This paper proposes a neural network-based optimized robust nonlinear Kalman filter for concurrent fault estimation of high-speed trains using descriptor systems. Considering the displacement and speed time delays and multi-source disturbances acting on every car, and regarding the concurrent actuator and sensor faults as the auxiliary variables of the train, a time-delay nonlinear descriptor system is established. Fuzzy and fuzzy inference layers are added to the hyper basis function neural network; meanwhile, an accelerated gradient algorithm is used for optimizing the basis function of the network. These strategies derive an improved fuzzy hyper basis function neural network to achieve a higher nonlinear multi-source disturbance approximation accuracy, and robust upper bounds are proposed to enhance the filtering accuracy of descriptor systems. Furthermore, a fusion intelligent optimization algorithm using a circle chaotic mapping method to improve the population initialization is proposed to better estimate the unknown system noise. These measures implement better Kalman filtering of descriptor systems to achieve better concurrent fault estimation, even under unknown noise. Simulation results show that, compared with the improved radial basis function neural network-based optimized robust nonlinear Kalman filter, using the proposed method, the displacement, speed, and fault estimation errors are comprehensively decreased, depending on more accurate multi-source disturbance and noise estimations.

This paper focuses on the velocity tracking control problem of high-speed trains (HSTs), in the presence of unpredictable partial loss of actuator effectiveness faults, actuator output magnitude and rate saturations, unknown time-varying model parameters, as well as the unmeasurable additional resistance. An augmented plant of train longitudinal dynamical system is constructed to restrict the actuator dynamics and facilitate the development of the control law. A train fault-tolerant velocity tracking control algorithm is designed by combining the dynamic surface control (DSC) technique, adaptive control technology and auxiliary system. Within the proposed controller, the auxiliary signal is introduced to compensate for the influence of actuator output magnitude saturation; Adaptive technology is adopted to estimate the unknown nominal values of train model parameters, the upper bound of the change of actuator output due to actuator faults and the lumped system uncertainties composed of model parameter uncertainties and additional resistance; The adopted DSC approach is applied to avoid the derivative of virtual control signal, which simplifies the controller. The stability of the closed-loop system is analyzed based on Lyapunov stability theory, and good performance of the presented controller is verified by simulation results.

Bearings are critical components of mechanical equipment, and predicting their remaining useful life (RUL) is important in industry. This paper proposes a RUL prediction method based on the assessment of a bearing’s health status. Features from the time, frequency, and time–frequency domains of the bearing’s vibration signal are extracted to construct a feature set. A multibranch encoder and restricted Boltzmann machine are used to improve the stacked autoencoder to reduce dimensionality. Local weights and health-sample means are introduced into the Mahalanobis distance to improve the health index. Subsequently, the weighted convolutional Euclidean distance serves as the distance metric in K -means clustering to achieve a more accurate health status assessment and provide historical data for RUL prediction. An improved self-attention (ISA) mechanism is proposed by incorporating depthwise separable convolutions and residual-like connections into self-attention mechanisms, enhancing the global and local dependencies of the temporal convolutional network (TCN). Thus, a more accurate RUL prediction is achieved. Comparative and ablation experiments confirm that the proposed ISA-TCN achieves superior predictive accuracy. Generalization experiments further demonstrate its strong adaptability, while anti-noise experiments demonstrate its strong robustness to uncertainties. Finally, experiments on multi-output RUL predictions validate the model’s effectiveness. This approach offers valuable insights for RUL prediction of rotating machinery under real-world scenarios involving variable operating conditions, noise interference, and multi-device environments.

