2026-10-10T14:46:49Z https://data.mendeley.com/oai

oai:data.mendeley.com/svc3tyct2t.1 2026-09-24T23:26:03Z

Huang, Guokai Experimental data and Simulink models for knowledge-guided DNN-based PMSM harmonic compensation Mendeley Data This dataset provides the experimental data, simulation models, and source codes associated with the study on a knowledge-guided deep neural network (KG-DNN) with an enhanced adaptive notch filter (EANF) for harmonic compensation in permanent magnet synchronous motor (PMSM) drives. Experimental data were collected on a 1.5 kW surface-mounted PMSM testbench controlled by a dSPACE DS1202 real-time system. The control algorithm was executed at a 10 kHz interrupt rate, and signals were recorded at 8192 Hz. The dataset includes two CSV files. experimental_sample.csv contains time-series measurements for six control configurations (A–F), with columns for timestamp (s), reference speed (rpm), measured speed (rpm), d/q-axis currents (A), d/q-axis voltages (V), electromagnetic torque (N·m), average torque error (N·m), total harmonic distortion (%), and configuration label. sensitivity_scan.csv reports the parameter sensitivity scan results, including parameter name, value, average torque error, THD, convergence epochs, and stability status. Reproduction instructions are provided in the Supplementary Material of the associated paper. Engineering Automation Engineering Embedded Control Systems Design Dataset doi:10.17632/svc3tyct2t.1 oai:data.mendeley.com/svc3tyct2t.1 Creative Commons Attribution 4.0 International http://creativecommons.org/licenses/by/4.0 https://data.mendeley.com/datasets/svc3tyct2t 2026-09-24T23:26:03Z