Study On Methods And Application Of Energy-saving Diagnosis Of Gas Turbine

Large gas generator unit is developing rapidly in our country, but operation andmaintenance technology of gas turbine is still in groping stage, which is the core equipment ofa gas generator unit. Therefore, to do energy saving diagnosis study on gas turbines of powerstation is of great significance for achieving the goal of economic operation and energy savingin power industry.For the calculation of gas turbine standard parameter, characteristic models of gasturbine components and overall performance are established through multivariate functionfitting and dynamic mechanical characteristics of typical analytic method. Taking an actualM701F gas turbine in operation as an example, the model of unit under different workingconditions is constructed using the operation data to identify the parameters of the model, andvarious standard performance parameters are calculated.For the performance degradation problem of gas turbine, gas turbine performancecalculation model of degradation conditions was established based on the theory of SupportVector Machine (SVM) and the simulation of an M701F gas turbine was conducted.Performance degradation diagnosis of gas turbine was studied in this paper. Degradationdiagnosis of location identification was tried using the method of Probabilistic NeuralNetworks, and simulation was conducted utilizing the data in the literature.On the basis of gas turbine performance calculation model, sensitivity analysis of maininfluence factors for gas turbine was conducted and the influence of various parameters on theefficiency of gas turbine was calculated. The operation parameters of gas turbine wereoptimized using genetic algorithm. of degradation conditions of gas turbine operation tocalculate The energy saving potential of gas turbine in degradation operating condition wascalculated using M701F gas turbine as an example and the results show that it has muchenergy saving space.For optimal load dispatch problem of operation optimization for combined cycle units, ofwhich gas turbine is the core equipment, the influence of environmental factors on theirperformance was analyzed when establishing the dispatch model. Therefore more reasonableand accurate objective function and constraint conditions were obtained. In the process ofsolving the load dispatch model, Multiple Population Genetic Algorithm (MPGA) was usedand the calculation results were compared with Standard Genetic Algorithm. Results showthat MPGA has obtained the very good optimization results. It can provide guidance for the energy saving operation of gas generator units.