[Paper Review] Flexible unit commitment of a network-constrained combined heat and power system

Alvaro González-Castellanos, Priyanko Guha Thakurta+1 more|arXiv (Cornell University)|Sep 25, 2018

Integrated Energy Systems Optimization20 references3 citations

TL;DR

This paper proposes a Mixed Integer Linear Programming (MILP) model for day-ahead flexible unit commitment in network-constrained Combined Heat and Power (CHP) systems, integrating electric and thermal networks with detailed CHP unit modeling, heat storage, and DC power flow constraints. The key contribution is demonstrating that joint operation of thermal and electric systems enhances operational flexibility, efficiency, and economic performance, especially when leveraging heat storage and respecting network limits.

ABSTRACT

Large Combined Heat and Power (CHP) plants are often employed in order to feed district heating networks, in Europe, in post soviet countries and China. Traditionally they have been operated following the thermal load with the electric energy considered as a by-product, while the modern trend includes them in the electric market to take advantage of the flexibility they could provide. This implies the necessity to consider the impact on the electric grid while filling the thermal load requests. A detailed Mixed Integer Linear Programming (MILP) optimization model for the solution of the network-constrained CHP unit commitment of the day-ahead operation is introduced. The developed model accounts for lossless DC network approximation of the electric power flow constraints, as well as a detailed characterization of the CHP units with useful effect, heat and power, function of one and two independent variables ("degrees-of-freedom"), and thermal energy storage. A computational validation of the outlined model on a CHP test system with multiple heating zones is presented in the form of computational test cases. The test cases illustrate the impact on the flexibility of the implementation of the energy storage, network constraints and joint multi-system operation. The conducted studies have highlighted the importance of a comprehensive and integrated analysis of multi-energy systems to exploit the operational flexibility provided by the cogeneration units. The joint operation of the thermal and electric system allows to reap economic, operational efficiency, and environmental benefits. The developed model can be easily extended to include diverse multi-energy systems and technologies, as well as more complex representations of the energy transmission networks, and the modeling of renewable energy resources dependent of one or more independent, weather-related, variables.

Motivation & Objective

• To develop a comprehensive optimization model for day-ahead unit commitment in network-constrained CHP systems.
• To integrate detailed CHP unit characteristics with one or two independent variables (e.g., fuel and power/heat output) for accurate representation.
• To incorporate thermal energy storage and network constraints (via DC power flow) to enhance system flexibility.
• To evaluate the impact of heat storage, network limitations, and multi-system operation on operational flexibility and cost efficiency.
• To enable extension to multi-energy systems and renewable integration through modular modeling.

Proposed method

• Formulates a Mixed Integer Linear Programming (MILP) model for day-ahead unit commitment in CHP systems with network constraints.
• Uses a lossless DC power flow approximation to model active power flow across transmission lines with line susceptance and node angle variables.
• Applies piecewise linearization of performance curves using breakpoints for units with one or two independent variables (e.g., fuel input vs. power/heat output).
• Incorporates binary variables for unit commitment, startup procedures, and minimum uptime constraints to model operational dynamics.
• Models thermal energy balance per zone with heat transfer efficiency, storage losses, and zonal storage capacity.
• Integrates electric and thermal systems via shared CHP units, with constraints ensuring load balance and line thermal limits.

Experimental results

Research questions

• RQ1 How does the inclusion of thermal energy storage affect the flexibility and cost of CHP unit commitment under network constraints?
• RQ2 To what extent do network constraints (e.g., line flow limits) limit the operational flexibility of CHP units in combined heat and power systems?
• RQ3 How does joint operation of electric and thermal networks improve system efficiency and economic performance compared to isolated operation?
• RQ4 What is the impact of modeling CHP units with two independent variables (e.g., fuel and power/heat) versus one on solution accuracy and flexibility?
• RQ5 Can the proposed MILP model be effectively extended to include renewable energy sources and more complex multi-energy system configurations?

Key findings

• The integration of thermal energy storage significantly improves system flexibility by enabling heat shifting, reducing curtailment, and lowering operational costs.
• Network constraints (line flow limits) have a measurable impact on unit commitment decisions, particularly in congested areas, and must be explicitly modeled for accurate dispatch.
• Joint operation of electric and thermal systems leads to higher operational efficiency and economic benefits, as CHP units can better balance both energy markets.
• The use of two independent variables (e.g., fuel and power/heat) in unit modeling enables a more accurate representation of CHP performance, improving solution fidelity.
• The model demonstrates that heat storage losses (assumed at 2% per hour) and heat transfer efficiency (92%) are critical parameters affecting system performance and cost.
• The computational validation on a multi-zone CHP system confirms the model’s ability to handle complex system interactions and deliver optimal, flexible dispatch solutions.