Fuzzy Optimization Neural Network Comprehensive Evaluation Of The Key Parameters Error Scheme Of Artillery

There are errors or uncertainties in artillery engineering practice,such as the material in the manufacturing process,the assemblage of mechanical components,launch environment and so on,which are coupled together influence the performance of whole systm of the artil-lery weapon.Based on this research background,the artillery key parameters and its errors were optimized using the modern optimization theory and method,the comprehensive evalua-tion of artillery key parametr error schemes was studied.The main contents are as follows:Firstly,a finite element model of a trucked-mounted howitzer top section was built.The methods of optimal Latin hypercube sampling Ding and BP neural network were used to es-tablish the approximate models for describing the nonlinear relation between the output like mass,stress,displacement and projectile initial disturbance,and the design variables of artil-lery top carriage key structure parameters and uncertain variables of material characteristic parameters.Secondly,the interval uncertain optimization model of the artillery top carriage key pa-rameters and its errors was established.It was transformed into a deterministic optimization model based on the interval number order.On the basis of the approximate model,the intelli-gent optimization algorithm was used to solve the interval uncertain optimization mathemati-cal model of the artillery key parameters and its errors.Thirdly,the analytic hierarchy process was used to obtain the weight values of different evaluation indexes for artillery key parameters error schemes evaluation index system.Char-acteristic values of artillery key parameters error schemes evaluation indexes were obtained using the methods of optimal Latin hypercube sampling Ding and approximate model.Based on the principle of fuzzy membership,characteristic values of evaluation indexes were stan-dardized handling.Fourthly,based on the fuzzy theory and artificial neural network theory,the fuzzy opti-mization BP neural network comprehensive evaluation model was constructed.The model was used in the comprehensive evaluation and optimization of key parameters error schemes for a large caliber gun,and the reasonable evaluation results was obtained.