results show that CDA is effective to find the global optimum reached by CA and presents a better dynamic response than FDA and can be used to solve multi-step economic dispatch problems in a globally distributed way in microgrids. [ABSTRACT FROM AUTHOR]

DC microgrids are desired to provide the electricity for the remote areas which are far from the main grid. The microgrid gets popularity because DC power sources such as photovoltaics (PVs), battery banks, and fuel cells can be interconnected without AC/DC converters. The stochastic nature of PV power and variation in the demand are responsible for voltage fluctuations in the DC microgrids. The voltage stability is a most important issue in the case of DC microgrids. The challenging thing is to maintain the microgrid voltage up to a reference value for the variable load. This paper focuses on designing a controller to obtain the voltage stability of the microgrid. Two controllers [PI and fuzzy logic PI (FL-PI)] are designed for the voltage stability analysis of the DC microgrid. Real-time simulation results show that upon appropriate choice of controller parameters, the DC microgrid voltage can be kept constant regardless of the wide range of voltage variations of the source and the load. The simulation results show the comparison of two controllers. The FL-PI controller shows much better performance than the conventional PI controller for the DC grid voltage control. With the inclusion of the proposed controllers, controlled voltage will be obtained, whatever may be the load and its variation. [ABSTRACT FROM AUTHOR]

Photovoltaic (PV) microgrids operating in the islanding mode are more susceptible to stability problems, due to the fluctuation of PV output power and weak system structures. In this paper, a coordinated optimization method considering the time-delay effect of islanded PV microgrids based on the modified backtracking search algorithm is proposed to improve the stability of PV on the condition of suffering small disturbances. First, the small signal model of PV microgrid considering time-delay characteristics of the inverter, which consists of a PV array, batteries, dc/ac inverters, and loads, is studied. Then, the range of controller parameters are obtained by analysing small signal stability using the technique of eigenvalue analysis. Finally, a coordinated optimization objective function is built and the modified Backtracking search algorithm is adopted to optimize the parameters. The simulation results show that the proposed method is effective for improving the stability of PV microgrids when suffering small disturbance. [ABSTRACT FROM AUTHOR]

Clean energy has become one of the more broadly acceptable options for the future transformation of existing energy systems. The energy transformation also requires comprehensive preparatory actions. Organizing a multi-chain clean energy value chain into a complex clean energy network system is likely to be critical to the successful utilization of clean energy in the future. This article proposes a novel concept for a clean energy value co-creation system covering the entire clean energy value chain with division on power generation, energy storage, and end utilization sections. This article mainly analyzes the self-organizing evolution mechanism and value co-creation of the clean energy value co-creation system (CEVCS), which is therefore critical to the further development of the renewable energy industry. First, this article analyzes the connotation and basic structure of CEVCS and introduces the self-organizing characteristics of CEVCS. Second, the self-organizing evolution mechanism of CEVCS is discussed, which is analyzed from four aspects: evolutionary power, evolutionary path, evolutionary form, and evolutionary process. Third, this paper analyzes the information cooperative evolution process of CEVCS, defines the value order parameter, and studies how the constituent elements of CEVCS realize value creation through a synergy mechanism based on the theory of synergy. Then the dissipative structure model and information resource collaborative model of CEVCS are presented. Finally, the conclusions are given in the end. The proposed CEVCS concept would support the effective operation of the clean energy value chain within municipalities by establishing virtual energy cooperatives aimed at meeting energy needs and supporting community development. [ABSTRACT FROM AUTHOR]

An active power dispatch method for a microgrid (MG) with multi-type loads, renewable energy sources, and distributed energy storage devices (DESDs) is the focus of this paper. The MG operates in a grid-connected model, and distributed power sources contribute to the service for load demands. The outputs of multiple DESDs are controlled to optimize the active power dispatch. Our goal with optimization is to reduce the economic cost under the time-of-use price and to adjust the excessively high or low loading rate of distributed transformers caused by the peak-valley demand and load uncertainties. To simulate a practical environment, stochastic characteristics of multi-type loads are formulated. Then, a finite-horizon Markov decision process model is established to describe the dispatch optimization problem. A learning-based technique is adopted to search the optimal joint control policy of multiple DESDs. Finally, simulation experiments are performed to validate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

