Jianfeng Huang

Smart manufacturing systems aim to enhance the efficiency, adaptability, and reliability of industrial operations through advanced data-driven approaches. Achieving these objectives hinges on accurate fault detection and timely maintenance, especially in highly dynamic industrial environments. However, capturing nonlinear and temporal dependencies in dynamic nonlinear industrial processes poses significant challenges for traditional data-driven fault detection methods. To address these limitations, this study presents an Incremental Pyraformer–Deep Canonical Correlation Analysis (DCCA) framework that integrates the Pyramidal Attention Mechanism of the Pyraformer with the Broad Learning System for incremental learning in a DCCA basis. The Pyraformer model effectively captures multi-scale temporal features, while the BLS-based incremental learning mechanism adapts to evolving data without full retraining. The proposed framework enhances both spatial and temporal representation, enabling robust fault detection in high-dimensional and continuously changing industrial environments. Experimental validation on the Tennessee Eastman (TE) process, Continuous Stirred-Tank Reactor (CSTR) system, and injection molding process demonstrated superior detection performance. In the TE scenario, our framework achieved a 100% Fault Detection Rate with a 4.35% False Alarm Rate, surpassing DCCA variants. Similarly, in the CSTR case, the approach reached a perfect 100% Fault Detection Rate (FDR) and 3.48% False Alarm Rate (FAR), while in the injection molding process, it delivered a 97.02% FDR with 0% FAR. The findings underline the framework’s effectiveness in handling complex and dynamic data streams, thereby providing a powerful approach for real-time monitoring and proactive maintenance.

Accurate energy consumption prediction in the injection molding process is crucial for optimizing energy efficiency in polymer processing. Traditional parameter optimization methods face challenges in achieving optimal energy prediction due to complex energy transmission. In this study, a data-driven approach based on the Rolling Learning Informer model is proposed to enhance the accuracy and adaptability of energy consumption forecasting. The Informer model addresses the limitations of long-sequence prediction with sparse attention mechanisms, self-attention distillation, and generative decoder techniques. Rolling learning prediction is incorporated to enable continuous updating of the model to reflect new data trends. Experimental results demonstrate that the RL-Informer model achieves a normalized root mean square error of 0.1301, a root mean square error of 0.0758, a mean absolute error of 0.0562, and a coefficient of determination of 0.9831 in energy consumption forecasting, outperforming other counterpart models like Gated Recurrent Unit, Temporal Convolutional Networks, Long Short-Term Memory, and two variants of the pure Informer models without Rolling Learning. It is of great potential for practical engineering applications.

Coal-fired power plants account for a large share of the power generation market in China. The mainstream method of desulfurization employed in the coal-fired power generation sector now is wet flue gas desulfurization. This process is known to have a high cost and be energy-/materially intensive. Due to the complicated desulfurization mechanism, it is challenging to improve the overall sustainability profile involving energy-, cost-, and resource-relevant objectives via traditional mechanistic models. As such, the present study formulated a data-driven many-objective model for the sustainability of the desulfurization process. We preprocessed the actual operation data collected from the desulfurization tower in a domestic ultra-supercritical coal-fired power plant with a 600 MW unit. The extreme random forest algorithm was adopted to approximate the objective functions as prediction models for four objectives, namely, desulfurization efficiency, unit power consumption, limestone supply, and unit operation cost. Three metrics were utilized to evaluate the performance of prediction. Then, we incorporated differential evolution and non-dominated sorting genetic algorithm-III to optimize the multiple parameters and obtain the Pareto front. The results indicated that the correlation coefficient (R2) values of the prediction models were greater than 0.97. Compared with the original operation condition, the operation under optimized parameters could improve the desulfurization efficiency by 0.25% on average and reduce energy, cost, and slurry consumption significantly. This study would help develop operation strategies to improve the sustainability of coal-fired power plants.

This paper proposes a method and an algorithm called Q-sorting for reinforcement learning (RL) problems with multiple cumulative constraints. The primary contribution is a mechanism for dynamically determining the focus of optimization among multiple cumulative constraints and the objective. Executed actions are picked through a procedure with two steps: first filter out actions potentially breaking the constraints, and second sort the remaining ones according to the Q values of the focus in descending order. The algorithm was originally developed upon the classic tabular value representation and episodic setting of RL, but the idea can be extended and applied to other methods with function approximation and discounted setting. Numerical experiments are carried out on the adapted Gridworld and the motor speed synchronization problem, both with one and two cumulative constraints. Simulation results validate the effectiveness of the proposed Q-sorting in that cumulative constraints are honored both during and after the learning process. The advantages of Q-sorting are further emphasized through comparison with the method of lumped performances (LP), which takes constraints into account through weighting parameters. Q-sorting outperforms LP in both ease of use (unnecessity of trial and error to determine values of the weighting parameters) and performance consistency (6.1920 vs. 54.2635 rad/s for the standard deviation of the cumulative performance index over 10 repeated simulation runs). It has great potential for practical engineering use.

