Reinforcement learning(RL) has roots in dynamic programming and it is called adaptive/approximate dynamic programming(ADP) within the control community. This paper reviews recent developments in ADP along with RL and ...Reinforcement learning(RL) has roots in dynamic programming and it is called adaptive/approximate dynamic programming(ADP) within the control community. This paper reviews recent developments in ADP along with RL and its applications to various advanced control fields. First, the background of the development of ADP is described, emphasizing the significance of regulation and tracking control problems. Some effective offline and online algorithms for ADP/adaptive critic control are displayed, where the main results towards discrete-time systems and continuous-time systems are surveyed, respectively.Then, the research progress on adaptive critic control based on the event-triggered framework and under uncertain environment is discussed, respectively, where event-based design, robust stabilization, and game design are reviewed. Moreover, the extensions of ADP for addressing control problems under complex environment attract enormous attention. The ADP architecture is revisited under the perspective of data-driven and RL frameworks,showing how they promote ADP formulation significantly.Finally, several typical control applications with respect to RL and ADP are summarized, particularly in the fields of wastewater treatment processes and power systems, followed by some general prospects for future research. Overall, the comprehensive survey on ADP and RL for advanced control applications has d emonstrated its remarkable potential within the artificial intelligence era. In addition, it also plays a vital role in promoting environmental protection and industrial intelligence.展开更多
Aiming at the tracking problem of a class of discrete nonaffine nonlinear multi-input multi-output(MIMO) repetitive systems subjected to separable and nonseparable disturbances, a novel data-driven iterative learning ...Aiming at the tracking problem of a class of discrete nonaffine nonlinear multi-input multi-output(MIMO) repetitive systems subjected to separable and nonseparable disturbances, a novel data-driven iterative learning control(ILC) scheme based on the zeroing neural networks(ZNNs) is proposed. First, the equivalent dynamic linearization data model is obtained by means of dynamic linearization technology, which exists theoretically in the iteration domain. Then, the iterative extended state observer(IESO) is developed to estimate the disturbance and the coupling between systems, and the decoupled dynamic linearization model is obtained for the purpose of controller synthesis. To solve the zero-seeking tracking problem with inherent tolerance of noise,an ILC based on noise-tolerant modified ZNN is proposed. The strict assumptions imposed on the initialization conditions of each iteration in the existing ILC methods can be absolutely removed with our method. In addition, theoretical analysis indicates that the modified ZNN can converge to the exact solution of the zero-seeking tracking problem. Finally, a generalized example and an application-oriented example are presented to verify the effectiveness and superiority of the proposed process.展开更多
This paper mainly focuses on the development of a learning-based controller for a class of uncertain mechanical systems modeled by the Euler-Lagrange formulation.The considered system can depict the behavior of a larg...This paper mainly focuses on the development of a learning-based controller for a class of uncertain mechanical systems modeled by the Euler-Lagrange formulation.The considered system can depict the behavior of a large class of engineering systems,such as vehicular systems,robot manipulators and satellites.All these systems are often characterized by highly nonlinear characteristics,heavy modeling uncertainties and unknown perturbations,therefore,accurate-model-based nonlinear control approaches become unavailable.Motivated by the challenge,a reinforcement learning(RL)adaptive control methodology based on the actor-critic framework is investigated to compensate the uncertain mechanical dynamics.The approximation inaccuracies caused by RL and the exogenous unknown disturbances are circumvented via a continuous robust integral of the sign of the error(RISE)control approach.Different from a classical RISE control law,a tanh(·)function is utilized instead of a sign(·)function to acquire a more smooth control signal.The developed controller requires very little prior knowledge of the dynamic model,is robust to unknown dynamics and exogenous disturbances,and can achieve asymptotic output tracking.Eventually,co-simulations through ADAMS and MATLAB/Simulink on a three degrees-of-freedom(3-DOF)manipulator and experiments on a real-time electromechanical servo system are performed to verify the performance of the proposed approach.展开更多
For unachievable tracking problems, where the system output cannot precisely track a given reference, achieving the best possible approximation for the reference trajectory becomes the objective. This study aims to in...For unachievable tracking problems, where the system output cannot precisely track a given reference, achieving the best possible approximation for the reference trajectory becomes the objective. This study aims to investigate solutions using the Ptype learning control scheme. Initially, we demonstrate the necessity of gradient information for achieving the best approximation.Subsequently, we propose an input-output-driven learning gain design to handle the imprecise gradients of a class of uncertain systems. However, it is discovered that the desired performance may not be attainable when faced with incomplete information.To address this issue, an extended iterative learning control scheme is introduced. In this scheme, the tracking errors are modified through output data sampling, which incorporates lowmemory footprints and offers flexibility in learning gain design.The input sequence is shown to converge towards the desired input, resulting in an output that is closest to the given reference in the least square sense. Numerical simulations are provided to validate the theoretical findings.展开更多
Reinforcement learning(RL)algorithms are expected to become the next generation of wind farm control methods.However,as wind farms continue to grow in size,the computational complexity of collective wind farm control ...Reinforcement learning(RL)algorithms are expected to become the next generation of wind farm control methods.However,as wind farms continue to grow in size,the computational complexity of collective wind farm control will exponentially increase with the growth of action and state spaces,limiting its potential in practical applications.In this Letter,we employ a RL-based wind farm control approach with multi-agent deep deterministic policy gradient to optimize the yaw manoeuvre of grouped wind turbines in wind farms.To reduce the computational complexity,the turbines in the wind farm are grouped according to the strength of the wake interaction.Meanwhile,to improve the control efficiency,each subgroup is treated as a whole and controlled by a single agent.Optimized results show that the proposed method can not only increase the power production of the wind farm but also significantly improve the control efficiency.展开更多
This paper examines the difficulties of managing distributed power systems,notably due to the increasing use of renewable energy sources,and focuses on voltage control challenges exacerbated by their variable nature i...This paper examines the difficulties of managing distributed power systems,notably due to the increasing use of renewable energy sources,and focuses on voltage control challenges exacerbated by their variable nature in modern power grids.To tackle the unique challenges of voltage control in distributed renewable energy networks,researchers are increasingly turning towards multi-agent reinforcement learning(MARL).However,MARL raises safety concerns due to the unpredictability in agent actions during their exploration phase.This unpredictability can lead to unsafe control measures.To mitigate these safety concerns in MARL-based voltage control,our study introduces a novel approach:Safety-ConstrainedMulti-Agent Reinforcement Learning(SC-MARL).This approach incorporates a specialized safety constraint module specifically designed for voltage control within the MARL framework.This module ensures that the MARL agents carry out voltage control actions safely.The experiments demonstrate that,in the 33-buses,141-buses,and 322-buses power systems,employing SC-MARL for voltage control resulted in a reduction of the Voltage Out of Control Rate(%V.out)from0.43,0.24,and 2.95 to 0,0.01,and 0.03,respectively.Additionally,the Reactive Power Loss(Q loss)decreased from 0.095,0.547,and 0.017 to 0.062,0.452,and 0.016 in the corresponding systems.展开更多
