Decentralized Control Architecture Research Articles

Increased Dynamic Drivetrain Performance by Implementing a Modular Design with Decentralized Control Architecture

This paper assesses the energy consumption, control performance, and application-specific functional requirements of a modular drivetrain in comparison to a benchmark drivetrain. A decentralised control architecture has been developed and validated using mechanical plant models. Simscape models have been validated with data from an experimental setup including an equivalent modular and benchmark drivetrain. In addition, the control strategy has been implemented and validated on the experimental setup. The results prove the ability of the control strategy to synchronize the motion of the different sliders, resulting in crank position tracking errors below 0.032 radians on the setup. The model and experimental data show an increased performance of the modular drivetrain compared to the benchmark drivetrain in terms of energy consumption, control performance, and functional requirements. The modular drivetrain is especially advantageous for machines running highly dynamic motion profiles due to the reduced inertia. For such motion profiles, an increased position tracking of up to 84% has been measured. In addition, it is shown that the modular drivetrain root mean square (RMS) torque is reduced with 32% compared to the benchmark drivetrain. However, these mechanical energy savings are partly counteracted by the higher motor losses seen in the modular drivetrain, resulting in potential electrical energy savings of around 29%.

A Novel Electric Vehicle Battery Management System Using an Artificial Neural Network-Based Adaptive Droop Control Theory

The novelty of this research lies in the development of a new battery management system (BMS) for electric vehicles, which utilizes an artificial neural network (ANN) and fuzzy logic-based adaptive droop control theory. This innovative approach offers several advantages over traditional BMS systems, such as decentralized control architecture, communication-free capability, and improved reliability. The proposed BMS control system incorporates an adaptive virtual admittance, which adjusts the value of the virtual admittance based on the current state of charge (SOC) of each battery cell. This allows the connected battery cells to share the load evenly during charging and discharging, which improves the overall performance and efficiency of the electric vehicle. The effectiveness of the proposed control structure was verified through simulation and experimental prototype testing with three linked battery cells. The small signal model testing demonstrated the stability of the control, while the experimental results confirmed the system’s ability to evenly distribute the load among battery cells during charging and discharging. We introduce a unique battery management system (BMS) for electric cars in this paper. Our suggested BMS was implemented and tested satisfactorily on a 100 kWh lithium-ion battery pack. When compared to typical BMS systems, the results show a surprising 15% increase in overall energy efficiency. Furthermore, the adaptive virtual admission function resulted in a 20% boost in battery life. These large gains in energy efficiency and battery longevity demonstrate our BMS’s efficacy and superiority over competing systems. Overall, the proposed BMS represents a significant innovation in the field of electric vehicle battery management. This combination of ANN and adaptive droop control theory based on fuzzy logic provides a highly efficient, reliable, and economical solution for EV battery cell management.

Counterfactual learning in enhancing resilience in autonomous agent systems.

Resilience in autonomous agent systems is about having the capacity to anticipate, respond to, adapt to, and recover from adverse and dynamic conditions in complex environments. It is associated with the intelligence possessed by the agents to preserve the functionality or to minimize the impact on functionality through a transformation, reconfiguration, or expansion performed across the system. Enhancing the resilience of systems could pave way toward higher autonomy allowing them to tackle intricate dynamic problems. The state-of-the-art systems have mostly focussed on improving the redundancy of the system, adopting decentralized control architectures, and utilizing distributed sensing capabilities. While machine learning approaches for efficient distribution and allocation of skills and tasks have enhanced the potential of these systems, they are still limited when presented with dynamic environments. To move beyond the current limitations, this paper advocates incorporating counterfactual learning models for agents to enable them with the ability to predict possible future conditions and adjust their behavior. Counterfactual learning is a topic that has recently been gaining attention as a model-agnostic and post-hoc technique to improve explainability in machine learning models. Using counterfactual causality can also help gain insights into unforeseen circumstances and make inferences about the probability of desired outcomes. We propose that this can be used in agent systems as a means to guide and prepare them to cope with unanticipated environmental conditions. This supplementary support for adaptation can enable the design of more intelligent and complex autonomous agent systems to address the multifaceted characteristics of real-world problem domains.

