NCKH – Pilot Scale Deep Learning
LSTM (Long Short-Term Memory) Forecast Dashboard
Deep learning model trained on historical time series data for accurate parameter predictions
Last updated:
Download CSV LSTM LSTM Hybrid PINN RL
Weather Data Source – Web API (Open-Meteo)
Air Temperature31,03°C
Air Humidity71,65%
Wind Speed13,82 km/h
Rainfall0,26 mm
ConditionsSunny intervals
Auto API
Weather Data Source – Field Sensor (Sensor)
Air Temperature31,03 °C
Air Humidity71,65 %
Wind Speed13,82 km/h
Rainfall0,26 mm
ConditionsSunny intervals
ESP32 + BME280 + BH1750 + Rain gauge
Water Temperature (°C)
28,36
PH Parameter
7,98
DO (mg/L)
6,73
Turbidity (NTU)
4,07
Air Temperature (°C)
31,03
Air Humidity (%)
71,65
Wind Speed (km/h)
13,82
Rainfall (mm)
0,26
Forecast Window:
━━ Observed   ╌╌ LSTM (Long Short-Term Memory)
Water Temperature (°C)
PH Parameter
Dissolved Oxygen (mg/L)
Turbidity (NTU)
Salinity (ppt)
TDS (mg/L)
ORP (mV)
Water Temperature (°C)
25,1
PH Parameter
7,52
DO (mg/L)
7,51
Turbidity (NTU)
2,56
Air Temperature (°C)
31,03
Air Humidity (%)
71,65
Wind Speed (km/h)
13,82
Rainfall (mm)
0,26
Forecast Window:
━━ Observed   ╌╌ LSTM (Long Short-Term Memory)
Water Temperature (°C)
PH Parameter
Dissolved Oxygen (mg/L)
Turbidity (NTU)
Salinity (ppt)
TDS (mg/L)
ORP (mV)
MAE
0.142
RMSE
0.198
R²
0.943
MAPE (%)
2.35
Training Loss vs Validation Loss MSE per epoch
MAE Convergence by Epoch
R² Score theo Epoch
Predicted vs Actual – Scatter (Temperature °C)
Model Architecture – LSTM (Long Short-Term Memory)
LSTM Network Architecture
  • Input: 48-step time series (sensor)
  • LSTM Layer 1: 128 units, dropout=0.2
  • LSTM Layer 2: 64 units, dropout=0.2
  • Dense: 32 → 1 (output)
  • Optimizer: Adam (lr=1e-3, decay ×0.9/10 ep)
  • Loss: MSE | Batch: 32 | Epochs: 200
Inputs from Observed
  • Water temperature (DS18B20)
  • pH (glass electrode)
  • DO (Clark cell)
  • Turbidity (optical OD)
  • Air temperature / humidity / pressure (BME280)
  • Light Intensity (BH1750)
  • Rainfall (rain gauge tipping-bucket)