Multispeed transmissions are helpful for improvement of the economy and drivability of electric vehicles (EVs). In this paper, we propose a two-speed transmission based on dual planetary gear mechanism, in which shifts are realized by torque transfer between two brakes located on ring gears. To synthesize the dynamic and economic performances of the vehicle, a multiobjective optimization problem is constructed for gear ratio optimization and Pareto-optimal solutions of gear ratio combinations are obtained by Nondominated sorting genetic algorithm-II (NSGA-II). In particular, the minimum electric energy consumption of the EV is calculated with a fast Dynamic Programming (DP) in each iteration. Following this, a constant-output-torque control (COTC) scheme is adopted for the torque phase and inertia phase of gearshift process to ensure constant output torque on the wheel. To enhance transient responses, the feedforward–feedback controller structure is applied and a disturbance observer is integrated to improve robustness. Simulation results demonstrate that the two-speed transmission has much better performance in terms of acceleration time and energy economy compared to the fixed-ratio transmission, and the proposed gearshift control method is able to achieve fast and smooth gear shift robustly while maintaining constant output torque.

Precise motor speed regulation control is essential to achieve a good gear shifting quality of the integrated motor-transmission (IMT) system, in which the relative speed between outgoing shaft and the gearwheel to be engaged can be eliminated directly through regulation of the motor speed. The speed regulation control confronts the difficulty that there exist external disturbances on the motor shaft, like the unknown load torque arised from bearing friction, oil shearing and oil churning, etc. To deal with these disturbances, a robust speed regulation controller combined a nominal proportional control and integral sliding mode control is proposed. The former is designed to achieve a good speed tracking performance and the latter provides functionality of disturbances rejection. The effects of different controller parameters for the robust controller design are assessed via simulations. Moreover, to verify the effectiveness of the combined control scheme in practical engineering use, experiments are carried out on a test bench with a real IMT powertrain system. Results indicate that the proposed approach can attain a rapid and smooth speed regulation process with a simple controller structure and good robustness.

This paper presents a gear shift method for the dual clutch transmission (DCT) with integrated electric motor in pure electric drive mode. In contrast to clutch-to-clutch shift in conventional DCT, a good gear shifting process relies on the coordinated control of the motor and synchronizer in electric drive mode of the hybrid DCT. To shorten the torque interruption time and reduce the wear of the synchronizer during engagement, the key point is to adjust the oncoming gear speed to the output shaft speed rapidly. This study provides a speed regulation control framework based on model predictive control (MPC) and disturbance observer (DO), where the MPC controller is designed to achieve a good tracking performance and the DO is to eliminate effects from exogenous disturbances. Simulation and experimental results demonstrate that the proposed approach can attain a rapid and robust gear shifting performance.

This paper compares several kinds of methods for motor speed synchronization control, which plays an important role in gear shifting of emerging clutchless automated manual transmissions for battery electric vehicles and other powertrain systems for hybrid electric vehicles. Specifically, four controllers, namely, proportional-integral control, disturbance observer–based proportional control, integral sliding mode–based proportional control, and disturbance observer–based act-and-wait control, are compared in transient responses regarding settling time, overshoot, and steady-state error. Each is formulated by combining one nominal controller for speed tracking and another for disturbance compensation. For disturbance observer–based proportional control, model-based determination of the feedback gain and disturbance observer gain is discussed. Simulations and experiments are carried out to study effects of different controller parameter settings and evaluate performances of different methods under different operation conditions with disturbances and uncertainties like plant parameter drift, communication time delay, unknown load torques, and so on. Based on the results, it is advised that disturbance observer–based proportional control with the reference signal set as the nominal speed trajectory be used in practice, since it comes with merits of few controller parameters to be determined as well as robust, consistent, and satisfactory performances.

Motor speed synchronization is important in gear shifting of emerging clutchless automated manual transmissions (AMT) for electric vehicles and other kinds of parallel shaft-based powertrains for hybrid electric vehicles. This paper proposes a speed synchronization controller design for a kind of system integrating a traction motor and a dual clutch transmission (DCT), using optimal control and disturbances compensation. Based on the relativity between magnitudes of different system parameters, the optimal control law is simplified into the proportional (P) one to ease design and analysis. Relationship between the feedback gain and the duration of speed synchronization process is derived in an explicit way to facilitate model-based determination of controller parameters. To alleviate overshoot while maintaining predesigned performances, the explicit nominal speed trajectory rather than the fixed setpoint speed is chosen as the reference signal. To improve robustness of the controller, a time-domain disturbance observer (DO) is added to cancel effects from parameter drift, unmodeled dynamics, and other exogenous disturbances. As a result, the proposed controller possesses merits of few controller parameters to be determined, good transient response, and robustness. These features make it suitable for practical engineering use. Simulation and experiment results verify its effectiveness in attaining both a fast and small-overshoot speed synchronizing process.