Search Results (35)

Finite-control-set model predictive control provides excellent torque–speed regulation for switched reluctance motor drives but requires an online combinatorial search at every control instant, making low-cost embedded implementation challenging. This article investigates whether offline reinforcement learning can distill policies from multiple classical controllers into a single feedforward policy requiring neither online optimization nor controller gain tuning. A replay buffer is populated with trajectories generated by three demonstrators—hysteresis current control, proportional–integral control with pulse-width modulation, and finite-control-set model predictive control—using a finite-element model of a four-phase 8/6 switched reluctance machine parameterized from measurements of the physical drive. An implicit Q-learning agent then learns a control policy without evaluating actions outside the offline dataset. The central finding is that demonstration diversity governs the stability of offline reinforcement learning on this problem: policies trained from a single demonstrator experience early mode collapse in all fifteen runs, whereas two- or three-demonstrator datasets converge stably in all fifteen. Behavior cloning trained on the identical buffer, split, architecture, and deployed controller provides the reference point for interpreting this result. It matches the offline RL policy on torque quality and improves on its speed regulation, exhibiting none of the seed-to-seed fragility seen at no load while requiring roughly $8\%$ more switching transitions. The stability requirement therefore appears to be a property of the advantage-weighted offline RL objective rather than the control task, and the measured benefit of that objective on this problem is confined to switching effort. We report this rather than claim a broader advantage. The characterization of the distilled controller shows that it generalizes to operating points that are not included in the training dataset, gains nothing systematic beyond approximately 60% of the replay buffer, remains insensitive to $\pm 20\%$ perturbations of all reward weights, and degrades gracefully under measurement noise while the current mask enforces the peak-current constraint throughout. A deployment analysis shows that the 18,432 multiply–accumulate policy meets a $50\mathsf{\mu }$s control period in its existing form at a measured cost of about $2\%$ in torque ripple. All the results are simulation-based on a finite-element model parameterized from a physical machine. Full article

Defibrillators require high-performance power electronic converters capable of rapidly charging a high-voltage capacitor and delivering controlled therapeutic waveforms while ensuring patient safety. This paper presents a predictive control strategy for the flyback converter of a defibrillator, including both its charging and discharging stages, together with the design and verification workflow for its implementation. The proposed system integrates the FPGA implementation of the charging-stage control with a hardware-in-the-loop (HIL) emulation of the flyback converter and discharge stage within a unified model-based design framework. The charging stage consists of a flyback converter regulated by a Finite Control Set Model Predictive Control (FCS-MPC) strategy, while the discharge stage employs a full-bridge converter to generate truncated exponential biphasic (BTE) waveforms. To regulate the switching frequency without sacrificing the fast dynamic response of predictive control, the Period Control Approach (PCA) is incorporated into the FCS-MPC. The proposed solution is benchmarked against conventional FCS-MPC and a hysteresis controller, highlighting the advantages of PCA-based predictive control in terms of switching-frequency regulation while preserving accurate current tracking. The proposed control system and the corresponding defibrillator model are developed in MATLAB/Simulink and automatically translated into synthesizable VHDL using HDL Coder. This approach enables FPGA implementation of the control strategy and HIL emulation of the power converters without manual HDL programming. The proposed methodology covers the entire workflow, from simulation to real-time FPGA implementation and HIL emulation. Simulation results demonstrate accurate current tracking, proper BTE waveform generation, and improved switching-frequency regulation compared with both conventional FCS-MPC and hysteresis-based control. HIL experiments on a Xilinx Artix-7 FPGA confirm the real-time operation of the implemented predictive controller interacting with the emulated flyback converter. The experimental results are consistent with the simulation results. This work provides a solid foundation for the development and validation of digitally controlled defibrillators based on advanced predictive control techniques. The results demonstrate the feasibility of the proposed approach in both simulation and reconfigurable hardware. Full article