Due to the serious jitter problem of traditional sliding mode controllers (SMC), it cannot effectively control permanent magnet synchronous motors (PMSM). An improved SMC method is designed based on a novel sliding mode reaching law (NSMRL) and applied to the PMSM vector control system to improve these problems. Firstly, the principle of the sliding mode reaching law (SMRL) and the design process are analyzed, and the NSMRL is constructed by designing the parameter adjustment function to realize the dynamic adjustment of the approach velocity. Secondly, the proposed NSMRL was used to replace the SMRL in the traditional SMC and constructing a SMC strategy based on NSMRL (NSMRL-SMC), and a PMSM speed loop controller based on NSMRL-SMC was optimized and designed. Then, the Lyapunov theory was used to analyze the stability of the NSMRL-SMC. Finally, the Matlab/Simulink simulation platform is used for verification. The simulation results show that compared with the traditional SMC, NSMRL-SMC has better dynamic response performance and anti-interference ability.

An adaptive finite time composite fault tolerant control strategy based on an optimized neural network for Attitude control systems (ACSs) of satellites is proposed considering the state time-varying delays, concurrent actuator and sensor faults, system uncertainties, modelable external disturbance and operating noise. An uncertain time-varying state space model for ACSs of satellites is established, and sensor faults are equivalent to actuator-like faults. A disturbance observer is designed for estimating the modelable external disturbance, and an improved dwarf mongoose optimization (DMO) algorithm based on the Levy flight distribution is utilized to optimize the basis function of hyperbasis function neural networks to better estimate the augmented actuator faults that include the actuator fault and the actuator-like fault. Furthermore, an adaptive finite time composite fault-tolerant controller is proposed, which includes the delay-dependent feedback control law, disturbance estimation based-disturbance compensation law and the adaptive fault compensation law based on the augmented fault estimation using the improved DMO-hyper basis function neural network. The finite time boundness of the close-loop dynamics to the uncertainties, operating noise, and augmented actuator faults and the robustness of the measurement to the uncertainties, operating noise and augmented actuator faults are analyzed, and the observer and controller design is formulated as the linear matrix inequalities. Simulation examples for ACSs in different working conditions are considered to exhibit the proposed method's effectiveness. An adaptive finite time composite fault-tolerant controller is proposedLevy flight distribution is introduced to improve the performance of the dwarf mongoose optimization algorithmDisturbance observer is designed to estimate the modelable external disturbance

A sliding mode controller-based control technique is devised to tackle the doubt of the high-speed train dynamics model and the challenge in attaining tracking control of the target trajectory due to outside disturbances. Firstly, a high-speed train dynamics model is established through force analysis of the train. The Lyapunov stability analysis, which guarantees the asymptotic stability of the closed-loop system, was employed to devise a sliding mode controller to tackle the uncertainty of the high-speed train model's nonlinear system. To validate the controller's efficacy, simulation examples were provided.

This article proposes an improved method for fatigue driving detection using the YOLOv5 algorithm, employing three techniques: Global Attention Mechanism (GAM), MPDIoU loss function, and Receptive Field Block (RFB). The core of this approach is to enhance the YOLOv5 algorithm's ability to recognize driver fatigue status, particularly through more accurate analysis of facial expressions and body language. The Global Attention Mechanism (GAM) weights key regions in the image, improving the algorithm's attention to driver facial features and, therefore, more accurately capturing subtle facial expressions related to fatigue. The MPDIoU loss function further optimizes the algorithm's training process, enhancing the model's ability to locate and recognize complex facial features. The introduction of Receptive Field Blocks (RFB) mimics the processing methods of the human visual system, improving the recognition of key features of fatigue driving behavior, such as eye closure, head position, and posture changes. This method has been tested and validated in various driving scenarios. The results demonstrate that, compared to the traditional YOLOv5 algorithm, this improved approach exhibits higher accuracy and reliability in fatigue driving detection.

