An adaptive event-triggered H∞ control method is proposed for the uncertain aero-engine distributed control systems with actuator fault,external disturbance,input-output quantization error and network induced time delay.Firstly,the actuator fault,external disturbance,quantization error and system uncertainty are mathematically described,an adaptive event-triggered mechanism is given to improve the utilization of network resources,and a closed-loop system mathematical model considering the above factors is established.Then,the Lyapunov-Krasovskii functional method is utilized to establish a sufficient condition to ensure the asymptotic stability of the system,and a co-design method of adaptive event trigger and H∞controller in the form of linear matrix inequality is given.Under the given simulation conditions,the adaptive event-triggered mechanism saves 94％of the network resources,which is an improvement of 46.4％compared to the ordinary event triggers.The simulation results show that the proposed event trigger and controller can ensure that the system is asymptotically stable despite the undesirable factors considered above.
As the marine environment is complex and changeable,ship navigation is easily affected by factors such as wind,waves,ocean currents and other factors,and ship motion is characterized by nonlinearity and coupling.Aiming at the problem that traditional ship motion prediction methods have insufficient efficiency in extracting time series data and are difficult to achieve high-precision prediction results,a sample convolution and interaction-channel attention(SCI-CA)neural network ship pitch motion prediction model is proposed.The model uses multi-category ship motion attitude data as input,splits the input into two subsequences,utilizes the recursive down sampling.convolution interaction structure of the sample convolution interaction network(SCI),and combines the rich features aggregated from multiple resolutions to improve the utilization of deep features of ship motion data.Then,channel attention mechanism(CA)is used to improve the weight ratio of effective channels,and the residual structure is input to the full connection layer to obtain the final prediction result.The simulation results of real ship data show that the prediction accuracy of the SCI-CA combined model is higher than that of other models,and its mean absolute percentage error(MAPE)and root mean square error(RMSE)are significantly reduced,verifying the effectiveness of the SCI-CA model in predicting ship motion.
When dealing with large-scale multi-objective optimization problem(LSMOP),the MOEA/D algorithm shows poor scalability in the decision space and a tendency to converge to local optima as the dimensionality of decision variables increases.To address this issue,this study proposes a large-scale MOEA/D algorithm with multiple strategies(MSMOEA/D).The MSMOEA/D algorithm introduces a hybrid initialization strategy based on autoencoders in the optimization process to expand the coverage of the initial population,thus promoting global search.Moreover,a neighborhood adjustment strategy based on aggregation functions is proposed,which can more accurately control the search range during the search process by adjusting neighborhood sizes,thereby avoiding low search efficiency caused by excessively large or small neighborhoods.Furthermore,a mutation-selection strategy based on non-dominated sorting is adopted during the optimization process.Different subproblems select their mutation strategies according to the number of individuals in the first level of non-dominated sorting to avoid the population falling into local optima and enhance the overall performance of the algorithm.Finally,the MSMOEA/D algorithm and other existing algorithms are evaluated using LSMOP and DTLZ test problems.Experimental results verify the effectiveness of the proposed algorithm for solving LSMOPs.
To reduce computational loads and save communication resources of multi-missile systems, event-triggered-based distributed differential game guidance laws for the nonlinear cooperative multi-missile system are investigated. First, the cooperative multi-missile interception problem is converted into multi-player zero-sum differential games, where all missiles minimize the cost function while the target maximizes the cost function. Second, event-triggered conditions are designed, for implementing the proposed guidance laws, Adaptive Dynamic Programming technique is adopted to approximate the couple event-triggered Hamilton-Jacobi–Isaacs equation, and adaptive weight-tuning laws are proposed. Furthermore, the proposed guidance laws can guarantee the stability of the nonlinear multi-missile system, which is proved by Lyapunov theory. The estimated weights of neural network and consensus tracking errors are ensured to be Uniformly Ultimately Bounded. Besides, a minimal inter-sample time is established to avoid the Zeno behavior. Finally, experiments show that the number of controller updates can be reduced to 65.36% by the proposed algorithm.