Wind direction time series exhibit angular periodic discontinuity, multi-scale non-stationary fluctuations and abrupt wind shifts, which hinder the precision of short-term forecasting for wind turbine yaw control. In this paper, a dual-branch forecasting framework based on Variational Mode Decomposition (VMD) and Transformer is developed to address the above drawbacks, with a hysteresis gating and zoning residual compensation module embedded for targeted error correction. First, sine–cosine encoding is adopted to eliminate the numerical discontinuity between 0° and 360° for wind direction angular data, and valid meteorological input features are screened to discard redundant covariates. Second, the sine–cosine-encoded wind direction sequence is decomposed into multiple band-limited intrinsic mode functions (IMFs) via VMD, extracting frequency-specific features that reduce non-stationarity and facilitate subsequent Transformer modeling. The standard Transformer encoder serves as the normal branch to capture long-range temporal dependencies across the whole time series, while a lightweight multilayer perceptron (MLP) constitutes the compensation branch to learn prediction deviations between baseline predictions and ground-truth values. The hysteresis gating unit activates residual compensation based on historical prediction errors and angular variation, without requiring access to the current ground-truth value, and compensation intensity is adaptively adjusted via the zoning strategy; relevant coefficients are optimized by random search. Verified on a real wind farm dataset consisting of 10,421 15 min sampling points, the proposed model achieves the lowest MAE of 9.64° among six benchmark models. For the improved genuine mutation samples (angle change > 70°), the model achieves a mean improvement of 6.02°. Ablation experiments verify that VMD preprocessing, the MLP compensation branch, and the hysteresis gating mechanism play indispensable roles in forecasting performance. The proposed framework can support accurate yaw control of wind turbines, and the decomposition–compensation workflow can also be generalized to other periodic non-stationary forecasting tasks. Full article

Memtranstors are emerging higher-order memory elements that establish a state-dependent constitutive relationship between electric charge and magnetic flux, making them attractive for adaptive analog electronics, neuromorphic computing, nonlinear dynamical systems, and memory-enabled signal processing applications. However, existing memtranstor emulators predominantly rely on operational amplifiers, analog multipliers, current conveyors, or behavioral models, leading to increased circuit complexity and limited suitability for monolithic CMOS integration. This paper presents a unified transistor-level dynamic-threshold MOS (DTMOS) framework for realizing both charge-controlled and flux-controlled memtranstor emulators. The proposed architectures synthesize direct and inverse memtranstances through capacitive state integration, state-dependent DTMOS conductance modulation, and current-domain affine processing, thereby eliminating the need for composite active building blocks. Closed-form analytical expressions are derived for both constitutive relations and explicitly related to transistor-level parameters, bias conditions, and state-storage elements. The theoretical framework is further supported by comprehensive analyses of channel-length modulation, finite output resistance, device mismatch, DTMOS body-effect deviations, leakage mechanisms, pseudo-resistor non-idealities, parasitic capacitances, and small-signal stability. Cadence Virtuoso simulations performed in a 180 nm triple-well CMOS technology validate the analytical predictions and demonstrate the characteristic butterfly shaped pinched hysteresis loops of both emulators. The proposed circuits operate from a single 0.8 V supply while dissipating approximately 22 μW and 36 μW for the charge-controlled and flux-controlled realizations, respectively, and exhibit electronic tunability, together with robustness against process and temperature variations. Representative implementations of reconfigurable frequency-selective circuits and a memtranstor-based envelope detector further demonstrate the practical applicability of the proposed architectures. To the best of the author’s knowledge, this work presents the first unified transistor-level DTMOS constitutive synthesis framework for realizing both direct and inverse memtranstive behavior, providing a scalable foundation for future adaptive mixed-signal integrated circuits, programmable analog memory systems, neuromorphic hardware, and in-memory computing platforms. Full article

This study investigates physics-informed and data-driven hybrid modeling strategies for an automotive-grade electrohydraulic (EH) semi-active damper system. Although deep sequence learning architectures such as Long Short-Term Memory (LSTM) networks and Transformers can provide high predictive accuracy, purely data-driven approaches may struggle to preserve physical consistency and maintain robustness under unseen operating conditions. These limitations become more pronounced for EH dampers, whose hysteretic characteristics exhibit highly nonlinear and non-proportional variations under different current and frequency excitations, unlike the more scalable behavior commonly observed in magnetorheological (MR) dampers. To address these challenges, two physics-informed integration strategies are investigated. The first strategy combines physical and data-driven models through parallel loss-function synthesis. The second strategy introduces a learnable physics layer (PINN-Hybrid), in which the coefficients of the extended hyperbolic tangent formulation are adaptively learned within the neural network architecture. In this framework, the physical model acts as a structural regularization mechanism that guides the learning process while preserving the flexibility of data-driven sequence modeling. The proposed models are evaluated under abrupt valve-control operating conditions. Comparative results indicate that the proposed physics-informed architectures improve hysteresis continuity, physical plausibility, and robustness compared with purely data-driven approaches, particularly in low-velocity and transition regions. The proposed framework therefore demonstrates the potential of physics-informed learning strategies for reliable real-time modeling of nonlinear automotive EH damper systems. Full article

