Application Of Improved Optimization Algorithm In Health Monitoring Of Liquid Rocket Power System

The development of aerospace technology is of great significance to the improvement of national defense security.Space vehicles are mainly propelled by liquid rockets.The liquid rocket power system is one of its important components.Its high reliability is an important guarantee for the smooth progress of the space launch mission.Therefore,the fault detection and diagnosis of the liquid rocket power system plays an important role.In this paper,two optimization algorithms are improved,and they are combined with the selected neural network and machine learning algorithm to carry out the research on the fault detection and diagnosis model of the liquid rocket power system.Then an intelligent algorithm diagnosis management system is proposed to realize the operation and result display of the algorithm in this paper.The main work is as follows:First of all,in the fault detection of the liquid rocket power system,the basic Harris Hawk Optimization(HHO)has the disadvantages of slow convergence and easy to fall into the local optimum.In this paper,a reverse learning strategy is introduced into the original HHO algorithm,and then the energy formula is turned into a non-linear.Using seven test functions to compare the performance of the four algorithms,the test results show that the improved Harris Hawk Optimization Algorithm(IHHO)improves the convergence speed.In addition,the improved Harris Eagle optimization algorithm is combined with probabilistic neural networks(PNN)to construct an IHHO-PNN model.Through the comparison of confusion matrix,receiver operating characteristic curve,AUC value(Area Under Curve),and true positive rate,it is concluded that the accuracy of fault detection of this model is improved,and it can be better applied to the judgment of whether the liquid rocket power system is faulty or not.Secondly,in the fault diagnosis of the liquid rocket power system,in order to overcome the limitations of the original Sparrow Search Algorithm(SSA),first use the Logistic chaotic map to initialize the population,and at the same time introduce the Levy flight random step into the joiner position update method,finally,after the sparrow search,random walk is used to perturb the optimal sparrow.Ten test functions are used to compare the performance of the four algorithms.The test results show that the improved Sparrow Search Algorithm(ISSA)can reduce the probability of the original algorithm entering the local optimum.In addition,the improved sparrow search algorithm is combined with a support vector machine(Support Vector Machine,SVM)to construct an ISSA-SVM model.The comparison of the classification error rate shows that the model improves the classification accuracy and is more conducive to the classification of fault types.Finally,in order to facilitate the application of the liquid rocket power system health monitoring project and realize the operation and result display of the algorithms in the first two chapters,a Web-based equipment intelligent diagnosis management system is designed and constructed.The system can realize the functions of data format conversion,algorithm import and operation,and original data recording and drawing.