This article considers the distributed economic model predictive control (DEMPC) scheme for addressing the load frequency control problem in a multiarea interconnected power system with wind turbines. The system is divided into multiple dynamically coupled subsystems, each subjected to state and control input constraints due to safety concerns. The overall optimal control problem is decomposed into several local optimal control problems based on the local information of each subsystem, meaning each area designs its own local DEMPC controller. Within this framework, the future state trajectories of neighboring subsystems are estimated from the transmitted information between neighbors. To enhance overall economic benefits, the economic stage cost, including load frequency regulation cost, fuel consumption cost, and wind generation cost, is incorporated into the cost function. Simulation results and analysis under different scenarios demonstrate potential improvements in computational burden, economic performance, and robustness of the designed DEMPC controller.
To alleviate the pressure of network data transmission,the output feedback H∞ controller design problem is investigated for a class of saturated 2-D systems described by the Roesser model under the dynamic event-triggered mechanism.First,the mathematical model of such an uncertain saturated 2-D system is developed.Second,in order to reduce the update frequency of the signal,a dynamic event-triggered mechanism is proposed to determine whether the current information can be transmitted to the network.Based on the Lyapunov function method,a sufficient condition is given to ensure that the corresponding closed-loop system meets the certain H∞ performance index,and the existence criterion for the output feedback controller is obtained by introducing a non-negative scalar μ.Moreover,with the aid of the cone complement linearization algorithm,the design problem of the output feedback controller can further be transformed into a nonlinear minimization problem constrained by linear matrix inequalities.Finally,the numerical simulation is presented to verify the effectiveness and feasibility of the output feedback control scheme.
>Notably, owing to the limited information available, the problem of output feedback stabilization/regulation for nonlinear systems has been explored based on state observers.In addition, the sensors accounting for state measurement could fail to detect the system states accurately due to the limitations of manufacturing techniques and instruments.
It is unquestionable that hesitant fuzzy set has developed into an important tool for handling uncertain data. As a measure between data, hesitant fuzzy entropy is significant in the decision-making process. However, existing entropy measures have a series of problems such as anti intuition. In the decision-making problem under hesitant fuzzy frameworks, traditional decision-making methods such as TOPSIS method assume that the decision-maker is absolutely rational, which also makes the calculation results unreasonable. Therefore, the present study provides a comprehensive hesitant fuzzy entropy. Additionally, the traditional TOPSIS method is enhanced based on the cumulative prospect theory and the entropy method, thereby generating a multi attribute decision-making model with unknown attribute weights. First, a new axiomatic definition of hesitant fuzzy entropy is given, then the comprehensive hesitant fuzzy entropy is defined based on the two features of fuzziness and unclarity. Besides, we prove that our method meets the new axiomatic definition. Finally, the improved TOPSIS method is applied to evaluate the comprehensive entropy measure, while simulations verify the effectiveness of the proposed method. The results showed that compared to existing entropy measures, the comprehensive hesitant fuzzy entropy more accurately reflects the uncertainty of hesitant fuzzy elements, and also has the advantages of simple calculation and easy understanding. Meanwhile, the improved TOPSIS method takes into account the psychological preferences of decision-makers, which is more reasonable than traditional methods.
This paper proposes a novel filter-based immersion and invariance (I&I) adaptive method for nonlinear systems with additive disturbances and parameter uncertainties. The key innovation of the proposed method is a filter construction that involves the dynamics of the system, based on which the I&I parameter estimator with a sigma modification term is designed. Comparing existing I&I methods, the expression for the novel filter-based I&I estimator is given in terms of total derivatives rather than partial derivatives, which is no longer subject to the integrability condition. In combining the sigma modification term, the proposed parameter estimator also guarantee the uniformly ultimately bounded stability of the parameter estimation error when the system is disturbed, which cannot be achieved by the original I&I method. Furthermore, Lyapunov theory demonstrates that the proposed method can guarantee the stability of the nonlinear systems discussed in this paper. Simulation and experimental results confirm the effectiveness of the proposed method.
