激波/激波的相互作用是超声速流动中的常见现象，其往往会在局部壁面形成极高的压力载荷和热载荷，是高超声速飞行器气动外形设计和进气道设计的重要参考依据。而高超声速强激波诱导的热化学反应会显著影响流动规律，当其和激波相互作用结合在一起时使得问题更加复杂，强激波相互作用规律及其极端热载荷的诱发机制是保障高超声速飞行安全的重要基础问题。本文针对激波相互作用的两个重要问题：定常激波反射和斜激波/弓形激波干扰，综合应用了理论分析、数值模拟、机器学习和风洞实验方法，系统地分析了强激波诱导的热化学反应对于激波反射流场参数和转变准则的影响，揭示了强激波干扰的几何结构特征与来流参数的关系，发现了一种新的激波干扰类型，进一步分析了它的流场结构和诱发极端热载荷的机制，并进行了大尺度的激波干扰实验，获得了高空间分辨率的热流分布。取得的主要创新性成果如下：

1）系统地分析了强激波诱导波后气体分子振动激发对激波反射及其转变准则的影响。应用考虑分子振动激发的真实气体热力学函数模型，得到了振动激发下的激波关系、等熵关系和P-M膨胀波关系，并以此求解强激波反射的流场参数。研究发现对于斜激波的压力极曲线，振动激发使得弱解部分的压力减小，强解部分的压力增加，上述不同的变化趋势使得整个激波极曲线的轮廓变大。极曲线的变化使得规则反射与马赫反射转变的von Neumann准则和脱体准则对应的角度发生变化，理论分析表明其脱体准则对应的角度明显增大。随后的数值模拟验证了上述理论发现。

2）通过机器学习方法预测了斜激波-弓形激波干扰结构的几何参数，并以此得到了射流结构及其冲击壁面的位置，即极端热载荷的位置。对于III型和IV型激波干扰，由于流场结构的复杂性以及高温气体热化学反应过程的复杂性，尚无方法表征干扰结构与来流参数的关系。因此，我们另辟蹊径，利用机器学习方法，MBB算法，从大量激波干扰数值模拟的数据中学习到了激波干扰几何参数与来流参数的关系。进一步结合理论方法获得了射流结构，并以此成功预测了射流冲击壁面的位置，也就是极端热载荷的位置。此外，我们通过数值模拟中激波干扰类型的数据，通过机器学习得到了III型、IV型和IVa型转变的准则。

3）发现了一种新的激波干扰类型，其热流峰值比IV型激波干扰更大。在分析激波干扰峰值热流随位置参数Ir的变化中发现了产生最大热流峰值的流场结构是一种未知干扰类型，其出现在IV型和III型激波干扰之间，我们将其命名为IIIa型。通过分析它的流场结构变化，我们发现它会产生上下两道超声速射流，中间为马赫杆后的亚声速流动结构。当上射流变小并冲击壁面时就会在局部形成极大的热载荷，这就是它产生最大热载荷的机理。

4）进行了大尺度激波干扰模型实验，获得了高清的激波干扰纹影照片和高空间分辨率的压力、热流分布。由于IV型激波干扰的压力、热流波峰尖锐，而传感器的大小有限制，为了布置足够多的测点捕捉到完整波峰，我们设计加工了大尺度的圆柱模型以安装足够多的热流传感器。我们进行了M7和M8两种工况工九个车次的实验，获得了不同的激波干扰结构的纹影照片和压力、热流分布。在纹影照片中观察到了射流中压缩和膨胀过程造成的明暗相间现象。而激波干扰引起的压力和热流放大率分别为3~5和6~12。