This paper proposes a modified iterative learning control(MILC)periodical feedback-feedforward algorithm to reduce the vibration of a rotor caused by coupled unbalance and parallel misalignment.The control of the vibr...This paper proposes a modified iterative learning control(MILC)periodical feedback-feedforward algorithm to reduce the vibration of a rotor caused by coupled unbalance and parallel misalignment.The control of the vibration of the rotor is provided by an active magnetic actuator(AMA).The iterative gain of the MILC algorithm here presented has a self-adjustment based on the magnitude of the vibration.Notch filters are adopted to extract the synchronous(1×Ω)and twice rotational frequency(2×Ω)components of the rotor vibration.Both the notch frequency of the filter and the size of feedforward storage used during the experiment have a real-time adaptation to the rotational speed.The method proposed in this work can provide effective suppression of the vibration of the rotor in case of sudden changes or fluctuations of the rotor speed.Simulations and experiments using the MILC algorithm proposed here are carried out and give evidence to the feasibility and robustness of the technique proposed.展开更多
In this paper, platoons of autonomous vehicles operating in urban road networks are considered. From a methodological point of view, the problem of interest consists of formally characterizing vehicle state trajectory...In this paper, platoons of autonomous vehicles operating in urban road networks are considered. From a methodological point of view, the problem of interest consists of formally characterizing vehicle state trajectory tubes by means of routing decisions complying with traffic congestion criteria. To this end, a novel distributed control architecture is conceived by taking advantage of two methodologies: deep reinforcement learning and model predictive control. On one hand, the routing decisions are obtained by using a distributed reinforcement learning algorithm that exploits available traffic data at each road junction. On the other hand, a bank of model predictive controllers is in charge of computing the more adequate control action for each involved vehicle. Such tasks are here combined into a single framework:the deep reinforcement learning output(action) is translated into a set-point to be tracked by the model predictive controller;conversely, the current vehicle position, resulting from the application of the control move, is exploited by the deep reinforcement learning unit for improving its reliability. The main novelty of the proposed solution lies in its hybrid nature: on one hand it fully exploits deep reinforcement learning capabilities for decisionmaking purposes;on the other hand, time-varying hard constraints are always satisfied during the dynamical platoon evolution imposed by the computed routing decisions. To efficiently evaluate the performance of the proposed control architecture, a co-design procedure, involving the SUMO and MATLAB platforms, is implemented so that complex operating environments can be used, and the information coming from road maps(links,junctions, obstacles, semaphores, etc.) and vehicle state trajectories can be shared and exchanged. Finally by considering as operating scenario a real entire city block and a platoon of eleven vehicles described by double-integrator models, several simulations have been performed with the aim to put in light the main f eatures of the proposed approach. Moreover, it is important to underline that in different operating scenarios the proposed reinforcement learning scheme is capable of significantly reducing traffic congestion phenomena when compared with well-reputed competitors.展开更多
In this paper,a new optimal adaptive backstepping control approach for nonlinear systems under deception attacks via reinforcement learning is presented in this paper.The existence of nonlinear terms in the studied sy...In this paper,a new optimal adaptive backstepping control approach for nonlinear systems under deception attacks via reinforcement learning is presented in this paper.The existence of nonlinear terms in the studied system makes it very difficult to design the optimal controller using traditional methods.To achieve optimal control,RL algorithm based on critic–actor architecture is considered for the nonlinear system.Due to the significant security risks of network transmission,the system is vulnerable to deception attacks,which can make all the system state unavailable.By using the attacked states to design coordinate transformation,the harm brought by unknown deception attacks has been overcome.The presented control strategy can ensure that all signals in the closed-loop system are semi-globally ultimately bounded.Finally,the simulation experiment is shown to prove the effectiveness of the strategy.展开更多
The potential for reducing greenhouse gas(GHG)emissions and energy consumption in wastewater treatment can be realized through intelligent control,with machine learning(ML)and multimodality emerging as a promising sol...The potential for reducing greenhouse gas(GHG)emissions and energy consumption in wastewater treatment can be realized through intelligent control,with machine learning(ML)and multimodality emerging as a promising solution.Here,we introduce an ML technique based on multimodal strategies,focusing specifically on intelligent aeration control in wastewater treatment plants(WWTPs).The generalization of the multimodal strategy is demonstrated on eight ML models.The results demonstrate that this multimodal strategy significantly enhances model indicators for ML in environmental science and the efficiency of aeration control,exhibiting exceptional performance and interpretability.Integrating random forest with visual models achieves the highest accuracy in forecasting aeration quantity in multimodal models,with a mean absolute percentage error of 4.4%and a coefficient of determination of 0.948.Practical testing in a full-scale plant reveals that the multimodal model can reduce operation costs by 19.8%compared to traditional fuzzy control methods.The potential application of these strategies in critical water science domains is discussed.To foster accessibility and promote widespread adoption,the multimodal ML models are freely available on GitHub,thereby eliminating technical barriers and encouraging the application of artificial intelligence in urban wastewater treatment.展开更多
Base isolators used in buildings provide both a good acceleration reduction and structural vibration control structures.The base isolators may lose their damping capacity over time due to environmental or dynamic effe...Base isolators used in buildings provide both a good acceleration reduction and structural vibration control structures.The base isolators may lose their damping capacity over time due to environmental or dynamic effects.This deterioration of them requires the determination of the maintenance and repair needs and is important for the long-termisolator life.In this study,an artificial intelligence prediction model has been developed to determine the damage and maintenance-repair requirements of isolators as a result of environmental effects and dynamic factors over time.With the developed model,the required damping capacity of the isolator structure was estimated and compared with the previously placed isolator capacity,and the decrease in the damping property was tried to be determined.For this purpose,a data set was created by collecting the behavior of structures with single degrees of freedom(SDOF),different stiffness,damping ratio and natural period isolated from the foundation under far fault earthquakes.The data is divided into 5 different damping classes varying between 10%and 50%.Machine learning model was trained in damping classes with the data on the structure’s response to random seismic vibrations.As a result of the isolator behavior under randomly selected earthquakes,the recorded motion and structural acceleration of the structure against any seismic vibration were examined,and the decrease in the damping capacity was estimated on a class basis.The performance loss of the isolators,which are separated according to their damping properties,has been tried to be determined,and the reductions in the amounts to be taken into account have been determined by class.In the developed prediction model,using various supervised machine learning classification algorithms,the classification algorithm providing the highest precision for the model has been decided.When the results are examined,it has been determined that the damping of the isolator structure with the machine learning method is predicted successfully at a level exceeding 96%,and it is an effective method in deciding whether there is a decrease in the damping capacity.展开更多