Dynamic multiobjective optimization problems exist in daily life and industrial practice. The objectives of dynamic multiobjective optimization problems conflict with each other. In most dynamic multiobjective optimization algorithms, the decision variables are optimized in the same way, without considering the different characteristics of the decision variables. To better track Pareto-optimal front and Pareto-optimal set at different times, a dynamic multiobjective optimization algorithm based on decision variable relationship (DVR) is proposed. Firstly, the decision variables are divided into two categories based on the detection mechanism of the contribution of decision variables to diversity and convergence. Secondly, different optimization methods are used for different types of decision variables. And a diversity maintenance mechanism is proposed. Finally, the individuals generated by these two parts and the perturbed individuals are combined. The combination individuals are nondominated sorted to form a population in the new environment. To verify the performance of the proposed algorithm, DVR is compared with five state-of-the-art dynamic multiobjective optimization evolutionary algorithms on 15 benchmark instances. The experimental results show that the DVR algorithm obtains 24 inverse generation distance optimal values in 45 groups of test data.
The coordinated design of comprehensive security control and communication is investigated for an industrial cyber-physical system (ICPS) that incorporates stealthy false data injection (FDI) and actuator faults, using data-driven and mechanism analysis methods. Firstly, replacing the conventional discrete event-triggered communication scheme with an adaptive discrete event-triggered communication scheme (ADETCS), a resilient ICPS framework is developed to combat network FDI attacks and actuator faults. Secondly, based on data-driven technology, the ELM prediction model of the FDI attack is established to reconstruct and compensate for the attack accurately. Then, by employing the affine Bessel-Legendre inequality and the augmented Lyapunov-Krasovskii (L-K) functional, the analytical solutions for the robust observer and integrated safety controller are derived. Lastly, the effectiveness of the proposed approach is validated through a quadruple-tank example.
Aiming at the cluster consensus problem for a class of unknown discrete time nonlinear multi-agent systems with denial of service(DoS) attacks, a model-free adaptive control algorithm is proposed. It is assumed that the system has a fixed topology and only part of the agents in each cluster can receive the leader information. Firstly, a dynamic linearization method is used to construct a data relationship model for multi agent systems, and the periodic DoS attack model conforms to Bernoulli distribution is constructed by limiting the corresponding attack time and frequency. A data-driven cluster tracking control protocol is designed by combining the cluster consensus error under the coupling effect among agents. Then, the sufficient conditions to ensure the convergence of tracking error under the sense of mathematical expectation are given by the method of compression mapping, and theoretically analyzes the convergence of the algorithm. Finally, simulation results verify the effectiveness of the proposed algorithm.
This paper investigates the finite-time stabilization of maglev system with suspension air gap limitation utilizing event-triggered technique. A subtle event-triggered controller is constructed on the basis of a finite-time trigger mechanism with a time-varying threshold and tangential barrier function. This control algorithm ensures that the suspension air gap is limited to a specific vicinity while the states of the maglev system converge to the origin within a finite time. The contribution lies in the fact that the resources consuming is reducing effectively while the design and the theory analysis for the constrained and unconstrained suspension air gap are unified. Finally, the practical simulation validates the efficiency and superiority of the proposed approach.
For most multivariate time series segmentation algorithms,the selection of segmentation points and the determination of the number of segments often need to be completed independently,which greatly increase the computational complexity of the algorithm.In order to solve the above problem,an adaptive greedy Gaussian segmentation algorithm based on multivariate time series is proposed.The algorithm interprets the data points from the segmentations of multivariate time series as independent samples of different multivariate Gaussian distributions,and then transforms the segmentation problem into a covariance-regularized likelihood maximization problem to solve.In order to improve the learning efficiency,the greedy search method is adopted to maximize the likelihood value of each segment to find the optimal segment point approximately.During the search process,the information gain method is adopted to adaptively obtain the optimal number of segments,which avoids from realizing the determination of the number of segments and the selection of segment points independently to reduce the computational complexity.The experimental analysis is carried out on real datasets in many different fields.Compared with traditional methods,the proposed method can obtain higher accuracy and efficiency,and is able to detect outliers in multivariate time series effectively.
In this paper, within the framework of nest algebra, we study the stability and the robust stability of the two-port networked control system(NCS), when the interference and distortion of communication channels are subject to norm boundedness. Using the phases of operators based on the numerical range, the stability for one-stage two-port NCS is obtained. For the cascaded two-port NCS, we prove that if the norms of communication channels are less than 1, the system is stable. And an example is presented to demonstrate the effectiveness of the proposed method. Further, we also obtain a sufficient condition for the stability of the two-port NCS when the nominal system is stable. Finally, the robust stability of the two-port NCS is derived by combining the equivalence of path-connected sets and complementary quadratic constraint conditions.
