diff --git a/Frontend/Analysis/analysis.html b/Frontend/Analysis/analysis.html
index f1b0098..52e7ced 100644
--- a/Frontend/Analysis/analysis.html
+++ b/Frontend/Analysis/analysis.html
@@ -242,6 +242,17 @@ 🌪 Wind Speed
+ 🤖 AI Predictive Insights
+
+
+
🧠 Rainfall Estimation
+
+
0.0 mm
+
Waiting for system query...
+
+
+
⚠ Climate Risk Analysis
diff --git a/Frontend/Analysis/analysis.js b/Frontend/Analysis/analysis.js
index 5dfa76a..49593bd 100644
--- a/Frontend/Analysis/analysis.js
+++ b/Frontend/Analysis/analysis.js
@@ -259,6 +259,16 @@ async function getWeatherData() {
document.getElementById("wind").innerText =
`${data.weather.wind_speed} km/h`;
+ //ML Rain Prediction
+ if (data.ai_rain_prediction){
+ const rain_val=data.ai_rain_prediction["Predicted Rain"];
+ const rain_status=data.ai_rain_prediction["Predicted Rain Status"];
+ const ml_rain= document.getElementById("ml-rain-mm")
+ const ml_rain_status=document.getElementById("ml-rain-status")
+ if (ml_rain) ml_rain.innerText = `${rain_val} mm`;
+ if (ml_rain_status) ml_rain_status.innerText = rain_status;
+ }
+
// Risks scores
const floodCard = document.querySelector(".risk-card.flood");
const floodScore = data.risks.flood_risk;
diff --git a/backend/alertsystem.py b/backend/alertsystem.py
index bde3472..9951611 100644
--- a/backend/alertsystem.py
+++ b/backend/alertsystem.py
@@ -1,10 +1,23 @@
import os
import sys
import requests
+import pandas as pd
+import numpy as np
+import joblib
from dotenv import load_dotenv
load_dotenv()
+#----loading the ml model for rain prediction
+CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
+model_path=os.path.join(CURRENT_DIR, "xgb_model (2).joblib")
+try:
+ if os.path.exists(model_path):
+ model=joblib.load(model_path)
+ print("imported")
+except:
+ model=None
+ print("Not imported")
GIS_ALERTS_URL = os.environ.get("GIS_ALERTS_URL", "https://example.com/gis/alerts")
@@ -278,6 +291,8 @@ def get_weather_insights():
weather_data["wind"]["speed"] * 3.6,
1
)
+ gust=round(weather_data["wind"].get("gust",0)*3.6)
+ dir=weather_data["wind"]["deg"]
rain_val = (
weather_data.get("rain", {}).get("1h")
@@ -303,6 +318,8 @@ def get_weather_insights():
forecast_response.raise_for_status()
forecast_data = forecast_response.json()
+ #-----------ML prediction-----------
+ ai_prediction_payload = ml_rain_prediction(lat, lon, temp_val, humid_val, wind_val, gust, dir)
# ----------------------------------------------------
# RISK CALCULATIONS
@@ -364,6 +381,7 @@ def get_weather_insights():
),
3
)
+
# ----------------------------------------------------
# ALERTS
@@ -502,7 +520,9 @@ def get_weather_insights():
"temperature": temp_val,
"humidity": humid_val,
"rainfall": rain_val,
- "wind_speed": wind_val
+ "wind_speed": wind_val,
+ "direction":dir,
+ "gust":gust
},
"risks": {
@@ -526,6 +546,7 @@ def get_weather_insights():
"forecast": forecast,
"alerts": calculated_alerts,
+ "ai_rain_prediction": ai_prediction_payload
})
except Exception as general_err:
@@ -676,6 +697,60 @@ def city_suggestions():
except Exception as e:
print("City Suggestions Error:", e)
return jsonify([])
+
+#---------XGBoost model predictions---------
+
+def ml_rain_prediction(lat,lon,temp_val,humid_val,wind_val,gust,dir):# extract the features from open weather
+
+ if not lat or not lon:
+ return {"Predicted Rain": 0, "Predicted Rain Status": "Coordinates not fetched."}
+
+
+ lat_rad = np.radians(lat)
+ lon_rad = np.radians(lon)
+ coord_x= np.cos(lat_rad) * np.cos(lon_rad)
+ coord_y= np.cos(lat_rad) * np.sin(lon_rad)
+ coord_z= np.sin(lat_rad)
+
+ live_features = pd.DataFrame({
+"latitude":lat,
+"longitude":lon,
+"wind_direction_10m_dominant": dir,
+"wind_gusts_10m_mean": gust,
+"wind_speed_10m_mean": wind_val,
+"temperature_2m_mean": temp_val,
+"relative_humidity_2m_mean": humid_val,
+"coord_x":coord_x,
+"coord_y":coord_y,
+"coord_z":coord_z}, index=[0])
+ try:
+ print(live_features)
+ result=model.predict(live_features)
+ rain=round(float(result[0]),2)
+ print(rain)
+ except Exception as e:
+ print(f"❌ MODEL CRASH ERROR LOG: {str(e)}")
+ rain= float(0)
+
+
+ rain_mm=rain
+ try:
+ if rain_mm <= 2.4:
+ rain_status= "No Rain / Light Drizzle ☀️"
+ elif 2.4 < rain_mm <= 15.5:
+ rain_status= "Light Rain 🌧️"
+ elif 15.5 < rain_mm <= 64.4:
+ rain_status="Moderate Rain ⛈️"
+ elif 64.4 < rain_mm <= 115.5:
+ rain_status= "Heavy Rain Alert 🚨"
+ elif 115.5 < rain_mm <= 204.4:
+ rain_status="Very Heavy Rain Warning 🌊"
+ else:
+ rain_status="Extremely Heavy Rain / Flood Risk ⚠️"
+ except Exception as e:
+ rain_status="An unexpected error occurred while determining rain status."
+ return {"Predicted Rain":rain,"Predicted Rain Status":rain_status}
+
# =========================================================
# CHATBOT API
diff --git a/backend/xgb_model (2).joblib b/backend/xgb_model (2).joblib
new file mode 100644
index 0000000..f10f603
Binary files /dev/null and b/backend/xgb_model (2).joblib differ