src/streamlit_app.py

import streamlit as st
import pandas as pd
import numpy as np
import math
import matplotlib.pyplot as plt
import numpy_financial as npf
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer
from reportlab.lib.pagesizes import A4
from reportlab.lib.styles import getSampleStyleSheet
import io
import xlsxwriter

# ===============================
# CONFIGURATION
# ===============================
SYSTEM_LOSSES = 0.20
PANEL_COST_PER_WATT = 55
INSTALLATION_COST_PER_WATT = 35
LITHIUM_BATTERY_COST_5KWH = 95000

CITY_SUNLIGHT = {
    "Karachi": 6.2,
    "Lahore": 5.5,
    "Islamabad": 5.2,
    "Peshawar": 5.6,
    "Quetta": 6.5,
}

APPLIANCES_RESIDENTIAL = {
    "LED Bulb (12W)": 12,
    "Fan (80W)": 80,
    "Refrigerator (200W)": 200,
    "LED TV (150W)": 150,
    "Air Conditioner 1.5 Ton (1500W)": 1500,
    "Washing Machine (500W)": 500,
    "Water Pump (750W)": 750,
    "Laptop (65W)": 65,
    "Iron (1000W)": 1000,
}

APPLIANCES_COMMERCIAL = {
    "CNC Machine (2kW)": 2000,
    "Industrial AC (5kW)": 5000,
    "Lighting System (1kW)": 1000,
    "Water Pump 3HP (2.2kW)": 2200,
    "Server Rack (1.5kW)": 1500,
}

RESIDENTIAL_TARIFF = [
    (100, 22),
    (100, 32),
    (100, 38),
    (100, 42),
    (100, 48),
    (np.inf, 65),
]

COMMERCIAL_TARIFF = 72  # PKR/unit average

# ===============================
# FUNCTIONS
# ===============================

def calculate_residential_bill(units):
    remaining = units
    bill = 0
    for slab_units, rate in RESIDENTIAL_TARIFF:
        if remaining > slab_units:
            bill += slab_units * rate
            remaining -= slab_units
        else:
            bill += remaining * rate
            break
    return bill

def calculate_commercial_bill(units):
    return units * COMMERCIAL_TARIFF

def calculate_system(load_watts, hours, sunlight):
    daily_kwh = (load_watts * hours) / 1000
    adjusted_kwh = daily_kwh / (1 - SYSTEM_LOSSES)
    required_kw = adjusted_kwh / sunlight
    return daily_kwh, round(required_kw, 2)

def calculate_battery(daily_kwh, backup_hours):
    backup_kwh = (daily_kwh / 24) * backup_hours
    batteries = math.ceil(backup_kwh / 5)
    return batteries

def calculate_cost(system_kw, batteries, system_type):
    base_cost = system_kw * 1000 * (PANEL_COST_PER_WATT + INSTALLATION_COST_PER_WATT)
    battery_cost = batteries * LITHIUM_BATTERY_COST_5KWH
    if system_type == "On-Grid":
        return base_cost
    elif system_type == "Off-Grid":
        return base_cost + battery_cost
    else:
        return base_cost * 1.1 + battery_cost

def emi_calculator(principal, annual_rate, years):
    r = annual_rate / 100 / 12
    n = years * 12
    emi = principal * r * (1 + r)**n / ((1 + r)**n - 1)
    return round(emi)

def financial_projection(total_cost, daily_kwh, mode, years=25, inflation_rate=5, energy_price_increase=7):
    monthly_units = daily_kwh * 30
    cashflows = []
    for year in range(1, years+1):
        if mode == "Homeowner":
            monthly_bill = calculate_residential_bill(monthly_units * ((1 + energy_price_increase/100)**(year-1)))
        else:
            monthly_bill = calculate_commercial_bill(monthly_units * ((1 + energy_price_increase/100)**(year-1)))
        annual_savings = monthly_bill * 12
        cashflows.append(annual_savings)
    npv = npf.npv(inflation_rate/100, [-total_cost]+cashflows)
    irr = npf.irr([-total_cost]+cashflows)
    payback_year = next((i for i, cf in enumerate(np.cumsum(cashflows), 1) if cf >= total_cost), None)
    cumulative_savings = np.cumsum(cashflows)
    return cashflows, round(npv,2), round(irr*100,2), payback_year, cumulative_savings

def generate_pdf(report_data):
    file_path = "solar_report.pdf"
    doc = SimpleDocTemplate(file_path, pagesize=A4)
    elements = []
    styles = getSampleStyleSheet()
    elements.append(Paragraph("<b>Pakistan Solar Feasibility Report</b>", styles['Title']))
    elements.append(Spacer(1, 12))
    for key, value in report_data.items():
        elements.append(Paragraph(f"<b>{key}:</b> {value}", styles['Normal']))
        elements.append(Spacer(1, 8))
    doc.build(elements)
    return file_path

def generate_excel(report_data):
    output = io.BytesIO()
    workbook = xlsxwriter.Workbook(output)
    worksheet = workbook.add_worksheet("Solar Report")
    bold = workbook.add_format({'bold': True})
    row = 0
    for key, value in report_data.items():
        worksheet.write(row, 0, key, bold)
        worksheet.write(row, 1, str(value))
        row += 1
    workbook.close()
    output.seek(0)
    return output