According to the characteristics of an AC/DC hybrid microgrid, this paper presents a new method for network planning of an AC/DC hybrid microgrid based on clustering partition and multi-agent reinforcement learning algorithm. The planning method is divided into three steps, namely, sub-microgrid partition, sub-network planning, and main network planning. On the basis of the principle of partition, a clustering partition model considering the location, type, and capacity of distributed generators (DG) and load is established, to divide this system into different AC and DC sub-microgrids. Then, the optimization model of the sub-network considering the annual cost of the converter and the optimization model of the main network considering the change of network loss of the sub-microgrid are proposed, which form a bilayer optimization model of the AC/DC hybrid microgrid. The optimal clustering partition is obtained by combining the k-means clustering algorithm with the particle swarm optimization algorithm, and considering the interaction between main network planning and sub-network planning, the network planning model is solved by the multi-agent reinforcement learning algorithm based on the Stackelberg equilibrium game. A numerical example is given to verify the validity and accuracy of the above models and algorithms. [ABSTRACT FROM AUTHOR]

An effective regulating power management is needed in next generation interconnected microgrids to mitigate the effect of load-generation imbalance in the system. The objective of this paper is to present a two-stage regulating power management scheme in an interconnected microgrid (MG) system. First, in a day-ahead power management, the power generation set-points are determined in the individual MGs considering predicted demand and power from renewable energy sources. Second, in a real-time regulating power management, a network operator establishes a set of power flow interactions among MGs to mitigate the load-generation imbalances in MGs. These interactions are obtained using a convex multi-constraint optimization problem and a cooperative developed algorithm. A simple suboptimal greedy solution is also proposed. Numerical results demonstrate the effectiveness of the proposed real-time regulating power management in cooperation with day-ahead scheduling to reduce the system load-generation imbalance, in comparison with conventional power resources. [ABSTRACT FROM AUTHOR]

Model Predictive Control (MPC) uses the model of system at current time to predict the system behavior at the future sampling interval in the prediction horizon and sets a number of variables to their references due to their controllability from input variables. This paper concentrates on the design and analysis of a controller for Distributed Generation (DG) microgrids in islanding and grid connected operation modes using a Receding Horizon MPC scheme. In this contribution, active and reactive powers are used in a cost function as an inner control loop to produce switching states. Two outer voltage and frequency droop loops share active and reactive powers between DGs in the microgrid. The design concept of the proposed control system is evaluated through simulation studies and experiment under different test scenarios. The impact of the simulation and the experimental results shows that the operations of the DG units within the microgrid can be coordinated effectively under the proposed control system to ensure stable operation of the overall microgrid and power quality improvement. [ABSTRACT FROM AUTHOR]

An electrical power system is considered as a critical infrastructure (CI), the epicenter of a nation's economy, security, and health. It is interlinked with other CIs such as gas and water supplies and transportation and communication systems. A failure in the power system will immensely affect the functionality of these CIs. Therefore, enhancing power system resilience is crucially needed to ensure continuous operation of these CIs. One of the possible approaches to improve the resilience in a power system is by integrating microgrids in the power system. Microgrids have proven to have self-healing and resilient capabilities in such extreme events which inflict damage out of the conventional scope of failures. Operational flexibility and controllability make microgrids a viable solution for resilience enhancement. This paper reviews the concept of resilience in power systems and the functions of microgrids in enhancement of resilience. The most current studies in improving power system resilience through microgrids are reviewed by highlighting their advantages and limitations. [ABSTRACT FROM AUTHOR]