Decentralized control architecture for multi-authoring microgrids

Decentralized control architecture for multi-authoring microgrids

Energy management and harmonic compensation of micro‐grids via multi‐agent systems based on decentralized control architecture

AbstractHere, a new energy management strategy based on a multi‐agent structure is presented by introducing hybrid control of AC micro‐grid (MG), where current harmonics compensation is also considered. This energy management strategy is based on Mealy machine in a multi‐agent system (MAS), which is implemented by generating coordination signals and sending them to the renewable energy sources (RES). In RES, maximum power point tracking (MPPT) or reverse of Hill‐climbing (RHC) is used based on these received coordination signals to limit the exchange power between the MG and the grid. Also, the multi‐agent structure of shunt active power filters (SAPFs) and passive filters have been studied in power MGs. The comparison of MG current harmonic compensation by these filters by fast Fourier transform (FFT) algorithm has been investigated in the proposed network. The simulation results show that the proposed energy management system (EMS) with proper management of MG different agents has been able to have a suitable performance for hybrid distributed generation (DG) systems. Also, the SAPF with better and more appropriate harmonic compensation than the SC‐RC‐LCL passive filter can eliminate the harmonic pollution of MG and be used with high performance in hybrid DG systems.

Decentralized Model-Reference Adaptive Control Based Algorithm for Power Systems Inter-Area Oscillation Damping

Being the primary cause of inter-area oscillations and due to the fact that they limit the generation’s output, Low-Frequency Electromechanical Oscillations (LFEOs) represent a real threat to power system networks. Mitigating their effects is therefore crucial as it may lead to system collapse if not properly damped. As rotor angle instability is the primary cause of LFEOs, this paper presents a novel Model-Reference Adaptive Control (MRAC) scheme that enhances its stability. The proposed scheme is tested using the Single-Machine Infinite Bus (SMIB) network. The results obtained validate the proposed decentralized control architecture. The robustness of this oscillation damping controller is verified through simulations in MATLAB/SIMULINK. With Gaussian noise added to the structure of the generator to emulate small load variations responsible for the rotor angle instability, the results of the simulations show that the rotor angle remains stable. Furthermore, when subjected to faults, the recovery time is less than 500 ms.

Review of Electronics-Free Robotics: Toward a Highly Decentralized Control Architecture

In recent years, conventional model-based motion control has become more challenging owing to the continuously increasing complexity of areas in which robots must operate and navigate. A promising approach for solving this issue is by employing interaction-based robotics, which includes behavior-based robotics, morphological computations, and soft robotics that generate control and computation functions based on interactions between the robot body and environment. These control strategies, which incorporate the diverse dynamics of the environment to generate control and computation functions, may alleviate the limitations imposed by the finite physical and computational resources of conventional robots. However, current interaction-based robots can only perform a limited number of actions compared with conventional robots. To increase the diversity of behaviors generated from body–environment interactions, a robotic body design methodology that can generate appropriate behaviors depending on the various situations and environmental stimuli that arise from them is necessitated. Electronics-free robotics is reviewed herein as a paradigm for designing robots with control and computing functions in each part of the body. In electronics-free robotics, instead of using electrical sensors or computers, a control system is constructed based on only mechanical or chemical reactions. Robotic bodies fabricated using this approach do not require bulky electrical wiring or peripheral circuits and can perform control and computational functions by obtaining energy from a central source. Therefore, by distributing these electronics-free controllers throughout the body, we hope to design autonomous and highly decentralized robotic bodies than can generate various behaviors in response to environmental stimuli. This new paradigm of designing and controlling robot bodies can enable realization of completely electronics-free robots as well as expand the range of conventional electronics-based robot designs.

Energy management and control system for microgrid based wind-PV-battery using multi-agent systems

Energy generation is currently evolving into a smart distribution system that incorporates several green energy resources at a distributed level, ensuring that clean energy is generated without releasing harmful gases, that operational procedures are consistent, and that energy management and supervision arrangements are improved. This paper proposes a multi-agent system-based microgrid energy management and proper control in distributed systems. For the complexity of energy management in distributed systems, a multi-agent system-based decentralized control architecture was developed. The proposed technique is based on several smart agents, each agent is based on the microgrid data for energy management and frequency control. The proposed energy management system based on the multi-agent system was tested by simulation under renewable resource fluctuations and seasonal load demand. The simulation results show that the proposed energy management system proved to be more resilient and high-performance controls than conventional centralized energy control systems.

