The Investment Risk Analysis Of Intelligent Manufacturing Project Based On SA And GA Improved BP Neural Network Model

With the deep integration of information technology and modern manufacturing industry,the era of industrial transformation brought about by intelligent manufacturing is coming.Our country has also put forward national plans such as"Made in China 2025"to vigorously develop the intelligent manufacturing industry.Intelligent manufacturing factories can achieve fewer people and much precise production to improve production efficiency and reduce rejection rate.Meanwhile,they can improve customer loyalty by customize production.So intelligent manufacturing is widely favored by enterprises.Intelligent manufacturing proj ects are investing more and more,but the research on their risks has not received extensive attention.Compared with traditional manufacturing projects,the investment risks of intelligent manufacturing projects present new characteristics such as systematization and networking,and has great multi-node and mutual correlation.It is necessary to carry on in-depth research.Firstly,the thesis expounds the process of investment risk analysis of intelligent manufacturing project and constructively puts forward the theory and method of investment risk analysis of intelligent manufacturing project by defining the concepts of intelligent manufacturing project risk and investment risk analysis of intelligent manufacturing project,based on in-depth analysis of the characteristics of intelligent manufacturing projects.Secondly,the main risk factors of intelligent manufacturing projects investment risks are presented systematically through in-depth analysis of the investment characteristics of intelligent manufacturing projects and the summary of the investment risk characteristics of intelligent manufacturing projects.Thirdly,BP neural network is chosen as the main model on the basis of comparing the applicability of common risk analysis methods.In view of the blindness of the number selection of hidden neurons,the shortcomings of lack of theoretical guidance,easy to fall into local optimum and slow convergence speed in the model,simulated annealing algorithm(SA)and genetic algorithm(GA)are adopted to improve the model,so as to establish SA-GA-BP investment risk neural network analysis model.Finally,the above theory and model are applied to practice by using actual cases,and the accuracy of model prediction before and after improvement is compared in the result analysis,which makes the model more practical.In addition,the avoidance guidance methods are proposed for the main risks in the investment of practical intelligent manufacturing proj ects.In theory,the thesis systematically expounds the characteristics and main risk factors of intelligent manufacturing investment,and preliminarily explores the risk analysis theory of intelligent manufacturing project,which has certain reference significance and reference value for investment risk analysis of similar intelligent manufacturing projects.For modeling,the BP neeural network model is deeply improved by combining SA and GA,and a SA-GA-BP investment risk analysis simulation evaluation model,which is more suitable for the risk assessment field of intelligent manufacturing projects,is established.Furthermore,the algorithm design and simulation calculation are carried out based on the software platform of Matlab,to provide a reference basis for scientific project investment decision-making.In the example,the theoretical model of investment risk analysis is simulated and verified by using the case of intelligent manufacturing investment,and its main risks are pointed out.Moreover,a few feasible risk avoidance method are provided for its main risk factors,to promote the process of risk analysis and the application of the model,which has more practical guiding significance.