On-site hydrogen refueling stations (HRS) face significant operational challenges due to the stochastic nature of hydrogen demand, creating a severe supply–demand mismatch. Under traditional pressure-based hysteresis control, this volatility forces Proton Exchange Membrane (PEM) electrolyzers into frequent start–stop cycles, accelerating degradation and reducing efficiency. In response, this study introduces an automated control framework integrating macroscopic gas-state modeling with deep-learning-based demand prediction. First, a real-gas thermodynamic model was established. Monte Carlo simulations of 100 random filling scenarios identified a robust design benchmark of 4.5 kg per vehicle. A low filling stability coefficient (5.02%) confirmed that individual thermodynamic fluctuations are negligible, validating a traffic-flow-driven demand approach. Next, a deep Long Short-Term Memory (LSTM) network was developed to forecast short-term demand. Trained on an 8784 h dataset exhibiting “double-peak” traffic patterns, the model achieved high precision on the unseen test set, yielding a Root Mean Square Error (RMSE) of 6.75 kg and a normalized RMSE (nRMSE) of 0.0987, explaining 82% of the demand variance. Finally, an LSTM-informed demand-following control strategy was formulated to enable proactive, thermally bounded operation alongside a novel “Hot Standby” mechanism. Maintaining a minimal 3.0 kg/h holding current during idle periods sustains stack temperatures above 60 °C, effectively mitigating thermal stress. Comparative simulations over 1464 h demonstrated that the proposed framework reduces detrimental cold start–stop cycles by 98.4% (from 61 to 1) and suppresses power output fluctuations by 40.7% compared to the traditional baseline. These results confirm that data-driven control significantly enhances operational stability, facilitates grid integration, and extends core equipment service life. Full article

Accurate modeling of smart composite dampers is crucial for simulation and model-based control. This study focuses on the constitutive modeling of a novel damper that synergistically combines an Entangled Metallic Wire Mesh (EMWM) with a magnetorheological (MR) fluid. Unlike traditional MR dampers, the interaction between the field-responsive MR fluid and the rate-sensitive, deformable EMWM matrix introduces strong coupled current–frequency dependence. To capture this essential characteristic, a control-oriented, bivariate (current–frequency) hysteresis model is formulated, wherein all parameters are explicit, continuous functions of both the control current (I) and excitation frequency (f). A systematic two-step identification method is employed to derive these functions from dynamic tests. A key finding is that the identified damping exponent (α) consistently exceeds unity across the tested operational range. This quantitatively indicates a transition from viscous-dominated to inertial-flow-dominated dissipation within the EMWM matrix, a distinctive mechanism attributed to non-Darcian flow in its porous structure. The fully parameterized model demonstrates high fidelity (R² > 0.99) within the characterized low-frequency, small-amplitude regime and shows reliable predictive capability for interpolated conditions. The presented model serves as a ready-to-use constitutive tool for the simulation and design of low-frequency vibration isolation systems utilizing EMWM-MR composites, and the revealed inertial flow mechanism provides fundamental insight for the development of next-generation adaptive dampers. Full article

Module- and pack-level mechanical design of lithium-ion batteries in electric vehicles is a primary driver of swelling-induced stack pressure and spatially varying ageing. Current practice remains largely empirical or data-driven and configuration-specific, limiting the ability to predict how design changes translate into local pressure heterogeneity and state-of-health (SOH) loss. This motivates a compact chemo-mechanical model that maps packaging boundary conditions to pressure, swelling, and SOH evolution with few interpretable parameters. This study introduces finite-element-ready constitutive laws that couple reversible and irreversible swelling to SOH and through-thickness pressure, covering three boundary cases reported in literature: constant pressure, thickness clamp after an initial preload, and flexible support. Parameters are identified from different published datasets, and the model is validated against independent constraint scenarios. Good quantitative agreement is shown with averaged RMSE of 1.16% for SOH and 0.16 [MPa] for pressure evolution. Variance-based sensitivity analysis shows SOH uncertainty dominated by the damage-law parameters of the proposed constitutive relationship, whereas pressure evolution is primarily controlled by irreversible swelling and the non-linear through-thickness stiffness, indicating calibration priorities for engineering design studies. The framework is intended for fast comparative analyses of individual cells under a controlled environment. Further extensions, including SOC-dependent mechanics, refined hysteresis, temperature, and C-rate variations require dedicated datasets and are left for future work. Full article