This survey paper provides a review and perspective on intermediate and advanced reinforcement learning(RL)techniques in process industries. It offers a holistic approach by covering all levels of the process control ...This survey paper provides a review and perspective on intermediate and advanced reinforcement learning(RL)techniques in process industries. It offers a holistic approach by covering all levels of the process control hierarchy. The survey paper presents a comprehensive overview of RL algorithms,including fundamental concepts like Markov decision processes and different approaches to RL, such as value-based, policy-based, and actor-critic methods, while also discussing the relationship between classical control and RL. It further reviews the wide-ranging applications of RL in process industries, such as soft sensors, low-level control, high-level control, distributed process control, fault detection and fault tolerant control, optimization,planning, scheduling, and supply chain. The survey paper discusses the limitations and advantages, trends and new applications, and opportunities and future prospects for RL in process industries. Moreover, it highlights the need for a holistic approach in complex systems due to the growing importance of digitalization in the process industries.展开更多
Unmanned Aerial Vehicles(UAvs)as aerial base stations to provide communication services for ground users is a flexible and cost-effective paradigm in B5G.Besides,dynamic resource allocation and multi-connectivity can ...Unmanned Aerial Vehicles(UAvs)as aerial base stations to provide communication services for ground users is a flexible and cost-effective paradigm in B5G.Besides,dynamic resource allocation and multi-connectivity can be adopted to further harness the potentials of UAVs in improving communication capacity,in such situations such that the interference among users becomes a pivotal disincentive requiring effective solutions.To this end,we investigate the Joint UAV-User Association,Channel Allocation,and transmission Power Control(J-UACAPC)problem in a multi-connectivity-enabled UAV network with constrained backhaul links,where each UAV can determine the reusable channels and transmission power to serve the selected ground users.The goal was to mitigate co-channel interference while maximizing long-term system utility.The problem was modeled as a cooperative stochastic game with hybrid discrete-continuous action space.A Multi-Agent Hybrid Deep Reinforcement Learning(MAHDRL)algorithm was proposed to address this problem.Extensive simulation results demonstrated the effectiveness of the proposed algorithm and showed that it has a higher system utility than the baseline methods.展开更多
This article studies the effective traffic signal control problem of multiple intersections in a city-level traffic system.A novel regional multi-agent cooperative reinforcement learning algorithm called RegionSTLight...This article studies the effective traffic signal control problem of multiple intersections in a city-level traffic system.A novel regional multi-agent cooperative reinforcement learning algorithm called RegionSTLight is proposed to improve the traffic efficiency.Firstly a regional multi-agent Q-learning framework is proposed,which can equivalently decompose the global Q value of the traffic system into the local values of several regions Based on the framework and the idea of human-machine cooperation,a dynamic zoning method is designed to divide the traffic network into several strong-coupled regions according to realtime traffic flow densities.In order to achieve better cooperation inside each region,a lightweight spatio-temporal fusion feature extraction network is designed.The experiments in synthetic real-world and city-level scenarios show that the proposed RegionS TLight converges more quickly,is more stable,and obtains better asymptotic performance compared to state-of-theart models.展开更多
This paper studies a novel distributed optimization problem that aims to minimize the sum of the non-convex objective functionals of the multi-agent network under privacy protection, which means that the local objecti...This paper studies a novel distributed optimization problem that aims to minimize the sum of the non-convex objective functionals of the multi-agent network under privacy protection, which means that the local objective of each agent is unknown to others. The above problem involves complexity simultaneously in the time and space aspects. Yet existing works about distributed optimization mainly consider privacy protection in the space aspect where the decision variable is a vector with finite dimensions. In contrast, when the time aspect is considered in this paper, the decision variable is a continuous function concerning time. Hence, the minimization of the overall functional belongs to the calculus of variations. Traditional works usually aim to seek the optimal decision function. Due to privacy protection and non-convexity, the Euler-Lagrange equation of the proposed problem is a complicated partial differential equation.Hence, we seek the optimal decision derivative function rather than the decision function. This manner can be regarded as seeking the control input for an optimal control problem, for which we propose a centralized reinforcement learning(RL) framework. In the space aspect, we further present a distributed reinforcement learning framework to deal with the impact of privacy protection. Finally, rigorous theoretical analysis and simulation validate the effectiveness of our framework.展开更多
Static Poisson’s ratio(vs)is crucial for determining geomechanical properties in petroleum applications,namely sand production.Some models have been used to predict vs;however,the published models were limited to spe...Static Poisson’s ratio(vs)is crucial for determining geomechanical properties in petroleum applications,namely sand production.Some models have been used to predict vs;however,the published models were limited to specific data ranges with an average absolute percentage relative error(AAPRE)of more than 10%.The published gated recurrent unit(GRU)models do not consider trend analysis to show physical behaviors.In this study,we aim to develop a GRU model using trend analysis and three inputs for predicting n s based on a broad range of data,n s(value of 0.1627-0.4492),bulk formation density(RHOB)(0.315-2.994 g/mL),compressional time(DTc)(44.43-186.9 μs/ft),and shear time(DTs)(72.9-341.2μ s/ft).The GRU model was evaluated using different approaches,including statistical error an-alyses.The GRU model showed the proper trends,and the model data ranges were wider than previous ones.The GRU model has the largest correlation coefficient(R)of 0.967 and the lowest AAPRE,average percent relative error(APRE),root mean square error(RMSE),and standard deviation(SD)of 3.228%,1.054%,4.389,and 0.013,respectively,compared to other models.The GRU model has a high accuracy for the different datasets:training,validation,testing,and the whole datasets with R and AAPRE values were 0.981 and 2.601%,0.966 and 3.274%,0.967 and 3.228%,and 0.977 and 2.861%,respectively.The group error analyses of all inputs show that the GRU model has less than 5% AAPRE for all input ranges,which is superior to other models that have different AAPRE values of more than 10% at various ranges of inputs.展开更多
