ISA transactions最新文献

Multiple processes connected closely during the endless strip production (ESP) rolling, it is difficult to obtain the global optimal solution by multi-objective modelling of a single process, and the parameters to be optimized coupled with each other. To obtain the optimal solution, a multi-objective optimization model combining the power consumption, product quality, and loading balance was proposed for the design of an ESP rolling schedule. The thickness and heating temperature were simultaneously taken as the decision variables for coupling the temperature and loading in the rolling process, and the non-dominated sorting genetic algorithm-II (NSGA-II) based on differential evolution (NSGA-II-DE) was applied to obtain the Pareto solutions. To select an optimal solution, a satisfaction function was designed and applied to fully utilize the Pareto solutions. Furthermore, to prove the precision and efficiency of the method, the online schedule and that obtained by the NSGA-II method were compared. The results proved that the final selected solution had better quality and a more balanced loading force than the other two types, which could provide guidance for the actual production process.

Improving the flexible and deep peak shaving capability of supercritical (SC) unit under full operating conditions to adapt a larger-scale renewable energy integrated into the power grid is the main choice of novel power system. However, it is particularly challenging to establish an accurate SC unit model under large-scale variable loads and deep peak shaving. To this end, a data-driven modeling strategy combining Transformer-Extra Long (Transformer-XL) and quantum chaotic nutcracker optimization algorithm is proposed. Firstly, three models of the SC unit under once-through/recirculation/shut-down are built via analyzing its mechanism of the operation process, respectively. Secondly, the superior performance of Transformer-XL in obtaining global feature information is employed to effectively solve the problem of high information dependence caused by the strong coupling and nonlinearity of SC unit. Then, the improved quantum chaotic nutcracker optimization algorithm with higher search accuracy is proposed to obtain the optimal parameters of Transformer-XL based on the logistic chaotic mapping and quantum thinking. Feature information dependencies and optimal parameter settings are fully considered in the proposed modeling scheme, which results in an accurate model of SC unit under full operating conditions. Finally, various simulations and comparisons are conducted based on the on-site data of 600 MW SC unit to demonstrate the superiority of the proposed data-driven modeling strategy. According to the improved Transformer-XL, the mean square errors of the proposed SC unit model under once-through/recirculation/shut-down modes are less than 2.500E-03, which verifies the high accuracy of the model. Consequently, the developed model is suitable for application in the controller designing and the operating efficiency and flexibility improvement of SC unit.

Traditional data-driven models for predicting rare earth component content are primarily developed by relying on supervised learning methods, which suffer from limitations such as a lack of labeled data, lagging, and poor usage of a major amount of unlabeled data. This paper proposes a novel prediction approach based on the BiLSTM-Deep autoencoder enhanced traditional LSSVM algorithm, termed BiLSTM-DeepAE-LSSVM. This approach thoroughly exploits the implicit information contained in copious amounts of unlabeled data in the rare earth production process, thereby improving the traditional supervised prediction method and increasing the accuracy of component content predictions. Initially, a BiLSTM autoencoder is established for unsupervised training on the rare earth production process data, enabling the extraction of inherent time series characteristics. Subsequently, boolean vectors are introduced in the Deep autoencoder training process to perform masking operations on the input data, simulating scenarios with noise and missing data. This is facilitated by their adherence to Bernoulli distributions, which allow for the random setting of certain input vector dimensions to zero. Additionally, the Deep autoencoder is capable of extracting high-dimensional implicit features from the data. After that, the conventional supervised prediction technique, least squares support vector machine (LSSVM), is fused with the implicit characteristics derived from the well-constructed BiLSTM-Deep autoencoder, culminating in the creation of a prediction model for rare earth component content. Ultimately, the simulation verification using LaCe/PrNd extraction field data demonstrates the effectiveness of the proposed approach in harnessing substantial quantities of unlabeled data from the rare earth extraction production process, thereby bolstering the accuracy of model predictions.

This study investigates pigeon-like flexible flapping wings, which are known for their low energy consumption, high flexibility, and lightweight design. However, such flexible flapping wing systems are prone to deformation and vibration during flight, leading to performance degradation. It is thus necessary to design a control method to effectively manage the vibration of flexible wings. This paper proposes an improved rigid finite element method (IRFE) to develop a dynamic visualization model of flexible flapping wings. Subsequently, an adaptive vibration controller was designed based on non-singular terminal sliding mode (NTSM) control and fuzzy neural network (FNN) in order to effectively solve the problems of system uncertainty and actuator failure. With the proposed control, stability of the closed loop system is achieved in the context of Lyapunov's stability theory. At last, a joint simulation using MapleSim and MATLAB/Simulink was conducted to verify the effectiveness and robustness of the proposed controller in terms of trajectory tracking and vibration suppression. The obtained results have demonstrated great practical value of the proposed method in both military (low-altitude reconnaissance, urban operations, and accurate delivery, etc.) and civil (field research, monitoring, and relief for disasters, etc.) applications.

