Abstract:In order to realize effective monitoring of naval battlefield situation,this paper constructs a set of network deployment model based on energy balancing strategy,aiming at the problems of poor connectivity,short life and low perception of traditional perception network deployment model,and adds network connectivity to the constraint term,and proposes an energy consumption balancing strategy for network nodes to improve the life cycle and total perception of the network.An improved artificial hummingbird algorithm(IAHA)is proposed to solve the model.Differential variation strategy and reverse mapping strategy are introduced to improve the global optimization ability of the algorithm,and adaptive foraging strategy is introduced to improve the convergence speed of the algorithm.The experimental results show that the network generated by the proposed model has strong connectivity,and compared with other deployment strategies,the network generated based on energy balancing strategy has a longer lifetime and a higher perception total.Compared with the comparison algorithm,IAHA has stronger global optimization ability and faster convergence speed.

Abstract:Microgrids can comprehensively utilize various energy sources and coordinate energy storage systems to compensate for the uncertainties in the output of distributed generators(DGs).a distributed pinning collaborative control strategy based on deep reinforcement learning for isolated microgrids is proposed to solve the problem of voltage and frequency deviation caused by droop control in isolated microgrids,which can effectively suppress frequency and voltage fluctuations under load disturbance or system topology changes.Firstly,aiming at the problem that the pinning target value in the pinning consistency algorithm is difficult to accurately preset and change,the double deep Q learning(DDQN)is used to adaptively modify the pinning target value by leveraging the adaptability and generalization capability of DDQN.It improves the adaptability and effectiveness of the pinning consistency algorithm.Furthermore,the structure of the distributed pinning collaborative controller based on DDQN is designed,and the definition of state space,action space,and reward function is completed.Finally,Simulation results demonstrate that the proposed control method outperforms traditional pinning control and reinforcement learning-based pinning control in responding to changes in network topology parameters and structure.

Abstract:The density peaks clustering algorithm can deal with datasets quickly and efficiently without iteration.How-ever,it can sometimes wrongly select cluster centers and misallocate samples when processing manifold data.Therefore,this paper proposes the density peaks clustering algorithm based on shared nearest neighbor and second-order K nearest neighbor for manifold data(DPC-SKNN)algorithm.Firstly,the algorithm introduces reverse nearest neighbors and shares nearest neighbors to redefine local density,fully considering both local and global information of samples,making the al-gorithm easier to identify correct cluster centers.Secondly,the association relationship of the samples is divided into three types:K nearest neighbors,second-order K nearest neighbors,and non-nearest neighbors,and design allocation strategies for K-nearest neighbors to enhance similarity among samples within the same cluster,thereby improving sample allocation accuracy.DPC-SKNN is compared with eight algorithms on manifold and UCI datasets,and the experimental results show that the DPC-SKNN algorithm obtains good clustering results on all the above datasets.

Abstract:In this paper,a novel fixed-time convergent sliding mode control method is proposed for the gait trajectory-tracking problem of lower limb rehabilitation exoskeleton robots.Firstly,the dynamic model of a lower limb rehabilitation exoskeleton robot is achieved through mechanism analysis.Then,a novel nonsingular fast terminal sliding surface is designed to achieve fixed-time convergence and effectively enhance the convergence rate.A new sliding mode control strategy for lower limb rehabilitation exoskeleton robots is designed based on this sliding surface,and the stability and global fixed-time convergence characteristics of the control system are proved by using the Lyapunov criterion.Finally,the proposed control method is experimentally validated via a lower limb rehabilitation exoskeleton robot experimental platform.The experimental results show that the control method proposed in this paper can achieve rapid convergence of lower limb exoskeleton trajectory tracking error,with the error converging to about 1°,strong robustness against wearers of different weights(61 kg and 80 kg)and different gait trajectories(with the period of 8 s and 6 s),and has a certain inhibitory effect on chattering phenomenon.

Abstract:Particle filter(PF)has good estimation performance for nonlinear and non-Gaussian systems,but the lack of particle diversity has always been the vital problem affecting the estimation accuracy of PF after the introduction of resampling technology.Therefore,an intelligent PF method based on hybrid adaptive resampling is proposed.Firstly,a function of computing the covariance matrix adaptively for Gaussian variation is designed in this method on the basis of hybrid adaptive Metropolis-Hastings(M-H)resampling.Secondly,an acceptance and rejection criterion function using the mode of survival of the fittest is developed.Finally,the effective particle set is updated in real time to improve the quality of the particle set and the accuracy of the PF.Two one-dimensional nonlinear models and one high-dimensional nonlinear model are used to verify the effectiveness of the proposed method.Experimental results show that the proposed method can effectively improve the quality of the particle after resampling and improve the estimation accuracy of the PF compared with the existing resampling methods.