Wind-farm flow control stands at the forefront of grand challenges in wind-energy science.The central issue is that current algorithms are based on simplified models and,thus,fall short of capturing the complex physic...Wind-farm flow control stands at the forefront of grand challenges in wind-energy science.The central issue is that current algorithms are based on simplified models and,thus,fall short of capturing the complex physics of wind farms associated with the high-dimensional nature of turbulence and multiscale wind-farm-atmosphere interactions.Reinforcement learning(RL),as a subset of machine learning,has demonstrated its effectiveness in solving high-dimensional problems in various domains,and the studies performed in the last decade prove that it can be exploited in the development of the next generation of algorithms for wind-farm flow control.This review has two main objectives.Firstly,it aims to provide an up-to-date overview of works focusing on the development of wind-farm flow control schemes utilizing RL methods.By examining the latest research in this area,the review seeks to offer a comprehensive understanding of the advancements made in wind-farm flow control through the application of RL techniques.Secondly,it aims to shed light on the obstacles that researchers face when implementing wind-farm flow control based on RL.By highlighting these challenges,the review aims to identify areas requiring further exploration and potential opportunities for future research.展开更多
This paper presents a machine-learning-based speedup strategy for real-time implementation of model-predictive-control(MPC)in emergency voltage stabilization of power systems.Despite success in various applications,re...This paper presents a machine-learning-based speedup strategy for real-time implementation of model-predictive-control(MPC)in emergency voltage stabilization of power systems.Despite success in various applications,real-time implementation of MPC in power systems has not been successful due to the online control computation time required for large-sized complex systems,and in power systems,the computation time exceeds the available decision time used in practice by a large extent.This long-standing problem is addressed here by developing a novel MPC-based framework that i)computes an optimal strategy for nominal loads in an offline setting and adapts it for real-time scenarios by successive online control corrections at each control instant utilizing the latest measurements,and ii)employs a machine-learning based approach for the prediction of voltage trajectory and its sensitivity to control inputs,thereby accelerating the overall control computation by multiple times.Additionally,a realistic control coordination scheme among static var compensators(SVC),load-shedding(LS),and load tap-changers(LTC)is presented that incorporates the practical delayed actions of the LTCs.The performance of the proposed scheme is validated for IEEE 9-bus and 39-bus systems,with±20%variations in nominal loading conditions together with contingencies.We show that our proposed methodology speeds up the online computation by 20-fold,bringing it down to a practically feasible value(fraction of a second),making the MPC real-time and feasible for power system control for the first time.展开更多
Reinforcement Learning(RL)based control algorithms can learn the control strategies for nonlinear and uncertain environment during interacting with it.Guided by the rewards generated by environment,a RL agent can lear...Reinforcement Learning(RL)based control algorithms can learn the control strategies for nonlinear and uncertain environment during interacting with it.Guided by the rewards generated by environment,a RL agent can learn the control strategy directly in a model-free way instead of investigating the dynamic model of the environment.In the paper,we propose the sampled-data RL control strategy to reduce the computational demand.In the sampled-data control strategy,the whole control system is of a hybrid structure,in which the plant is of continuous structure while the controller(RL agent)adopts a discrete structure.Given that the continuous states of the plant will be the input of the agent,the state–action value function is approximated by the fully connected feed-forward neural networks(FCFFNN).Instead of learning the controller at every step during the interaction with the environment,the learning and acting stages are decoupled to learn the control strategy more effectively through experience replay.In the acting stage,the most effective experience obtained during the interaction with the environment will be stored and during the learning stage,the stored experience will be replayed to customized times,which helps enhance the experience replay process.The effectiveness of proposed approach will be verified by simulation examples.展开更多
This paper is concerned with a novel integrated multi-step heuristic dynamic programming(MsHDP)algorithm for solving optimal control problems.It is shown that,initialized by the zero cost function,MsHDP can converge t...This paper is concerned with a novel integrated multi-step heuristic dynamic programming(MsHDP)algorithm for solving optimal control problems.It is shown that,initialized by the zero cost function,MsHDP can converge to the optimal solution of the Hamilton-Jacobi-Bellman(HJB)equation.Then,the stability of the system is analyzed using control policies generated by MsHDP.Also,a general stability criterion is designed to determine the admissibility of the current control policy.That is,the criterion is applicable not only to traditional value iteration and policy iteration but also to MsHDP.Further,based on the convergence and the stability criterion,the integrated MsHDP algorithm using immature control policies is developed to accelerate learning efficiency greatly.Besides,actor-critic is utilized to implement the integrated MsHDP scheme,where neural networks are used to evaluate and improve the iterative policy as the parameter architecture.Finally,two simulation examples are given to demonstrate that the learning effectiveness of the integrated MsHDP scheme surpasses those of other fixed or integrated methods.展开更多
基金supported in part by the National Natural Science Foundation of China(62222301, 62073085, 62073158, 61890930-5, 62021003)the National Key Research and Development Program of China (2021ZD0112302, 2021ZD0112301, 2018YFC1900800-5)Beijing Natural Science Foundation (JQ19013)。
文摘Reinforcement learning(RL) has roots in dynamic programming and it is called adaptive/approximate dynamic programming(ADP) within the control community. This paper reviews recent developments in ADP along with RL and its applications to various advanced control fields. First, the background of the development of ADP is described, emphasizing the significance of regulation and tracking control problems. Some effective offline and online algorithms for ADP/adaptive critic control are displayed, where the main results towards discrete-time systems and continuous-time systems are surveyed, respectively.Then, the research progress on adaptive critic control based on the event-triggered framework and under uncertain environment is discussed, respectively, where event-based design, robust stabilization, and game design are reviewed. Moreover, the extensions of ADP for addressing control problems under complex environment attract enormous attention. The ADP architecture is revisited under the perspective of data-driven and RL frameworks,showing how they promote ADP formulation significantly.Finally, several typical control applications with respect to RL and ADP are summarized, particularly in the fields of wastewater treatment processes and power systems, followed by some general prospects for future research. Overall, the comprehensive survey on ADP and RL for advanced control applications has d emonstrated its remarkable potential within the artificial intelligence era. In addition, it also plays a vital role in promoting environmental protection and industrial intelligence.
