Fuel gas utilization rate prediction method and system based on hybrid neural network

Abstract

The invention relates to a fuel gas utilization rate prediction method and system based on a hybrid neural network, and relates to the technical field of gas-steam combined cycle power generation. The method comprises the following steps: acquiring working condition parameters, gas consumption amount and generating capacity corresponding to compressed air amount of a gas turbine unit in a plurality of preset periods, establishing a single neural network model according to the divided working condition parameters, gas consumption amount, compressed air amount and corresponding generating capacity, and training the single neural network model by taking the gas consumption amount as an input amount; and after the training is completed, combining the single neural network models to generate a first hybrid neural network model, combining the first hybrid neural network models in a plurality of preset periods to generate a second hybrid neural network model, and taking the gas consumption as an input parameter of the second hybrid neural network model to obtain a power generation prediction value. The accurate prediction of the gas utilization rate is improved, so that the operation reliability of a gas turbine power plant is further improved.

Description

technical field [0001] The invention relates to the technical field of gas-steam combined cycle power generation, in particular to a method and system for predicting gas utilization rate based on a hybrid neural network. Background technique [0002] Gas-steam combined cycle generator set (hereinafter referred to as gas turbine) has the advantages of environmental friendliness, high energy utilization rate, large adjustment range, fast adjustment speed and quick load response. Good peak performance and other characteristics. [0003] However, due to the large-scale construction of thermal power and new energy, the peak-shaving capacity of the power grid is improved, and the gas-turbine power plant is mainly positioned to adjust the balance between supply and demand of urban natural gas, optimize the layout of urban power sources, enhance the safe operation of the power grid and natural gas pipeline network, and undertake double peak-shaving The important task: not only adju...

AI Technical Summary

Problems solved by technology

[0005] To this end, the present invention provides a method and system for predicting gas utilization rate based on a hybrid neural network to overcome the problem in the prior art that the gas utilization rate of gas turbine units cannot be accurately predicted, resulting in low reliability of power supply for gas-fired power plants

Images