Conventional finite control set model predictive control (FCS-MPC) for permanent magnet synchronous motor (PMSM) drives suffers from a fundamental trade-off: shortening the control period improves current tracking but increases switching frequency and losses. This paper proposes a hysteresis-based variable control period MPC (HBVCP-MPC) to break this compromise. Unlike methods like direct torque control (DTC) and model predictive direct torque control (MPDTC) that use hysteresis to select voltage vectors (VV), our approach first selects the optimal VV via a cost function that balances current tracking accuracy and switching frequency. Hysteresis on the $dq$-axis currents is then employed solely to dynamically determine the application time of this pre-selected VV, which defines the variable control period. This grants continuous adjustment over the VV duration, enabling superior current tracking without a proportional rise in switching frequency. Experimental results confirm that the proposed method achieves enhanced steady-state performance at a comparable switching frequency. Full article

Steel slag aggregate (SSA), as a high-performance and sustainable material, has demonstrated significant potential in enhancing the mechanical properties of concrete and improving the bond behavior between reinforcement and the concrete matrix, thereby contributing to the seismic resilience of steel slag concrete columns (SSCCs). Nevertheless, the underlying mechanism through which SSA influences the seismic performance of SSCCs remains insufficiently understood, and current analytical models fail to accurately capture the effects of bond strength on structural behavior. In this study, a comprehensive experimental program comprising central pull-out tests and quasi-static cyclic loading tests was conducted to investigate the influence of SSA on bond strength and the seismic response of SSCCs. Key seismic performance indicators, including the hysteresis curve, equivalent viscous damping ratio, and ductility coefficient, were evaluated. The role of bond strength in governing energy dissipation and ductility characteristics was elucidated in detail. The results indicate that bond strength significantly affects the seismic performance of SSCC components. At an SSA replacement ratio of 40%, the specimens show optimal performance: energy dissipation capacity increases by 11.3%, bond–slip deformation in the plastic hinge region decreases by 10%, and flexural deformation capacity improves by 9% compared to the control group. However, when the SSA replacement exceeds 60%, the performance metrics are similar to those of ordinary concrete, showing no significant advantages. Based on the experimental findings, a modified bond–slip constitutive model for the steel slag concrete–reinforcement interface is proposed. Furthermore, a finite element model incorporating bond–slip effects is developed, and its numerical predictions exhibit strong agreement with the experimental results, effectively capturing the lateral load-carrying capacity and stiffness degradation behavior of SSCCs. Full article

Active magnetic bearings (AMBs), pivotal in high-speed rotating machinery for their frictionless operation and precise control, demand power amplifiers with exceptional dynamic performance and minimal current ripple. However, conventional amplifier designs often overlook eddy current effects, a critical oversight given the high-frequency switching inherent to pulse-width modulation (PWM). These induced eddy currents distort output waveforms, amplify ripple, and degrade system bandwidth. This paper bridges this critical gap by proposing a comprehensive methodology to model, quantify, and mitigate eddy current impacts on three-level half-bridge power amplifiers. A novel mutual inductance-embedded circuit model was developed, integrating winding–eddy current interactions under PWM operations, while a discretized transfer function framework dissects frequency-dependent ripple amplification and phase hysteresis. A voltage selection criterion was analytically derived to suppress nonlinear distortions, ensuring stable operation in high-precision applications. A Simulink simulation model was established to verify the accuracy of the theoretical model. Experimental validation demonstrated a 212% surge in steady-state ripple (48 mA to 150 mA at 4 A DC bias) under a 20 kHz PWM operation, aligning with theoretical predictions. Dynamic load tests (400 Hz) showed a 6.28% current amplitude reduction at 80 V DC bus voltage compared to 40 V, highlighting bandwidth degradation. This research provides a paradigm for optimizing AMB power electronics, enhancing precision in next-generation high-speed systems. Full article

The sedimentation process of coal slurry water is influenced by numerous factors and has complex mechanisms. Its nonlinear and large hysteresis characteristics pose great challenges to process optimization control, making it a current research hotspot. This paper takes the typical slime water treatment process of a coal preparation plant as the object, and, on the basis of selecting raw coal quantity, flocculation dosage, coagulation dosage, overflow turbidity, raw coal ash content, underflow concentration, and slime quantity as the key variables, establishes a quality control method for process detection data consisting of data acquisition → data anomaly detection → data filling and noise reduction; subsequently, different machine-learning algorithms are used to predict the performance of coal-slurry-settling agents. It was found that Long Short-Term Memory shows the highest prediction accuracy for coagulants, with corresponding root mean square errors of 2.72% and 6.23%. Finally, using iFix software (version 5.5), an intelligent control system for the settling process of coal slurry water was constructed, which reduced the usage of coagulants by 31.56% and 37.21%. Full article