AT-DIFC + : Toward Adaptive and Trust-Aware Decentralized Information Flow Control

Modern software systems and their corresponding architectures are increasingly decentralized, distributed, and dynamic. As a consequence, decentralized mechanisms are required to ensure security in such architectures. Decentralized Information Flow Control (DIFC) is a mechanism to control information flow in distributed systems. This article presents and discusses several improvements to an adaptive decentralized information flow approach that incorporates trust for decentralized systems to provide security. Adaptive Trust-Aware Decentralized Information Flow (AT-DIFC + ) combines decentralized information flow control mechanisms, trust-based methods, and decentralized control architectures to control and enforce information flow in an open, decentralized system. We strengthen our approach against newly discovered attacks and provide additional information about its reconfiguration, decentralized control architectures, and reference implementation. We evaluate the effectiveness and performance of AT-DIFC + on two case studies and perform additional experiments and to gauge the mitigations’ effectiveness against the identified attacks.

Optimal Renewable Resource Allocation and Load Scheduling of Resilient Communities

This paper presents a methodology for enhancing community resilience through optimal renewable resource allocation and load scheduling in order to minimize unserved load and thermal discomfort. The proposed control architecture distributes the computational effort and is easier to be scaled up than traditional centralized control. The decentralized control architecture consists of two layers: The community operator layer (COL) allocates the limited amount of renewable energy resource according to the power flexibility of each building. The building agent layer (BAL) addresses the optimal load scheduling problem for each building with the allowable load determined by the COL. Both layers are formulated as a model predictive control (MPC) based optimization. Simulation scenarios are designed to compare different combinations of building weighting methods and objective functions to provide guidance for real-world deployment by community and microgrid operators. The results indicate that the impact of power flexibility is more prominent than the weighting factor to the resource allocation process. Allocation based purely on occupancy status could lead to an increase of PV curtailment. Further, it is necessary for the building agent to have multi-objective optimization to minimize unserved load ratio and maximize comfort simultaneously.

Cyber-security of centralized, decentralized, and distributed control-detector architectures for nonlinear processes

Decentralized and distributed control systems provide an efficient solution to many challenges of controlling large-scale industrial processes. With the expansion in communication networks, vulnerability to cyber intrusions also increases. This work investigates the effect of different types of standard cyber-attacks on the operation of nonlinear processes under centralized, decentralized, and distributed model predictive control (MPC) systems. The robustness of the decentralized control architecture over distributed and centralized control architectures is analyzed. Moreover, a machine-learning-based detector is trained using sensor data to monitor the cyber security of the overall system. Specifically, detectors built using feed-forward neural networks are used to detect the presence of an attack or identify the subsystem being attacked. A nonlinear chemical process example is simulated to demonstrate the robustness of decentralized control architectures and the effectiveness of the neural-network detection scheme in maintaining the closed-loop stability of the system.

Soft pneumatic actuator-driven origami-inspired modular robotic “pneumagami”

This article presents a new modular robotic platform for enabling reconfigurable, actively controlled, high-degree-of-freedom (high-DoF) systems with compact form factor. The robotic modules exploit the advantages of origami-inspired construction methods and materials, and soft pneumatic actuators (SPAs) to achieve an actuator embedded, parallel kinematic mechanism with three independently controlled “waterbomb” base legs. The multi-material, layer-fabricated body of the modules features selectively compliant flexure hinge elements between rigid panels that define the module as a kinematic 6R spherical joint. The precision layer-fabrication technique is also used to form embedded distribution channels within the module base to connect actuators to onboard control hardware. A decentralized control architecture is applied by integrating each module with small-scale solenoid valves, communication electronics, and sensors. This design approach enables a single pneumatic supply line to be shared between modules, while still allowing independent control of each leg joint, driven by soft, inflatable pouch actuators. A passive pneumatic relay is also designed and incorporated in each module to leverage the coupled, inverted inflation, and exhaust states between antagonistic actuator pairs allowing both to be controlled by a single solenoid valve. A prototype module is presented as the first demonstration of integrated modular origami and SPA design, or pneumagami, which allows predefined kinematic structural mechanisms to locally prescribe specific motions by active effect, not just through passive compliance, to dictate task space and motion. The design strategy facilitates the composition of lightweight, high-strength robotic structures with many DoFs that will benefit various fields such as wearable robotics.