Abstract:Aiming at the complexity of the surface assembly production process,the production process is prone to equipment failures and process defects,this paper designs an intelligent fault diagnosis model based on fault knowledge graph for surface assembly production.At the same time,the key technology of knowledge graph construction process-fault entity extraction is studied,and a fault entity extraction model based on BERT-Residual-BiLSTM-CRF for surface assembly production fault logs is designed and implemented.Firstly,the training and testing datasets of the fault entity extraction model are constructed based on the text of surface mount technology(SMT)fault logs,secondly,the TensorFlow framework is used to build the SMT fault entity extraction model,and finally,the trained model is used to conduct controlled experiments.The results show that the fault entity recognition accuracy,recall and mean F-value of the designed fault entity extraction model are improved by about 0.26,0.28 and 0.24 compared with the base model BERT-BiLSTM-CRF,respectively.

Abstract:To improve the tracking control performance of electro-hydraulic servo systems,considering the lumped uncertainties,including parameter uncertainty,unmodeled dynamics,and unknown disturbances,a finite-time backstepping control method based on command filter and neural network is proposed.A Levant differentiator is used as the command filter to obtain the derivative of the differential signal between the virtual input variable and the virtual control law,which not only avoids the"explosion of complexity"in standard backstepping control,but also estimates unmatched uncertainty through reconstruction;Compared with traditional neural network-based backstepping control that requires multiple neural networks,this method only uses one neural network to approximate the matched uncertainty,avoiding the complexity and fragility of controllers caused by multiple neural networks.By introducing piecewise feedback functions composed of a fractional exponential power function and a polynomial function to accelerate convergence,the system achieves finite time stability while avoiding singularity problems.Finally,an experimental platform is established to conduct comparative experiments,and the effectiveness and superiority of the proposed new method are verified.

Abstract:This work focuses on the problem of uncertain parameter estimation and trajectory tracking for wheeled mobile robots.A method for estimating uncertain model parameters of mobile robots based on the convolutional neural network(CNN)is studied,and a primal-dual neural network(PDNN)model predictive control(MPC)tracking control algorithm for mobile robots is proposed.For wheeled mobile robots,tire lateral stiffness is affected by load disturbance,unmodelled dynamics and load changes,which is difficult to measure in real time during actual driving.CNN estimator of lateral stiffness is designed to eliminate uncertainty during robot operation considering the constraint conditions of front wheel deviation and acceleration.This work studies the design of predictive control for mobile robots based on CNN parameter estimation and proposes a PDNN based algorithm with CNN parameter estimation for solving the predictive control problem of mobile robots.The stability of the proposed PDNN-MPC algorithm is proved.Finally,to verify the effectiveness of the controller,the proposed PDNN-MPC algorithm is validated.

Abstract:In this paper,an adaptive anti-disturbance switched control strategy is proposed for switched T-S fuzzy systems subject to multiple sources of disturbance.The multi-source disturbances encompass two components:measurable but unmodeled disturbances and unobservable disturbances modeled by neural networks.Firstly,an adaptive disturbance observer is devised for approximating dynamic neural network modeled disturbances,specifically designed for switched T-S fuzzy systems under the constraint of an average dwell time switching signal.Subsequently,an adaptive anti-disturbance controller is formulated based on the observer.The attenuation performance from the output to the available disturbances is analyzed using the L2 gain performance index.Furthermore,under the constraint of the average dwell time-dependent switching signals,the sufficient conditions for the solvability of the fuzzy adaptive anti-disturbance switching control method are provided.Finally,the rationality and effectiveness of the established adaptive anti-disturbance switched control scheme are validated through a simulation example of a mass-spring-damper system.

Abstract:To address the challenge of achieving high-precision tracking control for piezoelectric actuators(PEAs)in the presence of model uncertainty,external interference,and measurement noise,a novel multi-objective complementary robust control method is proposed in this paper.First,a rate-dependent hysteresis nonlinear model,structured upon the Hammer-stein model,is formulated.This model represents the static hysteresis nonlinear component using the Prandtl-Ishlinskii(PI)model,while the dynamic linear component is characterized by an enhanced correlation identification method.Then,a multi-objective complementary robust control method is employed to achieve the precise tracking control of the piezo-electric actuator.This approach integrates PID control principles to optimize the closed-loop system's performance,in-corporates robust control strategies to ensure the stability of the closed-loop system,and utilizes Youla parameterization to address the conflict between the system's optimal performance and robustness.Finally,the system's tracking accuracy and anti-interference capabilities are verified through experiments,providing empirical evidence for the effectiveness of the proposed methodology in this paper.

