Research On Trajectory Optimization And Release Region Calculation Of Glide Guided Bomb

Improving the ability of attack far away from defense area is an important trend for the development of guided bomb.Research on trajectory optimization of guided bomb has a meaningful significance on improving the performance of weapon system and implementing precision bombing far away from defense area.The study of release region calculation is derived from actual engineering demands.Fast and accurate calculation of release region is the key to seize the initiative during the war.Taking the glide guided bomb as an example,this dissertation studies trajectory optimization and release region calculation problem.The main content is shown as follows:1.Kinetics and kinematics model of glide guided bomb are established.Reasonable assumptions are made in order to simplify the problem.Then the simplified motion equations in the vertical plane are established,which will provide the basis for the later research on the trajectory optimization and release region calculation.2.Aiming at different requirements of fixed ground target,trajectory optimization of glide guided bomb is carried out.Firstly,using Radau pseudospectral method,optimal control problem is transformed into an NLP problem which can be solved by SQP method.Then,an improved hp-adaptive mesh refinement algorithm is proposed.Accuracy of the solution is assessed by the residual of the differential-algebraic constraints,and pseudospectral method or mesh refinement method is chosen to improve accuracy.Finally,trajectory optimization model is established,numerical simulation verifies the accuracy and efficiency of the algorithm.By calculating the maximum range under different release conditions,the influence on release region by initial release condition is given.3.To give full play to the performance of guided bomb and reduce pilot's pressure,release region calculation problem is studied.Firstly,the relationship between the boundaries of release region and extreme range is given.Then the model of release region calculation is established combined with the consideration of the influence on release region by initial release condition.Two methods for calculating release region are proposed,namely multiple linear regression and neural network method based on Extreme Learning Machine(ELM).The parameters are obtained by offline fitting,and then can be loaded into the fire control system so the release region can be calculated in real time.Finally,numerical simulation proves the real-time and accuracy of two methods and the latter method is more accurate and more suitable for actual engineering demands based on the consideration of versatility and fitting efficiency.