Due to the huge volume of gearless ball mill direct drive permanent magnet synchronous machine (PMSM), an unwinding modular technology is convenient for its redundant power control, manufacture and maintenance. Several module units are wrapped around the ball mill roller for direct drive. The motor frame adopts a modular structure to achieve complete decoupling between modules. And the cooling problem of the motor is very important for its stable operation. The traditional waterway structure design and cooling analysis methods are difficult to meet the engineering requirements of large modular permanent magnet synchronous machine (MPMSM). A 3D equivalent thermal network method is proposed to study the cooling problem of MPMSM. The design program of modular frame cooling structure is written based on a 3D equivalent thermal network method. According to the temperature distribution of each node, the cyclic check is carried out to quickly design the modular waterway. The cooling effect of the designed waterway structure is analyzed by fluid-solid coupled temperature field simulation. An 80 kW prototype is developed and the temperature rise experiment is carried out. The simulation and experiment results prove the rationality of a 3D equivalent thermal network method for the MPMSM cooling structure design.

In this paper, an optimized adaptive robust extended Kalman filter is proposed based on random weighting factors and an improved whale optimization algorithm for fault estimation of the dynamics of high-speed trains with constant time delays, drastically changing noise and stochastic uncertainties. Robust upper bounds are proposed to improve the performance of the extended Kalman filter by decreasing the influence of the linearization error on filtering for the dynamics of high-speed trains with constant time delays, and its robustness is proven to guarantee the feasibility of the proposed upper bounds. Furthermore, considering drastically changing noise with unknown statistics, a random weighting adaptive algorithm is proposed to implement unbiased noise estimation so that the robust extended Kalman filter can still be implemented well. In addition, a differential evolution algorithm and adaptive parameter are introduced to improve the performance of the whale optimization algorithm so that the stochastic uncertainties are optimized, and the influence of the stochastic uncertainties on filtering is further decreased. The simulation results in the three conditions show that, compared with the variational Bayes adaptive iterated extended Kalman filter, using the proposed method, the position, speed and fault estimation errors are decreased by 31.8%, 33.2% and 28.3%, respectively, on average, which depends on more accurate noise estimation.

针对具有状态定常时滞、测量丢失、执行器故障和噪声测量偏差的惯性系统,提出一种基于白鲸优化算法的优化鲁棒扩展Kalman滤波器.首先,视执行器故障为系统的附加状态变量,建立增广时滞系统.其次,设计鲁棒扩展Kalman滤波器.给出鲁棒上界,降低非线性状态项和非线性状态时滞项的线性化误差;考虑测量丢失环节对滤波精度的影响,在滤波器中引入调节系数降低估计误差.最后,考虑噪声真实值和测量值可能存在的偏差,基于白鲸优化算法优化噪声测量值的协方差矩阵,降低噪声测量偏差对滤波性能的影响.仿真结果表明,相较于基于鲸鱼优化算法和基于灰狼优化算法的优化鲁棒扩展Kalman滤波器,所提方法对状态变量估计的均方根误差分别平均降低了 50.2%和 67.2%,执行器故障估计的均方根误差分别降低了 12.8%和 35.4%;状态估计误差均值分别平均降低了 21.2%和31.4%,执行器故障估计均值分别降低了 9.3%和 16.9%,验证了所提方法的有效性.

针对具有状态时滞和未知不确定性的卫星姿态控制系统,提出一种强跟踪鲁棒扩展Kalman滤波器,以实现执行器和传感器的并发故障估计.首先,考虑系统噪声,视故障为系统的辅助变量,建立增广时滞非线性系统.然后,提出鲁棒扩展Kalman滤波器,引入鲁棒上界以减少线性化误差.进一步,针对系统过程不确定性导致预测协方差精度较低的问题,引入基于多重次优渐消因子的强跟踪算法,以降低不确定性对滤波精度的影响.最后,给出仿真算例,将所提出方法与鲁棒扩展Kalman算法和扩展Kalman算法进行对比仿真.仿真结果表明,相较于其他两种算法,所提出方法的状态估计和故障估计均方根误差的平均值分别降低了 69.2％、60.6％和88.1％、78.9％,仿真结果验证了设计方案的有效性.

