app.py

# app.py

import streamlit as st
import pandas as pd
import numpy as np

# Set page title and layout
st.set_page_config(page_title="Water System Analysis", layout="centered")
st.title("Water Network Leak, Surge, and Gas Flow Analyzer")

# Allow user to upload an Excel or CSV file
uploaded_file = st.file_uploader("Upload Excel or CSV File", type=["xlsx", "csv"])

if uploaded_file:
    try:
        # Read the uploaded file into a pandas DataFrame
        if uploaded_file.name.endswith('.xlsx'):
            df = pd.read_excel(uploaded_file)
        else:
            df = pd.read_csv(uploaded_file)
        st.success("Data loaded successfully!")
    except Exception as e:
        st.error(f"Error reading the file: {e}")
        st.stop()
    
    # Detect if a "Serial Number" column exists (case-insensitive)
    serial_col = None
    for col in df.columns:
        if col.strip().lower().startswith("serial"):
            serial_col = col
            break

    # If serial number column is found, let user select one
    if serial_col:
        serial_list = df[serial_col].unique().tolist()
        selected_serial = st.selectbox(f"Select {serial_col}", serial_list)
        df_filtered = df[df[serial_col] == selected_serial]
        if df_filtered.empty:
            st.warning("No data found for the selected serial number.")
            st.stop()
    else:
        # If no serial number column, use the entire dataset
        df_filtered = df
        selected_serial = None

    # Let the user select which parameter (column) to analyze
    # Exclude the serial column itself from selection
    columns = [col for col in df_filtered.columns if col != serial_col]
    selected_column = st.selectbox("Select a parameter to analyze", columns)
    
    if selected_column:
        # Retrieve the value(s) for the selected parameter
        # If only one row is present, get the scalar; otherwise get list of values
        if len(df_filtered) == 1:
            value = df_filtered[selected_column].iloc[0]
        else:
            value = df_filtered[selected_column].tolist()
        rec = ""  # This will hold the recommendation string
        
        # Leak detection (Water Pressure)
        if "water pressure" in selected_column.lower():
            # Compute mean pressure if multiple readings, else use the single value
            if isinstance(value, list):
                mean_val = np.mean(value)
            else:
                mean_val = value
            # Apply simple threshold rules
            if mean_val < 20:
                rec = ("🚨 **Alert:** The water pressure is significantly low. "
                       "This suggests a strong likelihood of a leak or supply issue. "
                       "Inspect the pipeline and valves immediately. ")
            elif mean_val < 40:
                rec = ("⚠️ The water pressure is on the lower side. "
                       "There may be a leak or high demand. "
                       "Monitor the system closely and check for any anomalies. ")
            elif mean_val > 80:
                rec = ("⚠️ The water pressure is very high. "
                       "Ensure the pressure-regulating devices are functioning. "
                       "High pressure can stress the pipes. ")
            else:
                rec = ("✅ The water pressure is within normal range. "
                       "No immediate issues detected for leaks based on pressure. ")
        
        # Surge detection (Transient Pressure)
        elif "transient pressure" in selected_column.lower():
            # Use the highest transient pressure to check for spikes
            if isinstance(value, list):
                peak_val = np.max(value)
            else:
                peak_val = value
            if peak_val > 60:
                rec = ("🚨 **Alert:** Detected a high transient pressure (pressure surge). "
                       "This could indicate water hammer or sudden valve closure. "
                       "Check surge protection devices and ensure valves operate smoothly. ")
            else:
                rec = ("✅ Transient pressure levels are within a safe range. "
                       "No significant pressure surges detected. ")
        
        # Gas flow analysis
        elif "gas flow" in selected_column.lower():
            if isinstance(value, list):
                mean_flow = np.mean(value)
            else:
                mean_flow = value
            if mean_flow > 150:
                rec = ("⚠️ Gas flow rate is high. "
                       "This could indicate heavy usage or a potential leak. "
                       "Check the gas lines and usage patterns. ")
            elif mean_flow < 50:
                rec = ("⚠️ Gas flow rate is very low. "
                       "Ensure that valves are open or that the demand is correct. "
                       "No gas might indicate a supply issue. ")
            else:
                rec = ("✅ Gas flow rate is within normal operating range. "
                       "No obvious anomalies detected in gas flow. ")
        
        # Network Design Type advice
        elif "network design" in selected_column.lower():
            design = str(df_filtered[selected_column].iloc[0]).strip().lower()
            if "grid" in design:
                rec = ("✅ A grid network provides redundancy and reliability. "
                       "Ensure maintenance of all loops for continuous supply. ")
            elif "radial" in design:
                rec = ("⚠️ A radial network is simpler but has less redundancy. "
                       "A failure in a main line may isolate downstream areas. "
                       "Consider backup routes if possible. ")
            elif "ring" in design or "loop" in design:
                rec = ("✅ A ring (loop) network allows for good circulation and redundancy. "
                       "Keep valves balanced to maintain pressure. ")
            else:
                rec = ("✅ The network design type is noted. "
                       "Regular inspection ensures reliability. ")
        
        # Population Density advice
        elif "population density" in selected_column.lower():
            val = df_filtered[selected_column].iloc[0]
            if val > 1500:
                rec = ("⚠️ High population density indicates heavy demand on the water network. "
                       "Ensure the infrastructure is scaled properly to serve the community. "
                       "Consider capacity upgrades or demand management programs. ")
            elif val > 800:
                rec = ("✅ Population density is moderate. "
                       "Maintain regular monitoring to meet demand and plan for growth. ")
            else:
                rec = ("✅ Low population density suggests light demand. "
                       "Focus on efficient resource allocation and cost-effective maintenance. ")
        
        # Average Water Usage advice
        elif "water usage" in selected_column.lower():
            val = df_filtered[selected_column].iloc[0]
            if val > 200:
                rec = ("⚠️ Very high per-capita water usage detected. "
                       "Promote water conservation measures and check for leaks in consumption systems. ")
            elif val > 120:
                rec = ("✅ Average water usage is within a moderate range. "
                       "Continue to encourage efficient use of water. ")
            else:
                rec = ("✅ Water usage per person is low. "
                       "This is efficient, but ensure basic needs are met. ")
        
        # Available Water Resources advice
        elif "available water" in selected_column.lower():
            val = df_filtered[selected_column].iloc[0]
            if val < 80:
                rec = ("⚠️ Limited available water resources. "
                       "Implement conservation strategies and consider alternative sources. ")
            elif val > 150:
                rec = ("✅ Ample water resources are available. "
                       "Still practice sustainable usage to maintain supply levels. ")
            else:
                rec = ("✅ Water resource levels are moderate. "
                       "Monitor usage rates to avoid shortages. ")
        
        # Pipe Sizes advice
        elif "pipe sizes" in selected_column.lower():
            val = df_filtered[selected_column].iloc[0]
            if val < 30:
                rec = ("⚠️ Small pipe diameter detected. "
                       "This may limit flow capacity. "
                       "Check for potential bottlenecks in the network. ")
            elif val > 40:
                rec = ("✅ Pipe size is large. "
                       "Adequate flow capacity is available. "
                       "Ensure regular inspections for longevity. ")
            else:
                rec = ("✅ Standard pipe size. "
                       "Flow capacity is adequate for normal operation. ")
        
        else:
            # Default case if the column is unrecognized
            rec = ("✅ No specific recommendations for the selected parameter. "
                   "Data value is noted and appears within expected range. ")
        
        # Display the recommendations
        st.markdown(f"### Recommendations for **{selected_column}**")
        st.write(rec)