Abstract:The nonlinearity,dynamics and variable redundancy of complex industrial processes lead to increase model-ing difficulty and reduce model performance.Therefore,a gated recurrent unit(GRU)network based on attention mecha-nism and nonnegative garrote(NNG)estimation is proposed and applied to actual industrial process soft sensor modeling.Firstly,the temporal attention is introduced into the GRU network,and the attention weights are adaptively assigned ac-cording to the temporal correlation between the implied layers at different moments to improve the model of temporal feature characterization capability.Secondly,an attentional weight vector for process variables is designed and embed-ded with NNG algorithm constraints to approximate unbiased estimates of its coefficients.Then the NNG algorithm with variable attention is used to perform sparse optimization of the GRU network to reduce model complexity,improve its interpretability and prevent overfitting.The effectiveness and superiority of the algorithm are verified by numerical simu-lation.Finally,the proposed algorithm is applied to the soft sensor of SO2 concentration in net flue gas emissions from a coal-fired power plant desulphurization process.The experimental results show that the proposed algorithm outperforms other advanced comparative algorithms and improves its predictive performance while effectively eliminating redundant variables and simplifying the model structure.

Abstract:Uncertainty and variation of mechanical parameters cause the mismatch of control settings in mechatronics servo systems,therefore the dynamic performance and the accuracy cannot be guaranteed continuously.A rapid parameter identification method for moment of inertia,nonlinear friction and bias load is proposed in this paper.It circumvents the introduction of noise in inertial estimation and improves the identification accuracy by exploiting the phasic and kinematic orthogonal properties of a sinusoidal reference to decouple the identifications of inertia and friction.The utilization of separable least square method relieves the computational burden for optimization solution by separating the procedures for optimizing the nonlinear and linear parameters of the friction model.The proposed method could be generalized to identifications of similar mechanical systems.The feasibility and the accuracy of the identification method are verified with a hardware-in-loop simulation respectively,and the application of incremental algorithm of the proposed method in self-tuning and feedforward compensation of speed PI control is studied experimentally.

Abstract:In the field of intelligent transportation,the cooperative tracking control of vehicle platoon systems is an effective means to improve traffic inefficiency,and achieving accurate and fast tracking of vehicle platoon systems is of paramount importance.Therefore,a distributed prescribed-time observer is proposed for cooperative tracking control of vehicle platoon systems.The observer is designed to enable the follower vehicles to successfully follow the state of the leader vehicle at the prescribed time,where the leader vehicle's input is unavailable for some of the follower vehicles.In order to guarantee tracking performance,it is necessary to establish a prescribed performance function.However,stopping and restarting the vehicle platoon systems midway,or changing the dynamics of the leader vehicle,will result in a reselection of the performance function,thus requiring a redesign of the control scheme.In order to eliminate this limitation,a performance function independent of initial conditions is designed in this paper.Based on observed results,a tracking control protocol with prescribed performance is developed to achieve accurate and rapid tracking of the system while ensuring transient and steady state performance.Additionally,to enhance the safety of vehicle platoon systems,an adaptive compensation control scheme is proposed to address actuator faults.The proposed control scheme guarantees that all vehicles maintain a safe distance while the system reaches stability.Finally,the effectiveness of the proposed algorithm is verified by simulation results.

Abstract:A nonlinear adaptive control method is proposed to solve the trajectory tracking control problem for a dual quadrotor cooperative suspension system with unknown disturbances,achieving trajectory tracking of quadrotor unmanned aerial vehicles(UAVs)and suppression of load swing.Firstly,using the D'Alembert principle,the model of the dual quadrotor cooperative suspension system under unknown disturbances is established.Secondly,an appoint-time prescribed performance function is designed to constrain the trajectory tracking error of the quadrotor UAVs,avoiding collision be-tween the quadrotor UAVs.Subsequently,a novel auxiliary variable is constructed to design a nonlinear controller for the underactuated system.Additionally,a neural network is used to approximate system uncertainties and to design an adaptive law to handle external time-varying disturbances to enhance the robustness of the system.The boundedness of all errors in the closed-loop system is proven through Lyapunov stability analysis.Finally,an indoor dual quadrotor suspension transportation experimental platform is used to verify the effectiveness and practical feasibility of the proposed method.

Abstract:With the development of artificial intelligence technology and the popularity of intelligent sensors in the field of unmanned control system,the operational data of various types of unmanned equipment is enriched.Unmanned surface vessel(USV)as an important component of unmanned intelligent equipment,the key link of unmanned aerial vehicle is to control its safety and stability independently.Because of its complex structure and long time operation in unknown environment,it is unavoidable that various abnormal conditions will occur,which will directly affect the capability of unmanned aerial vehicle and reduce its safety and economy.Therefore,it is necessary to accurately predict the unmanned ship's posture.Firstly,time series data is decomposed into several components by using variational mode decomposition,then several types of data in unmanned ship are predicted by adopting a broad-based learning system method.At the same time,the pseudo-inverse solution regression parameters in broad learning are optimized by applying to a Harris Eagle optimization algorithm based on whale algorithm and simulated annealing algorithm.Simulation results show that the proposed method has certain advantages in accuracy and training speed.