This paper proposes a hybrid coordination control strategy to improve the low voltage ride-through (LVRT) capability of microgrids. During microgrid external failure, the overcurrent and the voltage sag of the microgrid can be effectively suppressed. Compared with traditional control strategies, the microgrid transient performance can also be enhanced by adopting a generalized control algorithm, performing the active switching of the energy storage (ES) operation mode and triggering the inductance-type flux coupling fault current limiters (FCLs). Among them, according to the hierarchical control structure of the grid-connected inverter, the generalized control algorithm is applied to ES. Through the two degrees of freedom's control principle and inverse plant modeling techniques, the generalized control algorithm can make the control layer of the ES inverter operate in all control targets (i.e., PQ control, droop control and Vf control), with a single control structure. Also, it eliminates the effects of microgrid distortion voltage and ES surge current on the control system of the ES inverter. For the inverter application layer, by introducing microgrid voltage and angular frequency feedforward compensation, the voltage and frequency fluctuation of the microgrid can be suppressed efficiently under the fault conditions. Therefore, with the help of the generalized control algorithm, the active switching function of ES can be carried out successfully, which is the key technology for microgrids to realize their operational mode smooth switching. Meanwhile, the voltage sag and overcurrent of microgrids can be significantly mitigated through generating reactive power from microgrids and injecting it into inductance-type FCLs with lower cost. The related theoretical derivation and simulation analysis under various scenarios (including asymmetric faults) confirm that the proposed control strategy can not only enhance the LVRT capability, but also strengthen the microgrid transient performance. [ABSTRACT FROM AUTHOR]

The purpose of this paper is to present a method for the probabilistic economic dispatch problem between the grid and a hybrid-microgrid. The probabilistic economic dispatch problem is solved for reducing the fossil fuel usage and the total cost of electricity consumption. A scenario tree method based on day-ahead load and wind power forecasting errors is proposed to solve the probabilistic dispatch problem. A hybrid-microgrid, comprising a wind turbine, two diesel generators, and an energy storage system, has been used to analyze the proposed method. Utilizing a multi-price electricity tariff has been shown as the advantage of the energy storage system and the energy arbitrage. A new form of formulation for power and energy arbitrage equations and economic benefits of energy arbitrage is provided. The probabilistic power generation costs of the grid are calculated. With simulation results and comparison of different strategies in the proposed hybrid-Microgrid model illustrated the effectiveness and advantages of the proposed approach using MATLAB. By using energy arbitrage formulation in the probabilistic economic dispatch, the day-ahead probabilistic active power of the grid is calculated and depicted in a new way. The proposed method in many micro-grids, including probabilistic power generators, can be used to reduce the cost of exploiting distribution networks. [ABSTRACT FROM AUTHOR]

Interconnected microgrids or a microgrid cluster (MGC) can enable mutual power support among microgrids and can improve the utilization of renewable energy sources. However, most of the distributed optimization focuses only on day-ahead or day-in dispatch, and few studies have attempted to study the integrated two-level distributed optimization. On the basis of the alternating direction method of multipliers (ADMM), this paper describes an efficient two-level distributed algorithm framework to solve multi-area ED problems in MGC considering unit commitment and transmission loss. Specifically, to overcome the non-convexity, a choice and comparison approach is introduced to cooperate with power-based ADMM for searching feasible binary variables in the first level, i.e., day-ahead dispatch. Voltage-based ADMM is adopted to obtain the relevant intra-day scheduling plan considering tie-line transmission loss among different areas in the second level. Moreover, a spinning reserve-based chance-constrained programming procedure is added to address uncertainty factors such as renewable energy fluctuations and load forecasting errors in isolated microgrids. Finally, a consensus-based cost allocation is proposed to fairly clear the expenses among microgrids. A case study with several networked microgrids is tested to demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