The shock-shock interference is a common phenomenon in supersonic flow, which often forms extremely high-pressure loads and thermal loads on local walls. It is an important reference for the aerodynamic shape design and inlet design of hypersonic aircraft. The thermochemical reaction induced by the hypersonic strong shock wave will significantly affect the flow law. When it is combined with the shock-shock interference, the problem is more complicated. It is an important basis for ensuring the safety of hypersonic flight to accurately predict the flow law of the strong shock-shock interference and the mechanism of extreme thermal load. This dissertation focuses on the two important issues of shock wave interference: steady oblique shock reflection and the interaction of oblique shock with the bow shock. Theoretical analysis, numerical simulation, machine learning, and shock tunnel experimental methods were applied in the study. In this work, the effects of thermochemical reactions induced by strong shock waves on the flow field parameters and transition criteria of shock wave reflection were revealed. The relation function between the geometric structure characteristics of strong shock-shock interaction and incoming flow parameters was revealed. Moreover, a new type of shock-shock interaction was discovered and its flow field structure and the mechanism of generating extreme heat loads were analyzed. At last, we conducted large-scale shock-shock interaction experiments to obtain heat flow distribution with high spatial resolution. The main innovative achievements are as follows:

1) The influence of vibration excitation induced by strong shock waves on steady shock wave reflection and its transition criterion was systematically analyzed. Based on thermodynamic functions of the real-gas model considering vibrational equilibrium flow, the shock wave relation, isentropic relation, and P-M expansion wave relation under vibration excitation were deduced. Then the flow field parameters of strong shock wave reflection were solved by these theoretical relations. It was found that for the pressure shock polar of the oblique shock wave, vibration excitation makes the pressure of the weak solution part decrease and the pressure of the strong solution part increases. Two different changing trends make the profile of the entire shock polar larger than that of a calorically thermal perfect gas flow. The change of the shock polar makes the angles corresponding to the von Newman criterion and the detach criterion of shock reflection transition change accordingly. Theoretical analysis shows that the angle corresponding to the detach criterion increases obviously, while the von Newmann criterion increases slightly. Subsequently, this conclusion was verified by numerical simulations.

2) The machine learning method was used to predict the locations of the shock intersecting points. The jet structure and its impinging position, which corresponds to the position of the extreme thermal load, were obtained accordingly. For Type III and Type IV shock interactions, there is no theoretical formula to characterize the relation between the interaction configurations and the incoming flow parameters at this time due to the complex flow structures and thermochemical processes of the high-temperature shocked gases. Therefore, we took a different approach, using machine learning methods, namely MBB algorithms to learn the relation between the locations of shock intersecting points and incoming flow parameters from a large amount of shock-shock interaction numerical simulation data. The jet structure was obtained and the jet impinging place with extreme thermal load was successfully predicted using the machine learning-theoretical analysis combined method. In addition, through the data of shock-shock interaction types obtained in the numerical simulations, we learned the criteria for the transitions among Type III, Type IV, and Type IVa.

3) A new type of shock-shock interaction was discovered, which features a peak heat flux even greater than that of type IV shock-shock interaction. In analyzing the variation of the shock-shock interaction peak heat flux with the position parameter Ir, it was found that the flow field structure that produces the maximum peak value of heat flux is an unknown type, which appears between the type IV and type III shock-shock interactions. We named it type IIIa shock-shock interaction. By analyzing its flow field structure, we found that it forms two supersonic jet structures, separated by a subsonic flow structure downstream of a Mach stem. When the upper jet shrinks, it will locally form a large thermal load when it hits the wall, which reveals the aerodynamic mechanism of the largest thermal load.

4) Shock-shock interaction experiments were carried out with a large-scale wedge-cylinder combined test model, and high-resolution shock-shock interaction schlieren images and high spatial resolution pressure and heat flow distributions were obtained. Due to the sharp peaks of pressure and heat flux caused by Type IV shock-shock interaction and the restriction of the sensor size, we designed a large-scale cylindrical model to install enough heat flux sensors to capture the peak value of heat flux. We conducted a total of nine experiments in two operation conditions of M7 and M8. We obtained schlieren images and pressure and heat flux distributions of different shock-shock interaction patterns. In the schlieren images, the phenomenon of light and dark phases caused by the compression and expansion in the jet is observed. The magnification ratios of pressure and heat flow induced by shock-shock interactions are 3~5 and 6~12 respectively.