基金supported by the National Natural Science Foundation of China(U21A20166)in part by the Science and Technology Development Foundation of Jilin Province (20230508095RC)+1 种基金in part by the Development and Reform Commission Foundation of Jilin Province (2023C034-3)in part by the Exploration Foundation of State Key Laboratory of Automotive Simulation and Control。
文摘Aiming at the tracking problem of a class of discrete nonaffine nonlinear multi-input multi-output(MIMO) repetitive systems subjected to separable and nonseparable disturbances, a novel data-driven iterative learning control(ILC) scheme based on the zeroing neural networks(ZNNs) is proposed. First, the equivalent dynamic linearization data model is obtained by means of dynamic linearization technology, which exists theoretically in the iteration domain. Then, the iterative extended state observer(IESO) is developed to estimate the disturbance and the coupling between systems, and the decoupled dynamic linearization model is obtained for the purpose of controller synthesis. To solve the zero-seeking tracking problem with inherent tolerance of noise,an ILC based on noise-tolerant modified ZNN is proposed. The strict assumptions imposed on the initialization conditions of each iteration in the existing ILC methods can be absolutely removed with our method. In addition, theoretical analysis indicates that the modified ZNN can converge to the exact solution of the zero-seeking tracking problem. Finally, a generalized example and an application-oriented example are presented to verify the effectiveness and superiority of the proposed process.
基金supported in part by the National Key R&D Program of China under Grant 2021YFB2011300the National Natural Science Foundation of China under Grant 52075262。
文摘This paper mainly focuses on the development of a learning-based controller for a class of uncertain mechanical systems modeled by the Euler-Lagrange formulation.The considered system can depict the behavior of a large class of engineering systems,such as vehicular systems,robot manipulators and satellites.All these systems are often characterized by highly nonlinear characteristics,heavy modeling uncertainties and unknown perturbations,therefore,accurate-model-based nonlinear control approaches become unavailable.Motivated by the challenge,a reinforcement learning(RL)adaptive control methodology based on the actor-critic framework is investigated to compensate the uncertain mechanical dynamics.The approximation inaccuracies caused by RL and the exogenous unknown disturbances are circumvented via a continuous robust integral of the sign of the error(RISE)control approach.Different from a classical RISE control law,a tanh(·)function is utilized instead of a sign(·)function to acquire a more smooth control signal.The developed controller requires very little prior knowledge of the dynamic model,is robust to unknown dynamics and exogenous disturbances,and can achieve asymptotic output tracking.Eventually,co-simulations through ADAMS and MATLAB/Simulink on a three degrees-of-freedom(3-DOF)manipulator and experiments on a real-time electromechanical servo system are performed to verify the performance of the proposed approach.
基金supported by the National Natural Science Foundation of China (62173333, 12271522)Beijing Natural Science Foundation (Z210002)the Research Fund of Renmin University of China (2021030187)。
文摘For unachievable tracking problems, where the system output cannot precisely track a given reference, achieving the best possible approximation for the reference trajectory becomes the objective. This study aims to investigate solutions using the Ptype learning control scheme. Initially, we demonstrate the necessity of gradient information for achieving the best approximation.Subsequently, we propose an input-output-driven learning gain design to handle the imprecise gradients of a class of uncertain systems. However, it is discovered that the desired performance may not be attainable when faced with incomplete information.To address this issue, an extended iterative learning control scheme is introduced. In this scheme, the tracking errors are modified through output data sampling, which incorporates lowmemory footprints and offers flexibility in learning gain design.The input sequence is shown to converge towards the desired input, resulting in an output that is closest to the given reference in the least square sense. Numerical simulations are provided to validate the theoretical findings.
基金supported by the National Natural Science Foundation of China (Grant No.12388101)the Science Challenge Project+1 种基金the Anhui NARI Jiyuan Electric Power Grid Technology Co.Ltd.through the Joint Laboratory of USTC-NARIthe advanced computing resources provided by the Supercomputing Center of the USTC
文摘Reinforcement learning(RL)algorithms are expected to become the next generation of wind farm control methods.However,as wind farms continue to grow in size,the computational complexity of collective wind farm control will exponentially increase with the growth of action and state spaces,limiting its potential in practical applications.In this Letter,we employ a RL-based wind farm control approach with multi-agent deep deterministic policy gradient to optimize the yaw manoeuvre of grouped wind turbines in wind farms.To reduce the computational complexity,the turbines in the wind farm are grouped according to the strength of the wake interaction.Meanwhile,to improve the control efficiency,each subgroup is treated as a whole and controlled by a single agent.Optimized results show that the proposed method can not only increase the power production of the wind farm but also significantly improve the control efficiency.
基金“Regional Innovation Strategy(RIS)”through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(MOE)(2021RIS-002).