This paper deals with an efficient real-time estimation (ERES) method for Doubly Fed Induction Generator (DFIG) parameters like resistances and inductances. The developed estimation technique is a sensorless method and is robust regarding measurement noises. Moreover, it estimates the rotating speed and the fluxes and their dynamics. The proposed estimation technique can easily serve to enhance the performances and the robustness of the related controllers. This technique can be employed for grid/micro-grid connected generation systems like DFIG-based Wind Energy Conversion Systems (WECSs) or other Energy Conversion Systems (ENCSs). The new approach is developed based on a DFIG state-space model and the recursive least squares algorithm. Simulations verify the viability of the proposed approach for a 1.5 MW DFIG-based WECS in the presence of measurement noises and parameter variations. A comparative study is also provided to show the superiority of the ERES-based adaptive sensorless sliding mode control approach compared to the classical one. It is observed that the motor drive ensures driving with a reduced number of sensors, lower tracking errors, and lower number of voltage fluctuations in the rotor. [ABSTRACT FROM AUTHOR]

The continuous growth of electricity demands microgrid distribution networks that use distributed energy sources. Therefore, the necessity for applying efficient methods to improve the performance of microgrids as a part of the power industry is felt more and more day by day. Microgrids are usually installed at locations where access to the power grid is not economical due to the far distance and losses. In this paper, the optimal allocation of distributed generation (DG) units using the craziness-based particle swarm optimization (CRPSO) algorithm based on the game theoretic formulation strategy was proposed to reduce the exchange of power between the macrostation and microgrids. The main motive for presenting this research is the combination of the advantages of coalition game theory, DG units and the CRPSO optimization algorithm. The advantages have caused the load of the microgrids to be implemented as much as possible through the exchanges between the microgrids and the cost of utilization and power supply of the loads minimized. Uncertainties in wind speed and solar radiation flux are also considered for the purpose of applying the random property of the distributed generation resources. The simulation results on the 33rd IEEE standard system in different scenarios indicate the desired performance of the proposed method. [ABSTRACT FROM AUTHOR]

Energy systems face problems such as increasing energy cost, greenhouse (GHG) emissions, and network load deviations. A wide range of load variations may cause network instability. Large commercial buildings with a heavy load, treated as microgrids (μGs), may contribute meaningfully to demand response and network stability. Therefore, a strategy for optimal energy consumption and bidirectional power trade between the building and the utility grid is required for this class of customers. The objective of this paper is to develop a linear bi-level macro energy hub (MEH) control for a group of large commercial building μGs, to minimize the network load deviation, while satisfying the customer specified total μG energy cost and the GHG emission constraints. The resulting constrained optimization problem is mapped in a flower pollination algorithm environment under deterministic and uncertain conditions such as the random presence of solar irradiance and electric and natural gas networks. The μG considers the solar photovoltaic panel, internal combustion engine, fuel cell, microturbine, and electric vehicle lot and battery. Using the results obtained under deterministic conditions, customer-utility relationship is developed by applying the multi-linear regression analysis technique, between the network load deviation, operation and maintenance cost of distributed generators, μG's spinning reserve value, and total μG energy cost. Analysis proves that under an MEH, customer-utility interest is non-linearly related. Building owned micro energy hub control reduces the energy cost and the GHG emissions; however, the load deviation inflates and vice versa for utility owned MEH control. To address this nonlinearity, biogas is integrated with the MEH, using a method termed as integrated demand response (IDR), which successfully brings the energy cost down at a low load deviation. The results show that if uncertainties are not severe, the IDR alone will be adequate to bring the cost down for simultaneous achievement of end-to-end stakes. However, if the cost is high, due to adverse conditions, in addition to the IDR, the proposed load curtailment scheme is applied. Simulations demonstrate that the cost, GHG emission, and load deviation drop by 22.448%, 11.716%, and 36.410%, respectively, for biogas integrated MEH compared to building owned micro EH control. [ABSTRACT FROM AUTHOR]

