To address the issue of uncertainty in the occurrence time of voltage sags in power grids, which affects power quality, a voltage state prediction method based on LSTM neural networks is proposed for predicting voltage states. For the problem of quickly and accurately compensating for voltage sags, a DVR system based on a new approach law of sliding mode variable structure control is proposed, which significantly reduces chattering, improves response speed, and enhances the robustness of the system. The stability of the system is proven based on Lyapunov stability theory. Simulation experiments are conducted to analyze the voltage state prediction effect based on the LSTM neural network and the compensation effect of the novel reaching law of sliding mode variable structure control under different levels of voltage sag, validating the effectiveness and correctness of the proposed solution.

Permanent Magnet Synchronous Motors (PMSMs) are nonlinear, multi-parameter systems that exhibit structural symmetry but are susceptible to parameter variations and external disturbances. These challenges can disrupt the inherent symmetrical characteristics of PMSM dynamics during real-world operations, posing difficulties for achieving efficient control. To address this issue, this paper proposes a Model-Free Generalized Super-Twisting Algorithm Fast Terminal Sliding Mode Control (MFFTSMC-GSTA) method. First, a novel ultra-local model incorporating PMSM uncertainties is established, and the MFFTSMC-GSTA controller is designed to address the system’s complex dynamic behavior. By integrating the generalized super-twisting algorithm with the nonsingular fast terminal sliding mode algorithm, the proposed controller ensures finite-time convergence and effectively mitigates chattering. Second, an extended sliding mode disturbance observer is developed to estimate the unknown components of the ultra-local model and provide feedforward compensation, further enhancing system robustness and dynamic performance. The experimental results show that the total harmonic distortion (THD) value of the proposed control method is 1.38%, demonstrating significant improvements in response speed and robustness for motor speed control, and verifying the algorithm’s superior performance under complex operating conditions.

To address the deteriorated control performance of permanent magnet synchronous motor (PMSM) drive systems for new energy vehicles (NEVs) under complex conditions caused by multi-source disturbances (internal parameter perturbations and external load mutations), this paper proposes an improved terminal integral sliding mode control (ITISMC-ADERL) strategy integrating a piecewise adaptive terminal integral sliding mode surface and an ADERL. The proposed sliding mode surface adopts interval-adaptive switching between high- and low-order power terms, completely eliminating singularity and integral saturation defects of traditional terminal sliding mode control while ensuring fast convergence, and achieving an optimal structural balance between convergence speed and chattering suppression. The state-dependent ADERL leverages the synergy of error-sliding variable coupled dynamic gain adjustment and variable exponential power compensation, realizing dual-mode adaptive switching of “strong driving for fast approaching far from the sliding surface, weak gain for smooth regulation near the sliding surface”, which significantly improves control accuracy and anti-disturbance robustness. The finite-time convergence of the closed-loop system is rigorously proved via Lyapunov stability theory. Full-operating-condition comparative tests on a TMS320F28379D DSP platform show that the proposed strategy outperforms SMC-ERL, ISMC-ERL and ITISMC-ERL in all test scenarios (no-load startup, acceleration/deceleration, sudden load changes, flux linkage perturbation), meeting the requirements of high-performance NEV drive systems and possessing important engineering application potential.

Energy-intensive enterprises lack a scientific and effective energy efficiency assessment framework and methodology. This lack leads to an inaccurate understanding of energy usage and its benefits. As a result, there is energy wastage and loss. This wastage and loss negatively affect product costs. They also present a challenge to effective energy management. To address these issues, this paper introduces a novel, comprehensive energy efficiency evaluation system. This system integrates both qualitative and quantitative measures. It proposes an evaluation model based on the Particle Swarm Optimization (PSO) combined with the Analytic Hierarchy Process (AHP) and Fuzzy Comprehensive Evaluation (FCE), wherein PSO is employed to optimize the weights determined by AHP, ensuring that the significance attributed to various indicators is scientific, objective, and rational. The FCE method is utilized to convert diverse factors affecting corporate energy efficiency, across different types and scales, into standardized 0–1 values, enabling a comparative analysis of the impact of each process and indicator on energy efficiency. Furthermore, the paper introduces an energy efficiency prediction model employing a multivariate linear regression algorithm, which demonstrates a good fit, facilitating the transition from retrospective energy efficiency evaluation to proactive improvements. Utilizing data on actual consumption of water, electricity, and steam from an enterprise, along with expert assessments on the implementation levels of new processes, technologies, equipment, personnel scheduling proficiency, steam recovery rates, and adherence to policies and assessments, a simulation experiment of the proposed model was conducted using Python. The evaluation yielded an energy efficiency score of 0.68; this is consistent with the real-world scenario of the studied enterprise. The predicted mean square error of 9.035416039503998 $\times {10}^{-9}$ indicates a high model accuracy, validating the practical applicability and effectiveness of the proposed approach.

In order to enhance the accuracy of ventilator parameter setting, this paper analyzes two identification methods for respiratory characteristic parameters of non-invasive ventilators and invasive ventilators. For non-invasive ventilators, a respiratory characteristic parameter identification method based on a respiration model is established. In this method, the patient’s respiratory sample set is obtained through non-invasive measurements. Experimental results demonstrate that the mean relative error of pulmonary elastance identification was 14.25%, and the mean relative error of intrapulmonary pressure identification was 12.33% using the Romberg integral algorithm. For chronic patients using non-invasive ventilators, the fault-tolerant space for ventilator parameter setting is large; this method meets the requirement of auxiliary setting of non-invasive ventilator parameters. For invasive ventilators, a respiratory characteristic parameter identification method based on the AVOV–BP neural network is established. In this method, the patient’s respiratory sample set is obtained through real-time invasive measurements. Even with small sample datasets, experimental results show that the mean relative error of pulmonary elastance identification and intrapulmonary pressure identification were both 0.22%. For critically ill patients using invasive ventilators, the fault-tolerant space for ventilator parameter setting is small; this method meets the requirement of auxiliary setting of invasive ventilator parameters.