Double-Loop Multi-Scale Control using Routh-Hurwitz Dimensionless Parameter Tuning for MIMO Processes

This paper presents a new approach to controlling MIMO processes by using the double-loop multi-scale control scheme in the decentralized control architecture. The decentralized PID control system has been used in process industry despite its several limitations due to process interactions, time-delays and right half plane poles. To overcome the performance limitation due to process interactions, decoupling controllers are often added to the decentralized PID control system. The proposed strategy based on the double-loop multi-scale control scheme has some advantages over the existing control strategies for MIMO processes. An advantage of the proposed scheme over the decentralized PID control with decoupling system is that, the proposed strategy has a fixed number of dimensionless tuning parameters that are easy to tune. For an n×n MIMO process, the proposed scheme requires the tuning of only 3 to 6 dimensionless parameters instead of the 3n original PID parameters.

Implementation of a partially decentralized control architecture using wireless active sensors

This paper deals with the development of a vibration control architecture in which many stand-alone active dampers are involved. Each device is able to independently perform the control task, since it is embedded with sensors, an inertial actuator and a microcontroller, in which the control algorithm is implemented. The aim of this architecture is to improve the performance of a decentralized control strategy through a partial sharing of data between devices. The proposed control solution is based on the optimal control theory; the Linear Quadratic Regulator works with the full-state of the system, which is not available in many applications. In order to provide the entire state vector, state estimation is implemented in each device. A state recovery algorithm is adopted to improve the quality of the estimation without placing a hefty burden on the wireless channel. Numerical analysis is made in order to study the advantages of this method. Finally, the proposed solution is validated with experimental results from a clamped–clamped beam.

Non-Linear Primary Control Mapping for Droop-Like Behavior of Microgrid Systems

Interconnecting microgrids in LV power system presents appealing features such as self-healing or power quality. When networked microgrids are not connected to a strong utility grid, their Point of Common Coupling (PCC) voltage and their power reserves vary with the operating point. An external droop control architecture is proposed that allows active and reactive power sharing among the different microgrids, thereby stabilizing the system frequency and PCC voltage, and the maximum achievable droop gains are supplied. Next, the design of appropriate primary controllers for the Distributed Energy Resources inside each microgrid allows to achieve a specified aggregated external droop controller at the connection point with little communication requirements. This methodology is applied to a modified CIGRE benchmark and shows good results while keeping a standard decentralized control architecture.

Three-Loop-Based Universal Control Architecture for Decentralized Operation of Multiple Inverters in an Autonomous Grid-Interactive Microgrid

This article presents a decentralized control architecture for inverters in a utility-interactive autonomous microgrid involving multiple inverter units. The architecture is novel as all the system requirements, viz., automatic/seamless-mode transitions between grid-tied to standalone mode of operation and vice versa, islanding/anti-islanding detection, low-voltage ride-through, and dynamic grid support by reactive-power injection, are organized into three cascaded control loops-voltage control, frequency control, and a PLL. The control loops are designed to operate in the synchronous reference frame so that the control structure becomes compact and free from any computational complexities and associated delays. An equivalent droop combining virtual resistance with the active (P) and reactive (Q) power-based droops is used to achieve proportional load-sharing among the inverters in the standalone mode. The references to these loops are self-modified by the controller according to the operational mode. The architecture is autonomous but straightforward, where each inverter operates on its terminal information eliminating the requirement of any centralized controller or intercommunication network. The performance of the proposed control strategy is validated through comprehensive experimental results on a laboratory prototype that tests all the functions of an autonomous microgrid.

