Optimization research on load dispatch in gas-steam combined cycle units

ABSTRACT

Gas-steam combined cycle units, with gas as fuel, are more efficient than the thermal units. The output of the combined system includes gas turbine and steam turbine load. With the load variation during the operation, it is difficult for the main equipment to keep running under the economic load all the time, and the fuel consumption will increase compared with the optimal working conditions. In this paper, special two-one economic consuming models are established as the base of dispatch problems, with the steam extraction flow and low-pressure water extraction flow as the model input, while the heat output as the model output. A hybrid algorithm is proposed in this paper by integrating the SA strategy into the basic BBO algorithm, called SABBO algorithm, to solve optimization problem. In this modified algorithm, a simulated annealing operator is used to randomly perturb the best individual retains after migration and mutation operating in the basic BBO. A benchmark function is then taken to test the proposed method’s optimal capacity. The proposed algorithm is then used in load optimal distribution model for gas-steam combined cycle units (one-one units and one two-one units). The optimal fuel flow calculated with SABBO algorithm by Optimal 2 is 202.17 t/h, lower than 206.63 t/h by Optimal 1 when power and heat demands are 1000 MW and 575 MW, respectively. Moreover, optimization results by BBO and SABBO algorithms under Optimal 2 solution with demands as [1000, 575], [1000, 700], and [800, 575] MW are listed in the paper. This case tells SABBO algorithm under Optimal 2 solution can be used in gas-steam combined cycle units dispatch problems for lower fuel flow consumption as 202.17 t/h, 211.01 t/h, and 171.76 t/h.

KEYWORDS:

• Gas-steam combined cycle
• gas turbine
• dispatch problem
• consuming model
• optimal

Nomenclature

Dthe heat output

Fthe fuel total cost of the power plant

Gflow quantity

Mthe number of units

Nthe power output

Tthe temperature

ccoefficient in cost function

Greek symbols

$\alpha$Lagrange multipliers

$\lambda$immigration rate

$\mu$emigration rate

$\xi$prediction error value

$\sigma$kernel parameter

Subscripts

$max$maximum value

$min$minimum value

‘ anew sample

*current optimal

1one-one cycle unit

2Atwo-one cycle unit-A

2Btwo-one cycle unit-B

Ccounter number

GTgas turbine

Llength of Markov chain

Nhabitat number

STsteam turbine

Rdemand value

Xindividual population

cheat quantity

mmain steam

$i$$i$th number

$j$$j$th number

ssteam extraction

wwater extraction

$n$sample number

xsample x

ysample y

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

All data, models, and code generated or used during the study appear in the submitted article.

Additional information

Funding

This work is supported by National Natural Science Foundation of China [51706093]; Science Foundation of Nanjing Institute of Technology [YKJ201711]; Research projects of Natural Science Foundation of Universities in Jiangsu [18KJB470012].

Notes on contributors

Hui Gu

Hui Gu, the lecturer in Nanjing Institute of Technology, focuses on optimization operation of units.

Hongxia Zhu

Hongxia Zhu, the associate professor in Nanjing Institute of Technology, researches on units' operation monitoring.

Fengqi Si

Fengqi Si, the professor in Southeast University, researches on units' operation optimization and monitoring.