Research On Beam Quality Control Technology Of Hybrid-cavity Slab Solid-state Laser Based On Adaptive Optics

The slab solid-state laser based on the hybrid cavity has the advantages of high efficiency,compact structure,long life,full electric operation,etc.However,the waste heat generated during the working process can cause undesirable thermal effects such as thermal distortion and optical component processing and adjustment errors,resulting in distortion of the wavefront of the output beam,which limits the improvement of the quality of the output beam of the slat solid laser.Adaptive optics is an effective means to solve the wavefront aberration introduced by thermal distortion and improve the beam quality.However,the traditional adaptive optics system based on wavefront sensing has complex structure and large volume,which is not conducive to the realization of the lightweight and miniaturization of the laser system.The wavefront sensor-less beam cleanup system based on the optimization algorithm does not require wavefront measurement and reconstruction,so it has the advantages of simple structure,small size,light weight,and low cost,which can effectively reduce the volume and weight of the system.The performance of wavefront sensor-less beam cleanup technology is affected by the selection of optimization algorithms.Currently,the optimization algorithm with the best comprehensive performance of commonly used algorithms is the stochastic random parallelism reduction algorithm.Compared with the traditional wavefront beam purification technology,the correction speed of this algorithm is slower.The selection of the iterative parameters and optimization index function will affect the speed of the algorithm,and the speed of the algorithm will gradually slow down as the number of optimized control parameters increases;at the same time,the algorithm has poor global search ability,and the algorithm is easy to converge to local extremes.In this paper,researches on the above-mentioned problems of wavefront sensor-less beam cleanup system are carried out,two efficient wavefront sensor-less optimization algorithms are proposed,and numerical simulation and experimental verification are designed and carried out,and a series of results have been obtained.First,this article introduces the structure of the slab solid-state laser based on the hybrid cavity,uses the wavefront reconstruction algorithm to analyze the laser wavefront aberration characteristics,and analyzes the basic principles of the wavefront sensor-less beam cleanup technology and common optimization algorithms.Derives the mathematical model of wavefront sensor-less beam cleanup,and builds a numerical simulation system for wavefront sensor-less beam cleanup.The basic principles and algorithm flow of hill climbing,genetic algorithm,particle swarm algorithm,simulated annealing algorithm and stochastic random parallel gradient descent(SPGD)algorithm are introduced in detail.Through the numerical simulation system,the algorithm speed,the correction effect and the performance of local extreme value problems of each algorithm are compared,and the advantages and disadvantages of each algorithm are compared and analyzed.Studies have found that the overall performance of the SPGD algorithm is the best,but it has the defect of converging to a local optimum.At the same time,compared to the traditional wavefront beam purification system,the correction speed is slower.Secondly,in view of the above problems,combined with the principle analysis and simulation of the SPGD algorithm,the main factors limiting the algorithm performance are deduced and simulated,and targeted improvement strategies are proposed.In order to improve the speed of the algorithm,an adaptive gain coefficient SPGD algorithm is proposed.This algorithm uses two-way indicators and adaptively adjusts the value of the gain coefficient,which can effectively improve the algorithm convergence speed and avoid the frequent modification of parameter settings during the algorithm iteration process..Inspired by the idea of adaptive gain coefficient,by introducing the 2th-order estimation value of the search point,the iteration step size is limited,and the adaptive gradient estimation modified stochastic parallel gradient descent(AGESPGD)algorithm is proposed,which can further improve the algorithm Speed,while effectively suppressing the problem of the traditional SPGD algorithm converging to a local extremum.On the basis of analyzing the influence of various laser beam performance indicators on the SPGD algorithm beam purification performance,combined with the idea of adaptive gain coefficient,an efficient SPGD algorithm based on joint index optimization is also proposed.The algorithm uses dual indicators to guide the iteration direction and step,and is an efficient optimization algorithm for solid-state lasers.At the same time,according to the characteristics of each algorithm,a hybrid optimization algorithm based on the particle swarm algorithm and the SPGD algorithm is proposed.The algorithm combines the excellent global search ability of the particle swarm algorithm and the characteristics of the local optimization of the SPGD algorithm,which can effectively improve the algorithm speed and convergence effect.Enhance the global optimization capability of the SPGD algorithm.Finally,simulation and experimental verification of the proposed algorithm are carried out,and the beam purification experiment of the k W-level hybrid cavity slab solid-state laser is completed by using the proposed high-efficiency SPGD algorithm.Using this article to build a wavefront sensor-less beam cleanup numerical simulation system to simulate and verify the above four algorithms.The simulation results show that,compared with the traditional SPGD algorithm,the adaptive gain coefficient SPGD algorithm is used as the basis of the adaptive gradient estimation modified SPGD algorithm and the high-efficiency SPGD algorithm of joint index optimization,and its improvement effect is not as good as the latter two.The adaptive gain coefficient SPGD algorithm is close,so it is determined that the adaptive gradient estimation modified SPGD algorithm and the high-efficiency SPGD algorithm of joint index optimization are used as the experimental research content.By building a k W-level hybrid cavity Nd:YAG slab solid-state laser beam purification experimental system,the wavefront-free beam purification experimental research was carried out.First of all,experiments are carried out on the efficient SPGD algorithm with joint index optimization,and compared with the traditional SPGD algorithm.The experimental results show that the algorithm speed and purification effect of the efficient algorithm proposed in this paper are far superior to the traditional SPGD algorithm,and the correction effect is comparable to that of a wavefront beam cleanup system.It provides a certain algorithm basis and experimental basis for the replacement of the traditional wavefront beam cleanup system with the wavefront sensor-less cleanup system based on optimization algorithm.Subsequently,an experimental study was carried out on the adaptive gradient estimation modified SPGD algorithm.The experiment uses a deformable mirror to fit the measured aberration as the input wavefront aberration.The experimental results show that the correction effect of the AGESPGD method is slightly improved compared with the traditional SPGD method.The algorithm speed is greatly improved.Experiments have verified that the algorithm is an efficient and feasible general optimization algorithm,which is expected to be applied to solve other optimization problems.