Design, Application, and Evaluation of a Multiagent System in the Logistics Domain

The increasing demand for flexibility of automated production systems also affects the automated material flow systems (aMFS) they contain and demands reconfigurable systems. However, the centralized control concept usually applied in aMFS hinders an easy adaptation, as the entire control software has to be re-tested, when manually changing sub-parts of the control. As adaption and subsequent testing are a time-consuming task, concepts for splitting the control from one centralized to multiple, decentralized control nodes are required. Therefore, this paper presents a holistic agent-based control concept for aMFS, whereby the system is divided into so-called automated material flow modules (aMFM), each being controlled by a dedicated module agent. The concept allows the reconfiguration of aMFS, consisting of heterogeneous, stationary aMFM, during runtime. Furthermore, it includes aspects such as uniform agent knowledge bases through metamodel-based development, a communication ontology considering different information types and properties, strategic route optimization in decentralized control architecture and a visualization concept to make decisions of the module agents comprehensible to operators and maintenance staff. The evaluation of the concept is performed by means of material flow simulations as well as a prototypical implementation on a lab-sized demonstrator.

Distributed Learning of Decentralized Control Policies for Articulated Mobile Robots

State-of-the-art distributed algorithms for reinforcement learning rely on multiple independent agents, which simultaneously learn in parallel environments while asynchronously updating a common, shared policy. Moreover, decentralized control architectures (e.g., CPGs) can coordinate spatially distributed portions of an articulated robot to achieve system-level objectives. In this work, we investigate the relationship between distributed learning and decentralized control by learning decentralized control policies for the locomotion of articulated robots in challenging environments. To this end, we present an approach that leverages the structure of the asynchronous advantage actor-critic (A3C) algorithm to provide a natural means of learning decentralized control policies on a single articulated robot. Our primary contribution shows individual agents in the A3C algorithm can be defined by independently controlled portions of the robot's body, thus enabling distributed learning on a single robot for efficient hardware implementation. We present results of closed-loop locomotion in unstructured terrains on a snake and a hexapod robot, using decentralized controllers learned offline and online respectively. Preprint of the paper submitted to the IEEE Transactions in Robotics (T-RO) journal in October 2018, and accepted for publication as a regular paper in May 2019.

Reliability Analysis of a Decentralized Microgrid Control Architecture

Reliability enhancement of microgrids is challenged by environmental and operational failures. Centrally controlled microgrids are susceptible to failures at high probability due to a single-point-of-failure, e.g., the central controller. True decentralization of microgrid architecture entails elimination of the central controller, attaining a parallel configuration for the system. In this paper, decentralized microgrid control architecture is proposed as a solution for reliability degradation over time, and analyzes the reliability aspects of centralized and decentralized control architectures for microgrids. Degree of importance of a single controller in centralized and decentralized architectures is determined and validated by Markov chain models. Results confirm that higher reliability is achieved when true decentralization of control architecture has been adopted. Challenges of implementing a true decentralized control architecture are discussed. Hardware-In-the-Loop simulation results for microgrid controller failure scenarios for both architectures are presented and discussed.

Hierarchical optimisation strategy for energy scheduling and volt/var control in autonomous clusters of microgrids

This study proposes a hierarchical optimisation strategy for the energy dispatch and volt/var control problem of a photovoltaic-battery microgrid cluster (MGC) operating autonomously. The proposed approach takes advantage of the decentralised control architecture existing in multi-microgrids (MMGs) framework by distributing the management responsibilities between the microgrid central controllers (MGCCs) and the central autonomous management controller (CAMC). In the first stage, the optimisation strategy solves a multi-temporal active power scheduling problem for the MGC based on consumption and generation forecasts. In the second stage, the reactive power and volt/var control are addressed by taking into account the medium-voltage (MV) and low-voltage levels independently. For this purpose, each MGCC computes the V(Q) capability area of operation at the boundary bus with the MV grid. Then, the CAMC performs an optimal power flow at the MV level for each time step, whose results at the boundary bus are considered in the last stage to schedule reactive power at the MGCC level. The effectiveness of the proposed strategy is demonstrated in a cluster of three microgrids. It keeps the modularity, interoperability and scalability characteristics of the MMG concept by clearly defining the roles and the information to be exchanged between the CAMC and the MGCC.