The uncertainty introduced by intermittent renewable energy generation and prosumer energy imports makes operational planning of renewable energy-assisted prosumer microgrids challenging. This is due to the difficulty in obtaining accurate forecasts of the energy expected from these renewable energy sources and prosumers. Operators of such microgrids therefore require additional grid-balancing tools to maintain power supply and demand balance during grid operation. In this paper, the impact of demand response aggregators (DRAs) in a prosumer microgrid is investigated. This is achieved by developing and solving a deterministic mathematical formulation for the operational planning of the grid. Also, taking a cue from CAISO's proposed tariff revision which allows the state-of-charge of non-generator resources (like storage units) to be submitted as a bid parameter in the day-ahead market and permits scheduling coordinators of these resources to self-manage their energy limits and state-of-charge, the proposed formulation permits prosumers to submit the battery energy content as a bid parameter and self-manage their battery energy limits. Furthermore, a robust counterpart of the model is developed. Both formulations are constrained mixed integer optimization problems which are solved using the CPLEX solver in the Advanced Interactive Multidimensional Modelling System environment. The results obtained from tests carried out on a hypothetical prosumer microgrid show that the operating cost of the microgrid reduces in the presence of DRAs. In addition, the storage facility owner may benefit from self-managing its energy limits, but this may cut the amount of grid-balancing resource available to the microgrid operator, thereby increasing the operating cost of the microgrid. [ABSTRACT FROM AUTHOR]

Due to the abundance of power electronic appliances, harmonic distortion has become one of the growing power quality concerns in recent years. Harmonic distortion is also caused by the nonlinearity of other power equipment such as transformers and rotating machines. Power systems, rich in harmonics, offer poor power factor and hence low efficiency operation. Due to the large concentration of nonlinear loads such as motor drives, investigation of harmonic distortion in industrial systems is getting special attention. Widespread integration of solar Photovoltaic (PV) systems into distribution systems brings additional challenges to the existing power quality scenario. Inclusion of solar PV systems in an industrial microgrid equipped with a large share of motor drives results in a significant increase in existing Total Harmonic Distortion (THD). To better understand the above issues, a methodology is proposed in this paper to identify the most suitable PV placement topology between the two schemes (namely, centralized and distributed) in terms of THD performance. Also, a technique is developed to estimate the maximum PV penetration level by maintaining the THD within the given acceptable limit. The proposed methodologies are applied to an industrial distribution system under various operating conditions. [ABSTRACT FROM AUTHOR]

Distributed generation (DG) is gaining popularity because of prevalent environmental issues and the need for better power quality. To address these concerns, the microgrid was introduced to utilize DG. Inverter controls are essential for improving power quality because DG units are connected to a grid through an inverter. In this paper, an enhanced droop control technique is proposed to properly operate parallel-connected inverters in an AC microgrid. The proposed control strategy is based on the droop control technique, which utilizes locally measurable feedback signals. The droop control method is used for power sharing because of the absence of critical communication links among parallel-connected inverters. The transient response of the conventional droop control of power sharing is slow, and its frequency and voltage deviations are high. These factors limit accurate power sharing among parallel-connected inverters. The enhanced droop control is introduced to mitigate this problem, through which the trade-off between voltage regulation and reactive power sharing decreases significantly. In addition, the proposed method improves the transient response and achieves higher output power without voltage and frequency deviations. The results indicate that the desired transient response and accurate reactive power sharing with the line impedance mismatch are achieved. The simulation results are analyzed using MATLAB/Simlink to verify the effectiveness of the proposed control strategy. [ABSTRACT FROM AUTHOR]

Low inertia of distributed energy resources (DERs), high penetration levels of renewable energy sources (RESs), and load demand variations put the islanded microgrids' (MGs) security at the risk of instability. This paper proposes a two-stage stochastic model for coordination of DERs and responsive loads in islanded MGs with regard to voltage and frequency security constraints. Based on the proposed model, scheduling of the controllable units in both supply and demand sides is done in a way not only to maximize the expected profit of the MG operator (MGO) but also to minimize the energy payments of customers under the premise of security and stability of MG. An AC optimal power flow procedure is also used to study the operating condition of the system under uncertainties and to guarantee acceptable nodal voltages and system frequency under different scenarios. The proposed stochastic optimization model is then applied to a typical autonomous MG, and its effectiveness is demonstrated through different scenarios under uncertainties in load consumption and renewable energy resource (RES) productions. Simulation results demonstrate that customers' participation in DR programs has a significant effect on the system's performance in terms of voltage and frequency stability. Moreover, optimal coordination of DERs and responsive loads can increase the expected profit of MGO significantly. The effectiveness of the proposed scheduling approach is verified on an islanded MG test system over a 24-h period. [ABSTRACT FROM AUTHOR]