# ===============================
# STREAMLIT APP
# ===============================
st.set_page_config(layout="wide")
st.title("🇵🇰 Pakistan Solar Engineering & Financial Dashboard")

audience = st.selectbox("Select Audience", ["Homeowner", "Solar Company", "Industrial Investor"])
city = st.selectbox("Select City", list(CITY_SUNLIGHT.keys()))
sunlight = CITY_SUNLIGHT[city]

if audience == "Homeowner":
    appliances = st.multiselect("Select Appliances", list(APPLIANCES_RESIDENTIAL.keys()))
elif audience == "Solar Company":
    appliances = st.multiselect("Select Residential / Commercial Appliances", 
                                list(APPLIANCES_RESIDENTIAL.keys()) + list(APPLIANCES_COMMERCIAL.keys()))
else:
    appliances = st.multiselect("Select Industrial Equipment", list(APPLIANCES_COMMERCIAL.keys()))

hours = st.slider("Usage Hours per Day", 1, 24, 8)
system_type = st.radio("System Type", ["On-Grid", "Off-Grid", "Hybrid"])
backup_hours = st.slider("Battery Backup Hours", 0, 24, 4)

if st.button("Calculate Solar System"):
    if not appliances:
        st.error("Please select at least one appliance or equipment")
    else:
        total_load = sum(APPLIANCES_RESIDENTIAL.get(a,0) + APPLIANCES_COMMERCIAL.get(a,0) for a in appliances)
        daily_kwh, system_kw = calculate_system(total_load, hours, sunlight)
        batteries = calculate_battery(daily_kwh, backup_hours)
        total_cost = calculate_cost(system_kw, batteries, system_type)

        interest = st.slider("Bank Interest Rate (%)", 5, 25, 15)
        years_loan = st.slider("Loan Duration (Years)", 1, 10, 5)
        emi = emi_calculator(total_cost, interest, years_loan)

        cashflows, npv, irr, payback_year, cumulative_savings = financial_projection(total_cost, daily_kwh, audience)

        # Display Results
        st.subheader("System Analysis")
        st.write(f"Total Load: {total_load} W")
        st.write(f"Daily Energy Consumption: {round(daily_kwh,2)} kWh")
        st.write(f"Required System Size: {system_kw} kW")
        st.write(f"Battery Units Required (5kWh each): {batteries}")
        st.write(f"Estimated System Cost: PKR {round(total_cost):,}")
        st.write(f"EMI (Monthly): PKR {emi:,}")
        st.write(f"25-Year Projection: NPV = PKR {npv:,}, IRR = {irr}%, Payback Year = {payback_year}")

        # Dashboard
        st.subheader("🔹 Daily Load vs Solar Generation")
        hours_day = np.arange(0,24,1)
        load_profile = np.array([total_load]*24)
        solar_profile = np.array([system_kw*1000/sunlight]*24)
        plt.figure(figsize=(10,4))
        plt.plot(hours_day, load_profile, label="Load (W)")
        plt.plot(hours_day, solar_profile, label="Solar Generation (W)")
        plt.xlabel("Hour of Day")
        plt.ylabel("Power (W)")
        plt.title("Daily Load vs Solar Generation")
        plt.legend()
        st.pyplot(plt)

        st.subheader("🔹 Cumulative Savings Over 25 Years")
        plt.figure(figsize=(10,4))
        plt.plot(range(1,26), cumulative_savings, marker='o')
        plt.axhline(total_cost, color='r', linestyle='--', label="Total System Cost")
        plt.xlabel("Year")
        plt.ylabel("Cumulative Savings (PKR)")
        plt.title("Payback & Savings Curve")
        plt.legend()
        st.pyplot(plt)

        st.subheader("🔹 Yearly Cashflows")
        df_cashflow = pd.DataFrame({"Year": range(1,26), "Annual Savings (PKR)": cashflows})
        st.dataframe(df_cashflow)

        st.subheader("🔹 Carbon Emission Reduction Estimate")
        co2_per_kwh = 0.85
        total_co2_saved = round(daily_kwh * 365 * 25 * co2_per_kwh)
        st.write(f"Estimated CO2 Reduction over 25 years: {total_co2_saved:,} kg (~{total_co2_saved/1000:,} tons)")

        # PDF & Excel
        report_data = {
            "Audience": audience,
            "City": city,
            "System Type": system_type,
            "Total Load (W)": total_load,
            "Daily Energy (kWh)": round(daily_kwh,2),
            "System Size (kW)": system_kw,
            "Battery Units": batteries,
            "Total Cost (PKR)": round(total_cost),
            "EMI (PKR)": emi,
            "25-Year NPV (PKR)": npv,
            "IRR (%)": irr,
            "Payback Year": payback_year
        }

        pdf_file = generate_pdf(report_data)
        excel_file = generate_excel(report_data)

        with open(pdf_file, "rb") as f:
            st.download_button("Download PDF Report", f, file_name="Solar_Report_Pakistan.pdf")
        st.download_button("Download Excel Report", data=excel_file, file_name="Solar_Report_Pakistan.xlsx")