文摘This paper examines the difficulties of managing distributed power systems,notably due to the increasing use of renewable energy sources,and focuses on voltage control challenges exacerbated by their variable nature in modern power grids.To tackle the unique challenges of voltage control in distributed renewable energy networks,researchers are increasingly turning towards multi-agent reinforcement learning(MARL).However,MARL raises safety concerns due to the unpredictability in agent actions during their exploration phase.This unpredictability can lead to unsafe control measures.To mitigate these safety concerns in MARL-based voltage control,our study introduces a novel approach:Safety-ConstrainedMulti-Agent Reinforcement Learning(SC-MARL).This approach incorporates a specialized safety constraint module specifically designed for voltage control within the MARL framework.This module ensures that the MARL agents carry out voltage control actions safely.The experiments demonstrate that,in the 33-buses,141-buses,and 322-buses power systems,employing SC-MARL for voltage control resulted in a reduction of the Voltage Out of Control Rate(%V.out)from0.43,0.24,and 2.95 to 0,0.01,and 0.03,respectively.Additionally,the Reactive Power Loss(Q loss)decreased from 0.095,0.547,and 0.017 to 0.062,0.452,and 0.016 in the corresponding systems.
基金Supported by National Natural Science Foundation of China(Grant Nos.51975037,52375075).
文摘This paper proposes a modified iterative learning control(MILC)periodical feedback-feedforward algorithm to reduce the vibration of a rotor caused by coupled unbalance and parallel misalignment.The control of the vibration of the rotor is provided by an active magnetic actuator(AMA).The iterative gain of the MILC algorithm here presented has a self-adjustment based on the magnitude of the vibration.Notch filters are adopted to extract the synchronous(1×Ω)and twice rotational frequency(2×Ω)components of the rotor vibration.Both the notch frequency of the filter and the size of feedforward storage used during the experiment have a real-time adaptation to the rotational speed.The method proposed in this work can provide effective suppression of the vibration of the rotor in case of sudden changes or fluctuations of the rotor speed.Simulations and experiments using the MILC algorithm proposed here are carried out and give evidence to the feasibility and robustness of the technique proposed.
文摘In this paper, platoons of autonomous vehicles operating in urban road networks are considered. From a methodological point of view, the problem of interest consists of formally characterizing vehicle state trajectory tubes by means of routing decisions complying with traffic congestion criteria. To this end, a novel distributed control architecture is conceived by taking advantage of two methodologies: deep reinforcement learning and model predictive control. On one hand, the routing decisions are obtained by using a distributed reinforcement learning algorithm that exploits available traffic data at each road junction. On the other hand, a bank of model predictive controllers is in charge of computing the more adequate control action for each involved vehicle. Such tasks are here combined into a single framework:the deep reinforcement learning output(action) is translated into a set-point to be tracked by the model predictive controller;conversely, the current vehicle position, resulting from the application of the control move, is exploited by the deep reinforcement learning unit for improving its reliability. The main novelty of the proposed solution lies in its hybrid nature: on one hand it fully exploits deep reinforcement learning capabilities for decisionmaking purposes;on the other hand, time-varying hard constraints are always satisfied during the dynamical platoon evolution imposed by the computed routing decisions. To efficiently evaluate the performance of the proposed control architecture, a co-design procedure, involving the SUMO and MATLAB platforms, is implemented so that complex operating environments can be used, and the information coming from road maps(links,junctions, obstacles, semaphores, etc.) and vehicle state trajectories can be shared and exchanged. Finally by considering as operating scenario a real entire city block and a platoon of eleven vehicles described by double-integrator models, several simulations have been performed with the aim to put in light the main f eatures of the proposed approach. Moreover, it is important to underline that in different operating scenarios the proposed reinforcement learning scheme is capable of significantly reducing traffic congestion phenomena when compared with well-reputed competitors.
基金supported in part by the National Key R&D Program of China under Grants 2021YFE0206100in part by the National Natural Science Foundation of China under Grant 62073321+2 种基金in part by National Defense Basic Scientific Research Program JCKY2019203C029in part by the Science and Technology Development Fund,Macao SAR under Grants FDCT-22-009-MISE,0060/2021/A2 and 0015/2020/AMJin part by the financial support from the National Defense Basic Scientific Research Project(JCKY2020130C025).
文摘In this paper,a new optimal adaptive backstepping control approach for nonlinear systems under deception attacks via reinforcement learning is presented in this paper.The existence of nonlinear terms in the studied system makes it very difficult to design the optimal controller using traditional methods.To achieve optimal control,RL algorithm based on critic–actor architecture is considered for the nonlinear system.Due to the significant security risks of network transmission,the system is vulnerable to deception attacks,which can make all the system state unavailable.By using the attacked states to design coordinate transformation,the harm brought by unknown deception attacks has been overcome.The presented control strategy can ensure that all signals in the closed-loop system are semi-globally ultimately bounded.Finally,the simulation experiment is shown to prove the effectiveness of the strategy.
基金the financial support by the National Natural Science Foundation of China(52230004 and 52293445)the Key Research and Development Project of Shandong Province(2020CXGC011202-005)the Shenzhen Science and Technology Program(KCXFZ20211020163404007 and KQTD20190929172630447).
文摘The potential for reducing greenhouse gas(GHG)emissions and energy consumption in wastewater treatment can be realized through intelligent control,with machine learning(ML)and multimodality emerging as a promising solution.Here,we introduce an ML technique based on multimodal strategies,focusing specifically on intelligent aeration control in wastewater treatment plants(WWTPs).The generalization of the multimodal strategy is demonstrated on eight ML models.The results demonstrate that this multimodal strategy significantly enhances model indicators for ML in environmental science and the efficiency of aeration control,exhibiting exceptional performance and interpretability.Integrating random forest with visual models achieves the highest accuracy in forecasting aeration quantity in multimodal models,with a mean absolute percentage error of 4.4%and a coefficient of determination of 0.948.Practical testing in a full-scale plant reveals that the multimodal model can reduce operation costs by 19.8%compared to traditional fuzzy control methods.The potential application of these strategies in critical water science domains is discussed.To foster accessibility and promote widespread adoption,the multimodal ML models are freely available on GitHub,thereby eliminating technical barriers and encouraging the application of artificial intelligence in urban wastewater treatment.
