Field-condition-aware cleaning-loss intensity estimation in rice combine harvesters using a lightweight 1D CNN_GRU virtual sensor with multi-sensor fusion

Cleaning-loss strongly affects rice combine harvesting efficiency, yet piezoelectric monitoring can become unreliable when the machine transitions between wet, dry, and combined paddy zones. This study developed a field-condition-aware lightweight 1D CNN_GRU virtual sensor to estimate voltage-based cleaning-loss intensity using machine-dynamics measurements. A multi-sensor platform was deployed on a ZOOMLION rice combine harvester, integrating dual IMUs (sieve and chassis), fan/rotor proximity RPM sensors, field-measured agronomic variables, and RTK-GNSS, while piezoelectric thin-film array and ceramic sensors provided training targets. Raw piezoelectric signals were sampled at 1000 Hz and down sampled to a synchronized 1 Hz logging rate for multi-sensor fusion. Field data were collected across wet, dry, and combined field conditions (15,000 time-aligned samples retained for each condition). Indoor grain-dropping calibration test confirmed the quantitative relationship between known grain impact mass and piezoelectric voltage response. The proposed 1D CNN_GRU model (∼73 k parameters, ≈ 0.28 MB) was benchmarked against machine-learning (ML) baselines (Ridge, SVR, Random Forest, XGBoost) and deep-learning (DL) baselines (1D_CNN, GRU, LSTM, CNN_LSTM, CNN_LSTM_GRU). The proposed model outperformed traditional ML and DL baselines, achieving (Coefficient of Determination (R 2 ) / Root Mean Squared Error (RMSE) values of 0.87/0.238 V and 0.90/0.211 V; 0.96/0.013 V and 0.97/0.012 V; 0.98/0.126 V and 0.98/0.105 V) under wet, dry and combined field conditions for the array and ceramic sensors, respectively. Under wet conditions, the virtual sensor outperformed the best ML baseline XGBoost (ΔR 2 = +0.25 for array and ΔR 2 = +0.23 for ceramic). Sensor ablation and SHAP analysis identified field condition dependent dominant predictors, while MC Dropout provided uncertainty aware prediction intervals. These findings demonstrate that the proposed virtual sensor provides an accurate and redundant cleaning-loss intensity estimation channel that remains operational when direct piezo measurements become unreliable, supporting embedded-oriented monitoring for smart rice harvesting.