Microgrids with integrated renewable energy sources (RESs) serve as a clean and befitting solution to meet the ever increasing energy demand of countries across the globe. The power supply reliability is maintained through bidirectional power exchange between the main grid and the green microgrid. The exchange of power between the main grid and the microgrid is often challenging if the uncertainty in the availability of renewable sources and the influence of market prices is not taken into consideration. This work considers a green microgrid comprising solar, wind, microturbine, and fuel cell generation, along with a pumped hydrostorage system to cater to uncertainties associated with RES. A novel double layer optimization approach aimed at the social benefit maximization of both the main grid and the green microgrid, with optimal utilization of available RES and strategic management of pumped hydrostorage (in the green microgrid) during real time dispatch, is proposed in this paper. The scheduled power exchange is submitted to the independent system operator in the Day Ahead Schedule Layer based on the bids submitted to the microgrid central controller by load agents and generation agents. The real time dispatch layer deals with the intermittencies in RES by optimally managing the pump hydrostorage system. The developed optimization approach is implemented on a test network simulated as the green microgrid interconnected by an IEEE 30 bus system as the main grid. [ABSTRACT FROM AUTHOR]

This paper proposes a robust control algorithm for a photovoltaic (PV) multifunctional grid tied inverter (MFGTI). The MFGTI is used to simultaneously inject the active power of a PV panel and enhance the power quality of a microgrid. The proposed algorithm will estimate the positive fundamental component of the source current even under strongly disturbed main voltage and load conditions. First, a Sliding Mode Controller is used to estimate the magnitude of the source current. Then, a positive fundamental component estimator is used to estimate the unity vector of the main voltage. By multiplying the output of the two blocks, we obtain a pure sinusoidal source current in phase with the main voltage. Using this algorithm, the MFGTI will inject the active power, compensate the reactive power, and enhance the power quality of the microgrid. Thus, the main grid will only supply some amount of active power with a unity power factor. The proposed algorithm is compared with the conventional method to verify its effectiveness. Several loads and main voltage scenarios are simulated and then experimentally tested. The results prove the superiority and effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Wind power exhibits frequent fluctuations, which can impact the stability of the power grid. The impact of wind power on grid stability can be mitigated by using energy storage system whose battery lifespan can be extended by control strategies. This study focuses on the design of a modular multilevel hybrid energy storage system that incorporates both battery packs and supercapacitors. This system aims to effectively smooth the fluctuation of wind power and prolong the lifespan of battery packs. The power distribution between the battery packs and the supercapacitors is achieved through a first-order low-pass filter. The low-frequency power variations are smoothed using the battery packs, while the time constant of the filter is adjusted based on changes in the state of charge of battery packs. This approach prevents overcharging and overdischarging of the battery packs, thereby increasing their lifespan. The remaining high-frequency power variations are handled using the supercapacitors. Simulation results and a comparison with fixed time constant low-pass filtering demonstrate that the grid-connected power meets national standards following the implementation of the energy storage configuration. The proposed control strategy effectively allocates high- and low-frequency energy storage power, resulting in an 18.34% increase in battery pack lifespan. [ABSTRACT FROM AUTHOR]

This paper presents an optimization of grid energy using a supervisory control and data acquisition (SCADA) system for DC microgrids including distributed energy resources and residential building. The proposed system is mainly focused on distributed energy resources (DERs) for source side management (SSM) and demand side management (DSM). A SSM agent performs cheaper (less emission) source selection based on the microgrid demand. The real time scheduling updates the power settings of DERs and generates the electricity prices of the microgrid, whereas a DSM agent performs load shifting in real time, based on microgrid prices, level of stored energy in the appliance, and latest start time. The incorporation of SCADA provides the common communication for all components of the microgrid to interconnect with the control room via wireless smart sensors to update the power setting. Furthermore, the use of SSM maximizes the use of DERs and optimizes the power exchange between utility and the microgrid subjected to the system and DER constraints. The simulation results show a reduction in the peak demand and electricity bill of the building. Moreover, the obtained results show that the size of the energy storage and the power losses with the adaptive load are small as compared to the non-adaptive load. [ABSTRACT FROM AUTHOR]