基金the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(2020R1A2C1A01011131)the Energy Cloud R&D Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Science,ICT(2019M3F2A1073164).
文摘Base isolators used in buildings provide both a good acceleration reduction and structural vibration control structures.The base isolators may lose their damping capacity over time due to environmental or dynamic effects.This deterioration of them requires the determination of the maintenance and repair needs and is important for the long-termisolator life.In this study,an artificial intelligence prediction model has been developed to determine the damage and maintenance-repair requirements of isolators as a result of environmental effects and dynamic factors over time.With the developed model,the required damping capacity of the isolator structure was estimated and compared with the previously placed isolator capacity,and the decrease in the damping property was tried to be determined.For this purpose,a data set was created by collecting the behavior of structures with single degrees of freedom(SDOF),different stiffness,damping ratio and natural period isolated from the foundation under far fault earthquakes.The data is divided into 5 different damping classes varying between 10%and 50%.Machine learning model was trained in damping classes with the data on the structure’s response to random seismic vibrations.As a result of the isolator behavior under randomly selected earthquakes,the recorded motion and structural acceleration of the structure against any seismic vibration were examined,and the decrease in the damping capacity was estimated on a class basis.The performance loss of the isolators,which are separated according to their damping properties,has been tried to be determined,and the reductions in the amounts to be taken into account have been determined by class.In the developed prediction model,using various supervised machine learning classification algorithms,the classification algorithm providing the highest precision for the model has been decided.When the results are examined,it has been determined that the damping of the isolator structure with the machine learning method is predicted successfully at a level exceeding 96%,and it is an effective method in deciding whether there is a decrease in the damping capacity.
基金supported in part by the Natural Sciences Engineering Research Council of Canada (NSERC)。
文摘This survey paper provides a review and perspective on intermediate and advanced reinforcement learning(RL)techniques in process industries. It offers a holistic approach by covering all levels of the process control hierarchy. The survey paper presents a comprehensive overview of RL algorithms,including fundamental concepts like Markov decision processes and different approaches to RL, such as value-based, policy-based, and actor-critic methods, while also discussing the relationship between classical control and RL. It further reviews the wide-ranging applications of RL in process industries, such as soft sensors, low-level control, high-level control, distributed process control, fault detection and fault tolerant control, optimization,planning, scheduling, and supply chain. The survey paper discusses the limitations and advantages, trends and new applications, and opportunities and future prospects for RL in process industries. Moreover, it highlights the need for a holistic approach in complex systems due to the growing importance of digitalization in the process industries.
基金supported in part by the National Natural Science Foundation of China(grant nos.61971365,61871339,62171392)Digital Fujian Province Key Laboratory of IoT Communication,Architecture and Safety Technology(grant no.2010499)+1 种基金the State Key Program of the National Natural Science Foundation of China(grant no.61731012)the Natural Science Foundation of Fujian Province of China No.2021J01004.
文摘Unmanned Aerial Vehicles(UAvs)as aerial base stations to provide communication services for ground users is a flexible and cost-effective paradigm in B5G.Besides,dynamic resource allocation and multi-connectivity can be adopted to further harness the potentials of UAVs in improving communication capacity,in such situations such that the interference among users becomes a pivotal disincentive requiring effective solutions.To this end,we investigate the Joint UAV-User Association,Channel Allocation,and transmission Power Control(J-UACAPC)problem in a multi-connectivity-enabled UAV network with constrained backhaul links,where each UAV can determine the reusable channels and transmission power to serve the selected ground users.The goal was to mitigate co-channel interference while maximizing long-term system utility.The problem was modeled as a cooperative stochastic game with hybrid discrete-continuous action space.A Multi-Agent Hybrid Deep Reinforcement Learning(MAHDRL)algorithm was proposed to address this problem.Extensive simulation results demonstrated the effectiveness of the proposed algorithm and showed that it has a higher system utility than the baseline methods.
基金supported by the National Science and Technology Major Project (2021ZD0112702)the National Natural Science Foundation (NNSF)of China (62373100,62233003)the Natural Science Foundation of Jiangsu Province of China (BK20202006)。
文摘This article studies the effective traffic signal control problem of multiple intersections in a city-level traffic system.A novel regional multi-agent cooperative reinforcement learning algorithm called RegionSTLight is proposed to improve the traffic efficiency.Firstly a regional multi-agent Q-learning framework is proposed,which can equivalently decompose the global Q value of the traffic system into the local values of several regions Based on the framework and the idea of human-machine cooperation,a dynamic zoning method is designed to divide the traffic network into several strong-coupled regions according to realtime traffic flow densities.In order to achieve better cooperation inside each region,a lightweight spatio-temporal fusion feature extraction network is designed.The experiments in synthetic real-world and city-level scenarios show that the proposed RegionS TLight converges more quickly,is more stable,and obtains better asymptotic performance compared to state-of-theart models.
基金supported in part by the National Natural Science Foundation of China(NSFC)(61773260)the Ministry of Science and Technology (2018YFB130590)。
文摘This paper studies a novel distributed optimization problem that aims to minimize the sum of the non-convex objective functionals of the multi-agent network under privacy protection, which means that the local objective of each agent is unknown to others. The above problem involves complexity simultaneously in the time and space aspects. Yet existing works about distributed optimization mainly consider privacy protection in the space aspect where the decision variable is a vector with finite dimensions. In contrast, when the time aspect is considered in this paper, the decision variable is a continuous function concerning time. Hence, the minimization of the overall functional belongs to the calculus of variations. Traditional works usually aim to seek the optimal decision function. Due to privacy protection and non-convexity, the Euler-Lagrange equation of the proposed problem is a complicated partial differential equation.Hence, we seek the optimal decision derivative function rather than the decision function. This manner can be regarded as seeking the control input for an optimal control problem, for which we propose a centralized reinforcement learning(RL) framework. In the space aspect, we further present a distributed reinforcement learning framework to deal with the impact of privacy protection. Finally, rigorous theoretical analysis and simulation validate the effectiveness of our framework.
