An efficient model and algorithm for the transmission-constrained multi-site combined heat and power system
| Autor: | Risto Lahdelma, Aiying Rong |
|---|---|
| Přispěvatelé: | Department of Mechanical Engineering, Energy efficiency and systems, Department of Mathematics and Systems Analysis, Aalto-yliopisto, Aalto University |
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
Mathematical optimization
ta222 Information Systems and Management General Computer Science Linear programming Computer science 020209 energy multi-site energy system 0211 other engineering and technologies 02 engineering and technology Management Science and Operations Research Power budget Industrial and Manufacturing Engineering law.invention Electric power system law combined heat and power production power transmission network 0202 electrical engineering electronic engineering information engineering Power-flow study Energy system Power transmission 021103 operations research business.industry linear programming energy optimization Demand forecasting Base load power plant Transmission (mechanics) Modeling and Simulation Distributed generation business Algorithm |
| Popis: | This paper deals with the transmission-constrained multi-site combined heat and power (CHP) problem and formulates it as a linear programming (LP) model with a special structure. CHP systems are treated as an extension of power-only systems. Each site can be treated as a regional energy system to supply both heat and power. Heat demand is satisfied by local production while power demand can be satisfied by local generation plus power exchange over the power network. The challenge of this problem is that power transmission needs to be coordinated with both power and heat production in each site. The transmission-constrained multi-site CHP system can be operated cost-efficiently according to hourly demand forecast for heat and power by coordinating production and transmission activities among different sites. An efficient network power Simplex algorithm is developed to this end. Numerical experiments with realistic test data show that the algorithm is 7–360 (with average 30) times faster than a commercial LPcode. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|