In this paper, an optimized long short-term memory (LSTM) network is proposed for the remaining useful life (RUL) prediction of the rolling bearings based on whale optimized algorithm (WOA). The multi-domain features are extracted to construct the feature dataset as the single domain features are difficult to characterize the performance degeneration of the rolling bearing. Considering the possible gradient explosion by training of the rolling bearing lifetime data and the difficulties in selecting the key network parameters, an optimized LSTM network, namely, WOA-LSTM network is proposed. Experiment results show that, compared with the LSTM network, the RUL prediction accuracy of the rolling bearing are improved by the proposed WOA-LSTM network.

In this paper, novel fault estimation and fault-tolerant control methods are proposed for dynamics of high-speed train based on descriptor systems with uncertainties in finite frequency domain. Dynamics of high-speed train is established based on multi-particle model considering that basic resistance is seen as the coefficient of state variables, and additive resistance and the operating noise are seen as multi-source disturbance. Concurrent actuator, sensor faults, and wind gust are considered simultaneously; wind gust is modeled as a disturbance generated by the exogenous system, and an uncertain descriptor system with actuator fault and the exogenous disturbance is established by seeing the sensor fault of high-speed train as the state variables. A robust disturbance-observer-based fault estimation method is proposed to decouple the non-linearity of the descriptor system, so that the combining estimation of the fault and wind gust is implemented. This observer has an unknown input structure, and its gain matrices are formulated as linear matrix inequalities. The observer not only guarantees the augmented state estimation error is asymptotic stable but also the actuator fault estimation and wind gust estimation errors are robust to the multi-source disturbance and the uncertainties. Based on the estimation results, the fault-tolerant controller associated with the state estimation, faults estimation, and wind gust estimation results is proposed to implement a stable close-loop fault-tolerant control for dynamics of high-speed train. Simulation examples are given to illustrate the effectiveness of this method.

In this study, an optimized long short-term memory (LSTM) network is proposed to predict the reliability and remaining useful life (RUL) of rolling bearings based on an improved whale-optimized algorithm (IWOA). The multi-domain features are extracted to construct the feature dataset because the single-domain features are difficult to characterize the performance degeneration of the rolling bearing. To provide covariates for reliability assessment, a kernel principal component analysis is used to reduce the dimensionality of the features. A Weibull distribution proportional hazard model (WPHM) is used for the reliability assessment of rolling bearing, and a beluga whale optimization (BWO) algorithm is combined with maximum likelihood estimation (MLE) to improve the estimation accuracy of the model parameters of the WPHM, which provides the data basis for predicting reliability. Considering the possible gradient explosion by training the rolling bearing lifetime data and the difficulties in selecting the key network parameters, an optimized LSTM network called the improved whale optimization algorithm-based long short-term memory (IWOA-LSTM) network is proposed. As IWOA better jumps out of the local optimization, the fitting and prediction accuracies of the network are correspondingly improved. The experimental results show that compared with the whale optimization algorithm-based long short-term memory (WOA-LSTM) network, the reliability prediction and RUL prediction accuracies of the rolling bearing are improved by the proposed IWOA-LSTM network.

In this paper, a spider monkey optimisation (SMO) algorithm is utilised to identify the parameters of the permanent magnet synchronous motor (PMSM), considering the parameters vary during the motor operation, which affects the sensorless control (SC) performance of the motor. An improved sliding mode observer (SOBS) is proposed for estimating the position and speed of the rotor. First, the SMO algorithm is used to identify the parameters of PMSM. Then, based on the identification results, an improved SOBS is proposed by a piecewise Sigmoid function. Furthermore, the stator position and speed are estimated by extended state observer (ESO) and phase-locked loop (PLL). Finally, a comparison simulation scenario is provided to demonstrate the efficacy of the suggested approach.