基金The authors thank the Yayasan Universiti Teknologi PETRONAS(YUTP FRG Grant No.015LC0-428)at Universiti Teknologi PETRO-NAS for supporting this study.
文摘Static Poisson’s ratio(vs)is crucial for determining geomechanical properties in petroleum applications,namely sand production.Some models have been used to predict vs;however,the published models were limited to specific data ranges with an average absolute percentage relative error(AAPRE)of more than 10%.The published gated recurrent unit(GRU)models do not consider trend analysis to show physical behaviors.In this study,we aim to develop a GRU model using trend analysis and three inputs for predicting n s based on a broad range of data,n s(value of 0.1627-0.4492),bulk formation density(RHOB)(0.315-2.994 g/mL),compressional time(DTc)(44.43-186.9 μs/ft),and shear time(DTs)(72.9-341.2μ s/ft).The GRU model was evaluated using different approaches,including statistical error an-alyses.The GRU model showed the proper trends,and the model data ranges were wider than previous ones.The GRU model has the largest correlation coefficient(R)of 0.967 and the lowest AAPRE,average percent relative error(APRE),root mean square error(RMSE),and standard deviation(SD)of 3.228%,1.054%,4.389,and 0.013,respectively,compared to other models.The GRU model has a high accuracy for the different datasets:training,validation,testing,and the whole datasets with R and AAPRE values were 0.981 and 2.601%,0.966 and 3.274%,0.967 and 3.228%,and 0.977 and 2.861%,respectively.The group error analyses of all inputs show that the GRU model has less than 5% AAPRE for all input ranges,which is superior to other models that have different AAPRE values of more than 10% at various ranges of inputs.
基金the financial support from the Independent Research Fund Denmark(DFF)under Grant No.0217-00038B。
文摘Wind-farm flow control stands at the forefront of grand challenges in wind-energy science.The central issue is that current algorithms are based on simplified models and,thus,fall short of capturing the complex physics of wind farms associated with the high-dimensional nature of turbulence and multiscale wind-farm-atmosphere interactions.Reinforcement learning(RL),as a subset of machine learning,has demonstrated its effectiveness in solving high-dimensional problems in various domains,and the studies performed in the last decade prove that it can be exploited in the development of the next generation of algorithms for wind-farm flow control.This review has two main objectives.Firstly,it aims to provide an up-to-date overview of works focusing on the development of wind-farm flow control schemes utilizing RL methods.By examining the latest research in this area,the review seeks to offer a comprehensive understanding of the advancements made in wind-farm flow control through the application of RL techniques.Secondly,it aims to shed light on the obstacles that researchers face when implementing wind-farm flow control based on RL.By highlighting these challenges,the review aims to identify areas requiring further exploration and potential opportunities for future research.
基金This work was supported in part by the National Science Foundation(NSF-CSSI-2004766,NSF-PFI-2141084).
文摘This paper presents a machine-learning-based speedup strategy for real-time implementation of model-predictive-control(MPC)in emergency voltage stabilization of power systems.Despite success in various applications,real-time implementation of MPC in power systems has not been successful due to the online control computation time required for large-sized complex systems,and in power systems,the computation time exceeds the available decision time used in practice by a large extent.This long-standing problem is addressed here by developing a novel MPC-based framework that i)computes an optimal strategy for nominal loads in an offline setting and adapts it for real-time scenarios by successive online control corrections at each control instant utilizing the latest measurements,and ii)employs a machine-learning based approach for the prediction of voltage trajectory and its sensitivity to control inputs,thereby accelerating the overall control computation by multiple times.Additionally,a realistic control coordination scheme among static var compensators(SVC),load-shedding(LS),and load tap-changers(LTC)is presented that incorporates the practical delayed actions of the LTCs.The performance of the proposed scheme is validated for IEEE 9-bus and 39-bus systems,with±20%variations in nominal loading conditions together with contingencies.We show that our proposed methodology speeds up the online computation by 20-fold,bringing it down to a practically feasible value(fraction of a second),making the MPC real-time and feasible for power system control for the first time.
基金supported by Imperial College London,UK,King’s College London,UK and Engineering and Physical Sciences Research Council(EPSRC),UK.
文摘Reinforcement Learning(RL)based control algorithms can learn the control strategies for nonlinear and uncertain environment during interacting with it.Guided by the rewards generated by environment,a RL agent can learn the control strategy directly in a model-free way instead of investigating the dynamic model of the environment.In the paper,we propose the sampled-data RL control strategy to reduce the computational demand.In the sampled-data control strategy,the whole control system is of a hybrid structure,in which the plant is of continuous structure while the controller(RL agent)adopts a discrete structure.Given that the continuous states of the plant will be the input of the agent,the state–action value function is approximated by the fully connected feed-forward neural networks(FCFFNN).Instead of learning the controller at every step during the interaction with the environment,the learning and acting stages are decoupled to learn the control strategy more effectively through experience replay.In the acting stage,the most effective experience obtained during the interaction with the environment will be stored and during the learning stage,the stored experience will be replayed to customized times,which helps enhance the experience replay process.The effectiveness of proposed approach will be verified by simulation examples.
基金the National Key Research and Development Program of China(2021ZD0112302)the National Natural Science Foundation of China(62222301,61890930-5,62021003)the Beijing Natural Science Foundation(JQ19013).
文摘This paper is concerned with a novel integrated multi-step heuristic dynamic programming(MsHDP)algorithm for solving optimal control problems.It is shown that,initialized by the zero cost function,MsHDP can converge to the optimal solution of the Hamilton-Jacobi-Bellman(HJB)equation.Then,the stability of the system is analyzed using control policies generated by MsHDP.Also,a general stability criterion is designed to determine the admissibility of the current control policy.That is,the criterion is applicable not only to traditional value iteration and policy iteration but also to MsHDP.Further,based on the convergence and the stability criterion,the integrated MsHDP algorithm using immature control policies is developed to accelerate learning efficiency greatly.Besides,actor-critic is utilized to implement the integrated MsHDP scheme,where neural networks are used to evaluate and improve the iterative policy as the parameter architecture.Finally,two simulation examples are given to demonstrate that the learning effectiveness of the integrated MsHDP scheme surpasses those of other fixed or integrated methods.