Where academic tradition
meets the exciting future

You are here: TUCS > PUBLICATIONS > Publication Search > Efficient Algorithms for Combi...

Efficient Algorithms for Combined Heat and Power Production Planning Under the Deregulated Electricity Market

Aiying Rong, Risto Lahdelma, Efficient Algorithms for Combined Heat and Power Production Planning Under the Deregulated Electricity Market. European Journal of Operational Research 176(2), 1219–1245, 2007.

Abstract:

Combined heat and power (CHP) production is an important energy production technology that can yield much higher total energy efficiency than separate heat and power generation. In CHP production, the heat and power production follows a joint characteristic, which means that the production planning must be done in coordination. Cost-efficient operation of a CHP system can be planned by using an optimization model. A long-term planning model decomposes into thousands of hourly models. Earlier, in the regulated electric power market, the planning problem was symmetrically driven by heat and power demand. The liberalization of the power market has created an asymmetrical planning problem, where heat production responds to the demand and power production to the volatile market price. In this paper, we utilize this asymmetry to develop novel envelope-based dual algorithms for solving the hourly CHP models efficiently. The basic idea is to transform the three-dimensional characteristic operating region for heat and power production of each CHP plant into a two-dimensional envelope by taking the power price as a parameter. Then the envelopes of each plant are used for looking up the optimal solution rapidly. We propose two versions of the algorithm: the on-line envelope construction algorithm (ECON) where the envelopes are constructed for each hour based on the power price and the off-line envelope construction algorithm (ECOFF) where envelopes are pre-computed for all different power price ranges. We derive the theoretical time complexity of the two algorithms and compare their performance empirically with realistic test models against the ILOG CPLEX solver and the Power Simplex (PS) algorithm. PS is an extremely efficient specialized primal algorithm developed for the symmetrical CHP planning problem under the regulated market. On average, when reusing previous basic solutions, ECON is 603 times faster than CPLEX and 1.3 times faster than PS. ECOFF is 1860 times faster than CPLEX and 4 times faster than PS.

BibTeX entry:

@ARTICLE{jRoLa07a,
  title = {Efficient Algorithms for Combined Heat and Power Production Planning Under the Deregulated Electricity Market},
  author = {Rong, Aiying and Lahdelma, Risto},
  journal = {European Journal of Operational Research},
  volume = {176},
  number = {2},
  publisher = {Elsevier},
  pages = {1219–1245},
  year = {2007},
  keywords = {Linear programming; Envelope; Combined heat and power production; Deregulated power market; Energy optimization.},
}

Belongs to TUCS Research Unit(s): Algorithmics and Computational Intelligence Group (ACI)

Publication Forum rating of this publication: level 2

Edit publication