Study On The Initial Pressure Optimization Of Steam Turbine Based On The Inproved Whale Optimization Algorithms

In recent years,with the rapid development of China's social economy and the continuous adjustment of electricity consumption structure and policies,more and more large-capacity steam turbine units have participated in the peak load operation.As the peak load operation is performed,the unit will work in the low load area for a long time.The thermal economy of the unit is significantly reduced,so it is necessary to study the optimal operation of the thermal power unit.In order to ensure that the unit still maintains high thermal economy during variable load operation,the sliding pressure operating curve of the steam turbine must be optimized to reduce the unit's heat rate.Because of the non-linear and strong coupling characteristics of the turbine unit,it is difficult to establish an accurate initial pressure optimization model in the conventional method.Therefore,in this paper,the artificial neural network and swarm intelligence optimization algorithms in the field of artificial intelligence are applied to the study of turbine initial pressure optimization to realize the economic operation of the unit.Its main content is described as follows:Firstly,aiming at the shortcomings of Whale Optimization Algorithm(WOA),such as low convergence accuracy and slow convergence rate,an adaptive Whale Optimization Algorithm(AWOA)based on inverse learning is proposed.Finally,in order to verify the effectiveness and efficiency of the AWOA algorithm,10 test functions were used to compare it with particle swarm optimization(PSO),differential evolution(DE),and basic WOA algorithms.It shows that AWOA algorithm is superior to the other three algorithms in convergence accuracy and convergence speed.Then,when the Fast Learning Network(FLN)is used to model the heat rate of a steam turbine,the FLN input weights and hidden layer thresholds are randomly initialized,which greatly reduces the stability and reliability of the model.Therefore,the AWOA algorithm is used to optimize the input weights and hidden layer thresholds of the FLN,and then the AWOA-FLN is used to comprehensively model the heat rate.The experimental simulation results show that the AWOA-FLN heat rate prediction model hashigher prediction accuracy and pan Stronger ability.Finally,based on the AWOA-FLN heat rate prediction model,the AWOA algorithm is used to target the lowest heat rate in the feasible pressure range to optimize the main steam pressure,and the optimal initial pressure and heat obtained by this method are used.The consumption rate has decreased to different extents and the thermal economy of the unit has been improved.According to the optimized sliding pressure running curve,it has a strong guiding significance for the economic operation of the power plant.