Fig 1 - uploaded by Mudhafar Al-saadi
Content may be subject to copyright.
A 48 V DC Autonomous microgrid with () number of distributed BESS agents, under a neighbor-to-neighbor multiagent communication.
Source publication
Multiagent reinforcement learning has proven remarkably effective at finding near-optimal solutions to complex non-linear control problems when compared to classical schemes. Such problems typically arise when considering power management problems related to advanced power distribution applications, such as micro/smart grids, smart buildings, elect...
Contexts in source publication
Context 1
... study considers the implementation of the new MARL adaptation on the number of BESSs, in the 48 V DC autonomous microgrid demonstrated in Fig. 1. Where a 24-hour excessive load variation and a 24-hour PV generation profile are existing. Accordingly, each BESS is formulated as an independent agent, balancing local output voltage for the mandatory load participation. In addition, collaborate in fulfilling an overall balanced power flow of the microgrid. Wherein the aim is to ...
Context 2
... The Decentralized Local Control: A 2-stage control was locally formulated to manage a 24V lithium-ion battery, as shown in Fig. 1. Accordingly, the output voltage () is balanced to under the mandatory load sharing. Where the role of the first-stage is to attain a real-time battery current reference ( __ℎ______) based on the error ( ____) between the immediate real-time adapted droop correction (__) and as explained in (1), (2), and the demonstration of the ...
Context 3
... (), multiplied by a nominated droop coefficient (), as explained in (5), and Fig. 3. Then an adaptive real-time reference for the local controller () is accomplished based on the average real-time droop drop due to the participation of the load demand of the neighbours' BESSs, (___(((_)_*) including the local (___(((___*), as demonstrated in (6), Fig. 1, and Fig. 3, is the number of the neighbours' BESSs. Hence, the requested load demand at any BESS is implemented collaboratively by all the BESSs. Therefore, accurate charge-discharge synchronization is accomplished. Thus, a reduced circulating current and overloading is verified and reflected in better, control stability and enhanced batteries' ...
Context 4
... aim of consensus correction here is the exploitation of multiagent neighbor-to-neighbor communication presented in Fig. 1 in establishing immediate real-time correction protocol. Accordingly, a correction is verified against deviations of the voltage and current, before sharing with the neighbors' BESSs to be corrected collaboratively [17]. Hence, a real-time balanced output voltage is attained, under decentralized control and ...
Context 5
... (__;;ℎ) and neighbours' ()__;;ℎ), current consensus correction. Then multiplying by the consensus gain ( ;;) , dividing by the number of neighbours' BESSs (), and then adding the correction to the measured current () to correct the participation level, before sending it to the neighbours to be corrected collaboratively, as explained in (15), and Fig. 3 [15][18]. ...
Context 6
... verify the success of the suggested adaptation and discuss the performance, the adapted control strategy was implemented on the distributed BESSs of the 48 V DC autonomous microgrid demonstrated in Fig. 1. The system parameters are given in Table I. A 24-hour excessively continuous load variation was implemented collaboratively by the distributed BESSs and a 24-hour PV generation. Furthermore, the existence of multiagent neighbor-toneighbor communication. Several case studies were conducted considering a variated number of BESSs, and ...
Citations
... Hence, the accuracy tends to reduce with the increase in the real-time utilization of the battery's capacity, constraining the effective capacity of the storage systems to maintain balance. Without artificial capacity constraints, the circulating current and temporary overloading of some storage systems in a network are existing drawbacks that upset the optimization and steadiness of the control, reduce the balance and sustainability of the power flow, deteriorate the health and life of the batteries, and limit the introduction and buffering capability of renewable energy [14][15][16]. DC infrastructure influences are a crucial magnifier for raising the impact of these drawbacks in real operation, potentially leading to the disparity of load participation due to the impact of the influence of the power electronics and transmission lines. This violates the charge-discharge synchronization accuracy if not adequately dealt with when designing the control system, and leads to the hypothesis that the compensation of these influences may lead to an effective capacity increase in the storage systems and improved performance under the MARL-based control. ...
... The accomplished control action is based on the current error (ei_pri) of IB_ref_Ch_Dis_i from the real-time locally measured battery current (Ib_i). This denotes the requested battery charge or discharge to verify the mandatory power-flow balance, as shown in (11) and (12) Table 1 [15,37,38]. ...
... The accomplished control action is based on the current error (ei_pri) of IB_ref_Ch_Dis_i from the real-time locally measured battery current (Ib_i). This denotes the requested battery charge or discharge to verify the mandatory power-flow balance, as shown in (11) and (12) Table 1 [15,37,38]. Implicitly, the primary local regulation of the power-storage flow is under the droop correction regulatory control with a supervisory trim signal. ...
The influence of the DC infrastructure on the control of power-storage flow in micro-and smart grids has gained attention recently, particularly in dynamic vehicle-to-grid charging applications. Principal effects include the potential loss of the charge-discharge synchronization and the subsequent impact on the control stabilization, the increased degradation in batteries' health/life, and resultant power-and energy-efficiency losses. This paper proposes and tests a candidate solution to compensate for the infrastructure effects in a DC microgrid with a varying number of heterogeneous battery storage systems in the context of a multiagent neighbor-to-neighbor control scheme. Specifically, the scheme regulates the balance of the batteries' load-demand participation, with adaptive compensation for unknown and/or time-varying DC infrastructure influences. Simulation and hardware-in-the-loop studies in realistic conditions demonstrate the improved precision of the charge-discharge synchronization and the enhanced balance of the output voltage under 24 h excessively continuous variations in the load demand. In addition, immediate real-time compensation for the DC infrastructure influence can be attained with no need for initial estimates of key unknown parameters. The results provide both the validation and verification of the proposals under real operational conditions and expectations, including the dynamic switching of the heterogeneous batteries' connection (plug-and-play) and the variable infrastructure influences of different dynamically switched branches. Key observed metrics include an average reduced convergence time (0.66-13.366%), enhanced output-voltage balance (2.637-3.24%), power-consumption reduction (3.569-4.93%), and power-flow-balance enhancement (2.755-6.468%), which can be achieved for the proposed scheme over a baseline for the experiments in question.
... However, loss of accuracy of coordinated charge-discharge operations for battery energy storage systems (BESSs) has been an existing issue for MARL approaches, particularly under sudden high or excessive continuous load variations and due to operational/infrastructural influences. Loss of accuracy in this way can also introduce increases in circulating eddies currents, and accidental storage overloading, negatively affecting system stabilization, power flow balance, batteries' health/life, and overall efficiency, including useable yield from local renewable energy sources [6], [8]- [10]. Environmental and operational variability and infrastructural influences such as a variable number of heterogeneous batteries, each with varying state-of-health/degradation behaviour and switch-on charge levels, subject to different temperature disparities, and with localized infrastructure and load effects, along with current and previous faults are generally outside of the designer's control. ...
... Hence, an immediate real-time regional multiagent power consumption/loss of load demand implementation (PLN), and DC power losses at connection (PTN_loss), load (PLN_loss), and BESS (PBN_loss), branches can be measured respectively, as explained in (6), (7), (8), and (9). ...
... Hence, a collaborative correction role is completed of voltage (4 ), participation current ( 4 5 ), and state of charge ( 4 678_ ), through a qualified correction platform. Thus, charge/discharge is requested to maintain the balance to 9:, as presented in (20) [5][6] [8]. (20) Accordingly, a decentralized secondary voltage correction was designed to determine an action (4 ), for deviations of the real-time consensus correction ( _ ; ℎ) from 9:. ...