In recent years, the application of machine learning methods has become increasingly common in atmospheric science, particularly in modeling and predicting processes that impact air quality. This study focuses on predicting hydrogen production from solid oxide electrolytic cells (SOECs), a technology with significant potential for reducing greenhouse gas emissions and improving air quality. We developed two models using artificial neural networks (ANNs) and support vector machine (SVM) to predict hydrogen production. The input variables are current, voltage, communication delay time, and real-time measured hydrogen production, while the output variable is hydrogen production at the next sampling time. Both models address the critical issue of production hysteresis. Using 50 h of SOEC system data, we evaluated the effectiveness of the ANN and SVM methods, incorporating hydrogen production time as an input variable. The results show that the ANN model is superior to the SVM model in terms of hydrogen production prediction performance. Specifically, the ANN model shows strong predictive performance at a communication delay time ε = 0.01–0.02 h, with RMSE = 2.59 × 10⁻², MAPE = 33.34 × 10⁻²%, MAE = 1.70 × 10⁻² Nm³/h, and R² = 99.76 × 10⁻². At delay time ε = 0.03 h, the ANN model yields RMSE = 2.74 × 10⁻² Nm³/h, MAPE = 34.43 × 10⁻²%, MAE = 1.73 × 10⁻² Nm³/h, and R² = 99.73 × 10⁻². Using the SVM model, the prediction error values at delay time ε = 0.01–0.02 h are RMSE = 2.70 × 10⁻² Nm³/h, MAPE = 44.01 × 10⁻²%, MAE = 2.24 × 10⁻² Nm³/h, and R² = 99.74 × 10⁻², while at delay time ε = 0.03 h they become RMSE = 2.67 × 10⁻² Nm³/h, MAPE = 43.44 × 10⁻²%, MAE = 2.11 × 10⁻² Nm³/h, and R² = 99.75 × 10⁻². With this precision, the ANN model for SOEC hydrogen production prediction has positive implications for air pollution control strategies and the development of cleaner energy technologies, contributing to overall improvements in air quality and the reduction of atmospheric pollutants. Full article

Efficient control of tunnel boring machine (TBM) tunneling along the designed tunnel axis in an unknown variable geological environment is a difficult and significant task. At present, the TBM attitude during tunneling is mostly manually controlled based on the deviation between the tunneling axis and the designed tunnel axis and their experiences. The tunneling axis from manual control is often the snakelike motion around the designed tunnel axis, even exceeding the deviation limit, for which this paper analyzed three reasons, the unknown geological environment, the hysteresis of TBM position response, and the unsolved overall optimization of tunneling axis. For these reasons, this paper proposed a real-time optimal control framework of TBM attitude based on reinforcement learning, which contains the geological information predictive model, TBM attitude and position (TBMAP) predictive model, and optimal attitude control policy (OACP). This framework can predict the current geological information in real-time and provide the corresponding real-time optimal attitude control that simultaneously considers the hysteresis of TBM position response and the overall optimization of the tunneling axis. This attitude control framework can be directly deployed to TBM without increasing costs and excessive modifications to the equipment. To verify the effectiveness of this attitude control framework, the Xinjiang Yiner Water Supply Phase II Project, using the TBM method, was adopted as a case study. The results revealed that the accuracy of geological environment recognition reached 94%, and OACP can significantly reduce the accumulated deviation of the tunneling axis from the designed tunnel axis by over 80% compared with the manual control and easily provide real-time decision support for attitude control in actual engineering. Full article

The integrated design of the motor and axial piston pump eliminates the coupling structure, resulting in a compact and lightweight motor-pump structure. The challenge of motors overheating has always been a major concern. To address this issue, the hydraulic oil throughout the motor pump is utilized for cooling the high-speed motor, effectively improving the power density and heat dissipation capability of the hydraulic power unit. This integrated design approach has successfully resolved the significant issue of overheating motors, leading to enhanced performance of the hydraulic power unit. To address this concern, the entire motor pump’s oil is utilized to cool the high-speed motor. Consequently, the thermodynamic prediction of high-speed motor pumps has become increasingly important. In this study, the impact of motor heat generation on hydrodynamics is analyzed, and the heat transfer of the motor pump is investigated using the control volume method. Furthermore, thermodynamic models of hysteresis loss, eddy current loss, alternating current loss, churning loss, and throttling loss are established for the oil-immersed motor pump. The change in oil viscosity is also considered. The instantaneous temperature change rule of the oil within the oil-immersed motor pump is derived. Additionally, the influence of various working conditions such as pressure and speed on the temperature of the motor pump’s key node is examined. The experimental results indicate the accuracy of the thermodynamic calculation, and the significant effect of motor loss on the leakage temperature. Full article