Ballistic wind parameter identification based on PSO-LM algorithm

Abstract To improve the meteorological support capability of weapon systems and satisfy the operational requirements of rapid, mobile, and precise strikes of artillery, aiming at the problem of lack of high-altitude meteorology in artillery shooting, taking ballistic wind as the research object, based on system identification method and least squares criterion, combined with the motion equation of projectile centroid under non-standard meteorological conditions and the meteorological model with standard distribution with height, a PSO-LM joint algorithm is proposed. The identification model of ballistic wind parameters is established. The number of particles is dynamically adjusted by the linear decreasing method. The inertia weight is adaptively adjusted by personal optimal value. The shortcomings of the LM algorithm sensitivity to initial value and the slow convergence speed of the standard PSO algorithm are improved. Using a large-caliber shell as a case study for simulation calculations, the simulation results show that the global search ability and local optimization ability of the algorithm can be effectively balanced by selecting the square sum of coordinate residuals as the objective function. Especially for the identification of range wind, the error is significantly smaller than other objective functions. The relative error of the range between the identified ballistic wind extrapolated ballistic falling point and the measured ballistic falling point is less than 2 %, and the relative error of the lateral deviation is less than 4 %, which has high accuracy in ballistic evaluation.