Abstract:This paper presents a cooperative turning control method inspired by jackdaw flock information to achieve rapid adjustment of unmanned aerial vehicle(UAV)swarm in dynamic and uncertain environments.The models of the individual interaction and information propagation are established by analyzing the interactive behaviors and information propagation process of jackdaw flock.The relationship between the information exchange and swarm motion is also eluci-dated.The swarm dynamics model based on the social force framework is proposed and mapped to the coordinated turning control of the UAVs driven by turning information.Simulation results indicate that under the influence of individual inter-action,information propagation,and swarm dynamics,the turning information triggered by some UAVs rapidly propagates,and enable the swarm to form a consistent response and thus achieve rapid turning maneuvers.

Abstract:Maintaining the stability of toner concentration to ensure printing quality is a key problem in the engineering application of copiers.In order to solve the problem of long time delay and uncertainty in the toner supply control system of copier,a predictive control strategy based on robust model is designed.Firstly,the mathematical model of the physical characteristics of the toner supply system is obtained by using the system identification method;Then,a robust model predictive controller is designed for the toner supply system of the copier,and an improved whale optimization algorithm(WOA)is introduced to optimize the controller parameters;Finally,the optimal toner supplement is calculated based on the cost function of the controller,and the on-line rolling optimization control of the system is implemented continuously.The experimental results show that compared with the widely used PI controller,the robust model prediction-based toner supply controller has decreased the root-mean-square error of toner concentration change by 8.0％,4.8％and 25.0％,and the standard deviation by 1.4％,5.4％and 14.4％in the tasks of fixed,incremental and stochastic image coverage,which indicates that this method can effectively reduce the dispersion of toner concentration in the printing process,and has a better dynamic performance control effect.

Abstract:Aiming at the problem of cooperative capture of maneuvering target by multiple spacecraft in three di-mensions(3D),a fixed-time convergence cooperative guidance law with terminal angle constraint and time consistency constraint is proposed.The 3D line-of-sight(LOS)coordinate system between spacecraft and target is established,and the acceleration is decomposed into three directions including the direction along the LOS and the directions perpendicular to the LOS.The acceleration along the LOS is designed by algebraic graph theory and distributed cooperative protocol algorithm to realize the time consistency constraint of multiple spacecraft.The accelerations perpendicular to the LOS are designed by sliding mode control theory and fixed-time convergence theory,so that the LOS angles converge to the expected value within a fixed time.In addition,the unknown target acceleration is estimated by a fixed-time observer and compensated in the guidance command.The simulation results show that the proposed cooperative guidance method can achieve the convergence of time-to-go and the terminal LOS angle within a fixed time.

Abstract:As a significant branch of covering problems on networks,many difficulties encountered in real-world complex systems can be viewed as instances of the k-path vertex cover problem.In distributed systems,one of the crucial research issues to achieve network covering optimization is how to design decentralized strategies for autonomous decision-making by agents.In this paper,the k-path vertex cover problem is modeled as a spatial game on networks,where individual vertices act as rational agents and communicate exclusively with their neighbors.This study analyzes the relationship between strong Nash equilibrium(SONE)and the k-path vertex cover state within the context of non-cooperative games.Additionally,the proposed game-based synchronous aspiration-driven algorithm(GSAA)is shown to converge to SONEs of the four-player coalitions within finite time.The effectiveness of the algorithm is validated through numerical simulations.In the context of the k-path vertex cover problem,the link between solutions and game equilibria is examined from a coalition-based perspective.This paper introduces a novel approach for solving distributed optimization problems with local coupling constraints on networks within the framework of game theory.

Abstract:To improve the safety and reliability of the permanent magnet synchronous motor(PMSM)drive system,for the issue of inverter open-circuit faults,this paper proposes an inverter fault diagnosis method based on set-membership estimation.Firstly,the mathematical model of the motor drive system is established using the discrete switch signals and the continuous current state variables of the motor drive system.Subsequently,based on the current estimated by ellipsoid set-membership estimation and the actual current,the ellipsoid residual current is determined.The major semi-axis of the estimated ellipsoid is used as the upper and lower bounds of the threshold.The six types of single-tube faults can be diagnosed and located by determining whether the three-phase ellipsoid residual currents exceed these bounds.This approach ensures that the ellipsoid residual current remains within the threshold during normal system operation while reducing diagnosis time in the event of a fault.In addition,the method only requires unknown but bounded noise,and does not require partial prior knowledge of noise,thereby reducing the conservatism.Simulation comparison shows that manual threshold setting has a high dependence on setting,and the adaptive threshold diagnosis time is relatively long.The proposed method has weak threshold dependence,fast detection speed,short positioning time and high accuracy,and the diagnosis time accounts for about 8％of the current cycle.