With the consideration of a line resistance in a DC microgrid, the state of charge (SoC) of batteries will not be consistent with each other due to the voltage drop in the line resistance. In this paper, a model of a DC microgrid including detailed distributed generation models is established first, which can present the mechanism characteristic of different distributed generations. Then, a self-convergence droop control of no communication based on double-quadrant SoC is proposed to solve the problem modifying the charge/discharge rate with SoC itself and line resistance. After operating for a while by adopting this proposed method, the SoC of batteries in the DC microgrid can converge to the same value itself without need for a communication system to exchange data with each other. Meanwhile, the output/input power of batteries can be equalized. Therefore, the lifetime of batteries can be prolonged and the overuse of batteries can be avoided in this situation. Also, the system reliability is improved. Finally, the test results of a DC microgrid with a photovoltaic cell, a fuel cell, and two batteries in the RT-LAB real-time platform are shown to verify this approach. [ABSTRACT FROM AUTHOR]

Micro-grids are small-scale grids, equipped with energy management and on-site energy generation. They offer low-cost power to consumers and combined with green energy, they are a promising solution to build the future smart city. This paper discusses the different components of a micro-grid and their mathematical equations. The model is built in Matlab/Simulink and respects the distribution rules of the Moroccan grid code. The modeled residential group represents the power consumption of an average Moroccan household in the mountainous region of Ifrane. Two simulation scenarios were designed in order to validate the model. The first scenario tests the microgrid during normal operation, while the second scenario introduces a grid fault, requiring disconnection from the utility grid. Simulation results show that the model responds to the specified weather and system conditions as predicted. [ABSTRACT FROM AUTHOR]

Nine-switch converter has been recently known for its higher compactness with reduced switch count as compared to the back-to-back converter. The readily available dc-link capacitor made it an appropriate alternative for ac/dc/ac conversion in microgrids. Regardless of the complicated Pulse Width Modulation control scheme, this paper proposes a novel predictive power control for the nine-switch converter which is simple and allows concurrent control of real and reactive power flow between two islanded microgrids. The proposed control scheme adopts the virtual flux vectors and a newly defined reactive power to improve the power quality under the existence of grid voltage disturbances, i.e., voltage unbalance and voltage distortion. Simulation results of the proposed predictive power control guarantee decoupled active and reactive power controls, fast response, accurate power and dc-link voltage tracking, and sinusoidal current waveforms with low harmonic distortions under the various grid conditions. [ABSTRACT FROM AUTHOR]

With the emergent need for clean and reliable energy resources, hybrid energy systems, such as the microgrid, are widely adopted in the United States. A microgrid can consist of various distributed energy resources, for instance, combined heat and power (CHP) systems. The CHP module is a distributed cogeneration technology that produces electricity and recaptures heat generated as a by-product. It is an energy-efficient technology converting heat that would otherwise be wasted to valuable thermal energy. For an optimal system configuration, this study develops a novel co-design optimization framework for CHP-based cogeneration microgrids. The framework provides the stakeholder with a method to optimize investments and attain resilient operations. The proposed co-design framework has a mixed integer programming (MIP) model that outputs decisions for both plant designs and operating controls. The microgrid considered in this study contains six components: the CHP, boiler, heat recovery unit, thermal storage system, power storage system, and photovoltaic plant. After solving the MIP model, the optimal design parameters of each component can be found to minimize the total installation cost of all components in the microgrid. Furthermore, the online costs from energy production, operation, maintenance, machine startup, and disruption-induced unsatisfied loads are minimized by solving the optimal control decisions for operations. Case studies based on designing a CHP-based microgrid with empirical data are conducted. Moreover, we consider both nominal and disruptive operational scenarios to validate the performance of the proposed co-design framework in terms of a cost-effective, resilient system. [ABSTRACT FROM AUTHOR]