Spindle thermal errors significantly influence the machining accuracy of machine tools, necessitating precise modeling. While deep learning methods are commonly used for this purpose, their generalization ability and performance largely depend on design of the network structure and the selection of hyperparameters. To address these challenges, this study proposes a neural network model that integrates Bayesian optimization (BO) with dilated convolution neural network (DCNN). Dilated convolutions enhance traditional CNN models by using a dilation rate, which allows the convolutional kernel to cover a larger receptive field without increasing parameter count or computational cost. To prevent local optima during hyperparameter tuning, a Bayesian algorithm based on Gaussian processes (GP) is utilized, which optimizes 9 critical hyperparameters in the DCNN. Experimental results demonstrate that the proposed model achieves over 95 % accuracy in predicting radial thermal errors for both heating and cooling states in the X and Y directions.

Accurate load-sharing and circulating current mitigation is a particular problem in parallel-connected inverters without intercommunication. This paper analyzes how using multiple-resonant controllers in the voltage regulation loop impacts the parallel operation of uninterruptible power supplies (UPSs) with droop control. It is shown how adding damping factors of all modes of the resonant controller reduces the circulating current between the UPSs, also improving the power-sharing closed-loop stability and performance. Moreover, the proper choice of these damping coefficients can replicate the effects of adding virtual resistance loops to the droop controller structure. An analysis of such coefficients on the stability margins of the voltage regulation system is carried out and the methodology is validated by experimental results considering the parallelism of two 3.5 kVA UPSs with parametric differences in their LC filters.

Sensor faults can contaminate system measurements such that reliable estimation cannot be achieved and further cause abnormalities in multiagent systems (MASs). Addressing this problem, this study develops an event-triggered fault-tolerant tracking control (FTTC) protocol. The descriptor approach is first used to construct extended states and eliminate sensor faults from the measurement equation. A sliding-mode observer is then tailored based on the transformed system to realize state estimation and fault diagnosis (FD) simultaneously. Subsequently, an event-triggered distributed estimator is developed to isolate the uncertainty's influence and reduce communication overhead while estimating the leader's state. By incorporating the estimator and observer outputs, an FTTC protocol is developed to ensure stability under actuator/sensor faults. Finally, the investigated FTTC method is validated with a numerical simulation of multiple quadrotors.

In this paper, we investigate the event-based tracking control for two-wheeled mobile robots using a sliding mode control strategy. To address the conflict between the singularity problem and finite-time performance, a new nonsingular terminal sliding mode controller enabling mobile robots to achieve the tracking goal through a wireless network is developed. Further, redesign the controller using sampling information, in which an event condition is introduced to determine the sampling sequence, and the event-triggered controller avoids the high gain situation through the proposed sliding variables. The Zeno phenomenon for event condition is excluded by proofing the existence of minimal positive interevent execution time. Finally, an experiment has been implemented on a remote computer transmitting control signals to a mobile robot, demonstrating the effectiveness and applicability of the designed controller.

The hydraulic secondary regulation drive system employs a hydraulic servo motor to achieve precise position tracking and zero throttling loss, but it faces challenges such as high inertia, low damping, and high system order, leading to suboptimal control accuracy. Traditional adaptive robust control methods struggle with the control challenges of such high-order systems. This paper introduces a cascaded control approach based on adaptive robust control to address these issues. A fifth-order model is developed to account for significant load inertia, dividing the system into inner and outer control loops. The outer loop applies adaptive robust control to handle uncertainties and load disturbances for accurate rotational position control, while the inner loop uses swashplate disturbance compensation robust control to manage torque disturbances and achieve precise displacement control. A cascaded Lyapunov function is designed to address the coupling effects between the errors of the inner and outer loop controllers, ensuring stability across both subsystems. Experimental results show that the proposed method's position tracking accuracy exceeds that of cascade dual-PID control methods by 50% to 80% and traditional adaptive robust control methods by 30% to 40% under sinusoidal frequency commands of 0.1 Hz and 0.25 Hz.

With the growing size of the system, this distributed Kalman filter (DKF) is widely used in multi-sensor networks. However, it is difficult for DKF to accurately estimate state values in non-Gaussian noise environments. In this paper, a regression equation is first constructed to contain all sensor node information. Then, by bringing the minimum error entropy with fiducial points (MEEF) standard into the process of information fusion, a robust algorithm named centralized MEEF KF (CMEEF-KF) is presented, which is robust to non-Gaussian noise and unusual data. Furthermore, to overcome the communication burden of CMEEF-KF in sensor networks, the distributed MEEF-KF (DMEEF-KF) is developed, which construct a framework of consensus average method for node information fusion. Specifically, each sensor only exchanges the key information with its neighborhoods. In addition, in order to make the algorithm able to cope with the nonlinear state estimation problem, the distributed MEEF extended Kalman filter is also proposed. Eventually, the effectiveness of the suggested algorithms is demonstrated by land vehicle navigation and power system tracking state estimation using a 10-node sensor network.