Search

In this paper, a scalable end-to-end tree boosting system called XGBoost has been applied for predicting the magnitude of an earthquake from the early part of earthquake waveform data. This model uses the features extracted from the early P wave phase of the records as an input. The model's effectiveness has been verified by using data on earthquakes occurring in the Eurasian plate of Japan Islands from 1996 to 2021. Feature engineering has given 29 new features identified from the early P wave phase of the record, which show a high correlation with the magnitude of an earthquake. The comparison of predicted and actual magnitude shows that a trained XGboost model, which uses a single input record for magnitude prediction, gives an average prediction error of 0.004 ± 0.57 for earthquakes in the test dataset. In contrast, the average prediction error of –1.1 ± 0.80 and –0.65 ± 0.69 has been obtained for the magnitude estimated from conventional τc and Pd methods using the same test dataset. It is further seen that the average predicted magnitude of a single earthquake of magnitude 4.5 and 6.1 (MJMA) obtained by using multiple nearfield records using XGBoost model is 4.58 ± 0.33 and 6.32 ± 0.29, which is close to the actual magnitude of the earthquake. The results presented in this paper clearly show that the structured data can be effectively used by complex machine learning or deep learning models to predict earthquake magnitude from single or multiple records. [ABSTRACT FROM AUTHOR]

The cascaded H‐bridge (CHB) converter can achieve rich functions, such as active rectification, active power filtering, distributed energy storage, etc. The control of the CHB converter is a multiobjective optimization problem. Finite control set model predictive control (FCS‐MPC) is an effective method developed in recent years to solve this problem. However, because of the large amount of calculation, traditional FCS‐MPC cannot be directly used to control the multilevel CHB converter. It is also difficult to tune the weighting factors in traditional FCS‐MPC. To solve these problems, this paper proposes an improved sequential MPC (IS‐MPC) method for the CHB converter. The proposed method divides the control objectives of the CHB converter into different layers according to their priorities and the control degrees of freedom of each layer can be adjusted according to the system state. Simulation and hardware‐in‐the‐loop (HIL) implementation on a three‐phase 7‐level CHB converter shows that the proposed method can effectively control the ac current and submodule voltage of the CHB converter and reduce the switching loss. Moreover, it significantly reduced the calculation burden of traditional FCS‐MPC, and the use of the weighting factors is also avoided. © 2024 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

A nonlinear multivariable highly coupled system that is susceptible to both internal and external perturbations is the permanent magnet synchronous motor. A new reaching law‐based control approach that combines a sliding‐mode disturbance observer and an enhanced two‐vector model predictive current control is suggested in order to enhance the dynamic performance of the permanent magnet synchronous motor and its robustness to system disturbances. First, a new reaching law is developed in this paper to resolve the conflict between the traditional reaching law's chattering of the system and the reaching time of the sliding‐mode surface. This new reaching law not only makes the system's response more rapid but also reduces chattering; second, taking into account the possibility of systems with internal parameter uptake and external load disturbance, a sliding‐mode disturbance observer based on t is developed. Finally, the q‐axis current obtained from the speed sliding‐mode control rate is added to the model predictive current control, and an optimal duty cycle model predictive current control based on space vector modulation is designed because the traditional two‐vector model predictive current control is to optimize the duty cycle and voltage vectors separately, and the switching frequency is not fixed. The method suggested in this research may successfully suppress the sliding‐mode control system chattering and increase the resilience and dynamic response performance of the speed control system, according to simulation findings compared to the traditional reaching law and traditional duty cycle. © 2024 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Aiming at the current control steady‐state error caused by sampling delay, parameters mismatch of direct‐drive permanent magnet synchronous motor (PMSM). This paper proposed a method of dual model predictive control (MPC) of direct‐drive PMSM based on sliding mode disturbance observer (SMDO). Deadbeat predictive current control (DPCC) is used for current loop, a SMDO is proposed to observe the system disturbance, the estimation is compensated to DPCC by feedback to decrease the current control steady‐state error. For speed loop, MPC with an internal model control observer (IMCO) can observe and compensate the load disturbance feedforward. The experimental results show that, the dynamic response speed of the proposed strategy is fast, the speed recovery time after sudden load torque is short, and the anti‐load disturbance performance and current tracking performance are effectively improved. © 2024 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Compared to the model predictive control (MPC) technique, the modulated MPC (M2PC) method applies multiple voltage vectors in order within one switching cycle, enabling a fixed switching frequency and better motor control performance. However, the M2PC technique suffers from a significant computational burden and imprecise allocation method of the vector action time, affecting the control performance. To address these two issues, this paper proposes an improved M2PC method that utilizes the information of projection of the reference vector on the selected voltage vectors for efficiently selecting the optimal active vectors and also accurately calculating their action time. Compared to the conventional M2PC, the proposed technique demonstrates both excellent dynamic performance with reduced execution time by 15.3% and superior steady‐state performance in terms of lower torque/speed ripple as well as fewer current harmonics. Those benefits have been proven by the results of a 2.5k rpm permanent magnet synchronous motor drive. © 2024 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Ground motion prediction is an important and complex research subject in earthquake engineering and traditional approaches based on statistical regression have much room for improvement in prediction accuracy. Utilizing the 67,164 ground motion records from KiK-net and K-Net for 777 shallow crustal earthquakes between 1997 and 2019 in Japan, this paper proposes a ground motion prediction model XGBoost-SC based on the machine learning algorithm of eXtreme Gradient Boosting (XGBoost) for Japan. Magnitude, focal depth, hypo-central distance, V s30 , site altitude, and focal mechanism were used as the feature parameters, and XGBoost, Random Forest, and Deep Neural Networks (DNN) algorithms were selected for model training while Bayesian optimization was used to search for optimized hyperparameters to improve the prediction accuracy. XGBoost algorithm was selected for further study based on the comparison of results from the three algorithms. Residual change with magnitude and hypo-central distance, the probability distribution of residuals, residual standard deviation (σ), residual mean squared error (MSE), and Pearson correlation coefficient (R) were used as the evaluation parameters, and a comparison study was performed against the ground motion prediction equation based on traditional approaches. Actual earthquake events were selected to compare the prediction results against the observation records. To further validate and explain the proposed model, SHapley Additive exPlanations (SHAP) analysis was performed to explain the impact of selected feature parameters on the proposed model. The results demonstrate that the proposed XGBoost-SC model has good prediction stability for all periods, and its residual errors are smaller than those of other models. Therefore, the proposed model can better reflect the ground motion attenuation for shallow crustal earthquakes in Japan and can serve as a better model for ground motion prediction in future aseismic design and earthquake disaster mitigation efforts. • A Ground Motion Prediction Model was proposed for shallow crustal earthquakes. • The proposed model was more precise than traditional approaches. • Explained the effect of feature parameters on the prediction results. [ABSTRACT FROM AUTHOR]