Towards Energy-Water Synergy: A Multi-Domain MILP Approach to Smart Microgrids
2040_Step Towards Energy-Water Smart Microgrids; Buildings Thermal Energy and Water Demand Management Embedded in Economic Dispatch.
The paper proposes a Mixed-Integer Linear Programming (MILP) framework for the day-ahead economic dispatch of smart microgrids, integrating interdependent energy, water, and building thermal networks. By utilizing piecewise linear approximation for nonlinear hydraulic and thermal functions, the model achieves a global optimum for co-optimizing resource allocation across three critical infrastructures.
TL;DR
Researchers have developed a new mathematical framework to bridge the gap between energy generation, water distribution, and building climate control. By converting complex nonlinear physical laws into a Mixed-Integer Linear Programming (MILP) model, they achieved a global optimum that reduces microgrid operational costs by over 30% through the intelligent use of variable-speed water pumps and thermal management.
Background: The Invisible Link
Water and energy are deeply intertwined. Power plants require water for cooling, while water treatment and distribution consume roughly 8% of global energy. Despite this, most "Smart City" algorithms manage power grids and water pipes as two separate universes. This paper argues that by treating them as a single, co-dependent system, we can unlock massive efficiencies that isolated systems miss.
The Core Problem: Complexity and Non-Linearity
The physics of a smart microgrid is messy.
- Water Hydraulics: The power consumed by a pump depends on flow rate and pressure head in a nonlinear, bivariate relationship.
- Thermal Equilibrium: Identifying the exact cooling needed for a building involves differential equations accounting for outdoor heat, solar radiation through windows, and even the heat generated by occupants.
Traditional methods often simplify these into linear models (losing accuracy) or use non-linear solvers (which can get stuck in "local optima").
Methodology: Taming Nonlinearity with MILP
The researchers' breakthrough lies in their "Linearization Strategy." They broke down the complex, curved surfaces of pump power and heat transfer functions into many small, flat "pieces" (Piecewise Linearization).
1. Unified Architecture
The model controls a microgrid containing:
- Generation: Wind, Solar, Battery Storage, and Diesel Generators.
- Loads: Residential, Office, and Apartment buildings.
- Water: A 7-node network with reservoirs, tanks, and variable-speed pumps.
2. The Bi-Variate Linearization
To handle pumps, the authors used a triangular grid approach. By dividing the flow (Q) and speed (r) axes into segments, they approximated the pump's curved power surface as a series of connected triangles, allowing the MILP solver to "walk" across the surface toward the mathematical best result.
Key Insights from Experiments
Variable Speed Pumps are Game Changers
The study highlights that "fixed speed" is the enemy of efficiency. When water pumps were allowed to vary their speed between 50% and 100%, the total energy cost dropped from 316.6 per day. This is because the system can "buffer" water in tanks when energy is cheap/renewable and slow down during peak price periods.
The "Pre-Cooling" Trap
One counter-intuitive finding was related to building thermal management. Many expect that letting a building warm up when no one is home (flexible settings) saves money. However, the study found that the energy required to "cool down" the building quickly before residents return creates massive power peaks. Maintaining a steady 20-25°C all day was actually $33 cheaper per day than fluctuating strategies.
Critical Analysis & Future Outlook
The strength of this work is its Global Optimality. Because it is MILP-based, we are guaranteed the best possible mathematical solution for the given constraints.
Limitations:
- Leakage: The model assumes a perfect water network. In reality, water loss (non-revenue water) would change the mass balance equations significantly.
- Computation: While MILP is efficient, as the number of buildings and pipes grows, the "piecewise" segments multiply, which could lead to a "curse of dimensionality" for mega-cities.
Future Work: The authors suggest moving toward Cyber-Resilience, ensuring that these interlinked systems cannot be taken down by a single hack on one of the networks.
Conclusion
This paper serves as a blueprint for the "Smart Microgrid" of 2030. It proves that by mathematically merging water distribution and building HVAC into the power dispatch routine, we can run our cities with significantly fewer fossil fuels and lower costs.