Thermoelectric Load Optimization of Gas-steam Combined Dycle Unit Based on Deep Learning

In order to optimize the thermoelectric load of the unit and control the gas consumption in the unit, the deep learning algorithm was introduced and the design and research of the thermoelectric load optimization method was carried out with the gas-steam combined cycle unit as an example. According to the unit structure, the thermodynamic function of top cycle is constructed. The deep learning algorithm is introduced to transform the optimal distribution of thermoelectric load into a multi-variable optimization problem, and the multi-dimensional dynamic planning of unit operation is carried out by combining thermodynamic functions. The optimal design of thermoelectric load is realized by decomposing the problem into several stages by combining the equal incremental rate method with multidimensional dynamic programming. The experimental results show that the designed method can optimize the distribution of thermoelectric load of gas-steam combined cycle unit and ensure the balance of supply and demand. At the same time, the method can also reduce the load fluctuation range on the basis of controlling the gas consumption, so as to optimize the unit operation.