As for the optimal scheduling issue of closely coupled the path planning of automated guided vehicles (AGVs) and task allocation in flexible manufacturing systems (FMSs), a heuristic optimization method is proposed based on Petri nets and an artificial potential field (APF). First, a manufacturing system with AGVs is described as a task Petri net and a path one, and then its Petri net model is obtained by composing the two nets together. Second, the topology of the Petri net model is used to design the potential energy parameters for the network junctions, and, consequently, the APF is designed for the Petri net model. Third, a heuristic functions is proposed to estimate the minimal terminate times of the Petri net model by means of its APF. Further, a heuristic beam search algorithm is presented to calculate a near optimal schedule by the Petri net model and its APF. Finally, numerical experiments are carried out, and the results show that optimal or near-optimal schedules can be calculated in the reasonable times by our algorithm.
基于深度学习的红外与可见光图像融合算法通常无法感知源图像显著性区域,导致融合结果没有突出红外与可见光图像各自的典型特征,无法达到理想的融合效果.针对上述问题,设计一种适用于红外与可见光图像融合任务的改进残差密集生成对抗网络结构.首先,使用改进残差密集模块作为基础网络分别构建生成器与判别器,并引入基于注意力机制的挤压激励网络来捕获通道维度下的显著特征,充分保留红外图像的热辐射信息和可见光图像的纹理细节信息;其次,使用相对平均判别器,分别衡量融合图像与红外图像、可见光图像之间的相对差异,并根据差异指导生成器保留缺少的源图像信息;最后,在TNO等多个图像融合数据集上进行实验,结果表明所提方法能够生成目标清晰、细节丰富的融合图像,相比基于残差网络的融合方法,边缘强度和平均梯度分别提升了64.56%和64.94%.
Aiming at the problem of poor control effect of active disturbance rejection control（ADRC） for thermal system with high order and large inertia characteristics, a design method of ADRC based on high-order system model information compensation is proposed. Based on the theoretical analysis, the physical significance of each parameter and the quantitative parameter tuning method are given, and the reason why the proposed ADRC can improve the control effect compared with the standard ADRC controller is analyzed from the perspective of observation error, open-loop frequency characteristic and stable region. The result of the comparative simulation and robustness test show that the proposed method has better control effect than PI/PID in set point tracking, disturbance rejection ability and performance robustness, and it can also significantly improve the control effect of low-order ADRC on high-order large inertia process,which has good engineering promotion potential.
当Q学习应用于路径规划问题时,由于动作选择的随机性,以及Q表更新幅度的有限性,智能体会反复探索次优状态和路径,导致算法收敛速度减缓.针对该问题,引入蚁群算法的信息素机制,提出一种寻优范围优化方法,减少智能体的无效探索次数.此外,为提升算法初期迭代的目的性,结合当前栅格与终点位置关系的特点以及智能体动作选择的特性,设计Q表的初始化方法;为使算法在运行的前中后期有合适的探索概率,结合信息素浓度,设计动态调整探索因子的方法.最后,在不同规格不同特点的多种环境中,通过仿真实验验证所提出算法的有效性和可行性.
针对机器视觉场景图像中由于雨线影响导致背景信息模糊、损失的问题,提出一种基于倍频卷积和注意力机制的图像去雨方法.首先,建立基于空-频域去雨模型,设计基于空间尺度变换和倍频卷积的频率特征分解模块,通过学习得到频率特征和雨线特征的映射关系,降低低频特征空间冗余,提高网络运行效率;其次,设计多层通道注意力模块映射雨线层权重信息,增强重要特征,挖掘雨线层之间的亮度差异,提高雨线检测性能;最后,通过序列操作迭代分解出不同成分的雨线信息,进而完成场景图像去雨.实验结果表明,所提方法对不同方向、形状的雨线和雨滴具有良好的去除性能,同时对于背景图像的细节与边缘信息也具有较好的保护作用.