AC/DC hybrid microgrids (MGs) have been developed rapidly in recent years due to their efficient integration and management of large-scale distributed energy resources. How to achieve the optimal operation and coordinated control of AC/DC hybrid MGs has become crucial. A multi-agent system (MAS) based hierarchical and decentralized method for coordination operation of multiple AC/DC hybrid MGs is proposed. First, a structure of AC/DC hybrid MGs including wind turbines, photovoltaics, micro-hydro, and biogas generators is designed, which aims at reducing the power conversion loss. Second, a three-level MAS including distribution network (DN) level, MG level, and distributed generation (DG) level is presented. At the DG level, the DG agents obtain the optimal results from the MG agent. At the MG level, the independent optimization is implemented based on the mixed integer linear programming, and the consensus algorithm is utilized to realize the coordination of MG with adjacent MGs through the communication among MG agents when there is imbalance power. At the DN level, the DN agent performs power exchange with AC/DC hybrid MGs to balance the load demand when needed. Finally, the distributed energy management platform is developed by the Java Agent Development Framework platform to simulate the optimal operation of AC/DC hybrid MGs. The results show the rationality and availability of the proposed optimization method. [ABSTRACT FROM AUTHOR]

Short-term load forecasting is an important part in the energy management of micro-grids, and the forecasting error directly affects the economic efficiency of operation. Considering the strong randomness and high volatility of loads on the user side of micro-grids, a short-term load forecasting method based on Empirical Mode Decomposition (EMD), Extreme Learning Machine with different Kernels, and Switching Delayed Particle Swarm Optimization (SDPSO) is proposed. First, the history load dataset is decomposed into several independent Intrinsic Mode Functions (IMFs) by EMD, and the sample entropy values of the IMFs are calculated. According to the approximation degree of sample entropy values, the IMFs are divided into three categories. Then, ELM with different kernels is adopted to forecast the three categories. Finally, the prediction results are summed to obtain the final prediction result. SDPSO is used to optimize the relevant parameters in the forecasting model. Three micro-grids with different capacities are used to verify the model, and the experimental results demonstrate that the proposed forecasting model can provide better accuracy than the other two methods. The proposed model can provide practical reference for the efficient operation of micro-grids. [ABSTRACT FROM AUTHOR]

We study the spreading of renewable power fluctuations through grids with Ohmic losses on the lines. By formulating a network-adapted linear response theory, we find that vulnerability patterns are linked to the left Laplacian eigenvectors of the overdamped eigenmodes. We show that for tree-like networks, fluctuations are amplified in the opposite direction of the power flow. This novel mechanism explains vulnerability patterns that were observed in previous numerical simulations of renewable microgrids. While exact mathematical derivations are only possible for tree-like networks with a homogeneous response, we show that the mechanisms discovered also explain vulnerability patterns in realistic heterogeneous meshed grids by studying the IEEE RTS-1996 test system. [ABSTRACT FROM AUTHOR]

Considering that traditional centralized scheduling cannot realize energy complementarity and benefit coordination between multi-energy hubs (EHs) and advantages of artificial intelligence technology in the application of integrated energy microgrid (IEM), a coordinated scheduling model and method of IEM with multi-EHs based on multi-agent deep deterministic policy gradient (MADDPG) is proposed. First, an IEM framework of multi-EHs is constructed, and the rationality of multi-EH construction is illustrated. Then, considering operation cost, environmental cost, and benefit of new energy, a two-layer economic optimal scheduling model of IEM is established. Furthermore, distributed deep reinforcement learning based on MADDPG and game theory is introduced, and a coordinated scheduling method of IEM with multi-EHs based on the algorithm is proposed. At the same time, based on the idea of transfer learning, an off-line training and online learning method is proposed, which can improve the training speed of online learning. Finally, a numerical example is constructed to verify the effectiveness of the proposed model and method. [ABSTRACT FROM AUTHOR]