This paper proposes a dual attention mechanism gate recurrent unit (DAGRU) neural network model for predicting the overload of the power system load. Based on the GRU neural network, the dual attention mechanism of feature and time is introduced to dig the potential correlation between the output and input characteristics of the power system load. Among which, the feature attention is utilized to analyze the relationship between historical information and input, and then timing attention is utilized to automatically extract the historical information of key points in GRU network to improve the stability of time series prediction. The simulation results show that compared with GRU and LSTM networks, the proposed DAGRU network model improves the prediction accuracy of electrical load.

针对具有状态时变时滞、系统不确定性、可建模扰动、运行噪声和执行器故障的卫星姿态控制系统,提出一种基于扰动观测器的自适应有限时间复合主动容错控制策略.针对可建模扰动设计扰动观测器,然后基于扰动估计误差设计了主动容错控制器.该时滞依赖控制器包含反馈控制项、扰动补偿项和快速自适应故障补偿项.提出的容错控制策略不仅保证闭环系统动态方程的有限时间有界性,而且保证闭环测量输出对于系统不确定性、运行噪声、执行器故障等的鲁棒性.给出控制器增益限制矩阵存在的充分条件及其线性矩阵不等式形式,进而给出仿真算例.仿真结果表明,基于扰动观测器方法,设计的自适应有限时间容错控制器能够快速估计可建模扰动,进而有效地实现系统的闭环容错控制.相较于基于非复合的自适应有限容错控制器,提出的方法对于状态变量的估计均方根误差分别降低了28.9％、4.7％和36.0％;对于可建模扰动估计的均方根误差降低了38.8％.仿真验证了所提方法的有效性.

In this paper, a finite time robust controller is proposed for tracking the speed and position of high speed trains. Considering the aerodynamic flag, mechanical rolling resistance, additive resistance and wind gust, the dynamics of high speed train with time-varying delay is established in longitude. Aerodynamic flag and mechanical rolling resistance are seen as the coefficient of the state variables, wind gust and additive resistance are seen as the external disturbance, and time-varying delay in speed are considered as the train is often running in bad weather or severe working conditions. Then a delay-dependent finite time robust controller is designed. This controller not only guarantees the closed-loop error dynamics error is finite time boundness, but also is robust to the external disturbance consisting of wind gust and additive resistance in finite time. Sufficient conditions of the controller design are given, and the gain matrices of the controller are formulated as linear matrix inequality. A 5-car simulation example is given to show the effectiveness of the proposed method.

本文针对一类非线性系统,提出基于广义系统的鲁棒增广扩展Kalman滤波器,结合改进鲸群优化算法寻优系统噪声,以精确估计系统状态量以及并发执行器和传感器故障.首先,视故障为系统的状态变量,建立广义系统,将非线性系统的故障估计转化为非线性广义系统的状态估计.其次,提出鲁棒上界以降低线性化误差对估计精度的影响.然后,利用改进鲸群算法寻优系统噪声,以优化鲁棒增广扩展Kalman滤波器.最后,给出F-16飞机的纵向运动数值模型,使用本文方法与自适应无迹Kalman滤波器以及基于鲸群算法的鲁棒增广扩展Kalman滤波器进行对比仿真,仿真结果表明,相较于其他两种算法,本文方法的故障估计均方根误差降低了50％左右,验证了其优越性.

This paper studies the problem of fast and accurate fault estimation for a class of descriptor systems with time-varying delays based on an adaptive finite time robust observer. This descriptor system is detectable, and time-varying delay is considered in finite time. Based on this model, the AFTRO is proposed. This observer has a non-singular structure which is easier to calculate, and by the adaptive algorithm, the fast fault estimation is implemented. Disturbance of the descriptor systems is considered, and finite time boundness of the observer is analyzed. Meanwhile, considering the accurate fault estimation, the robustness of the estimation error to the fault is guaranteed by H infinite technology in finite time. Sufficient conditions for the existence of the observer is proved, and using the LMI technology, the calculation of the constraint matrices of the observer is formulated as an optimization problem. Simulation example as three machine bus is given, and the simulation results are presented to illustrate the efficiency of the proposed method.