第四、五章研究了一类非线性抛物方程熵解的存在性。
In Chapter 4, 5, an existence result of entropy solutions to a class of nonlinear parabolic problems is established.
同时利用信息论中的不等式，直接地证明最小交互熵解就是对偶几何规划解；
Then, using the inequality of information theory, the paper directly proved that the minimum cross-entropy solution is exactly the dual geometric programming solution.
另外在第四章中，我们还研究了极端相对论方程组熵解的非相对论整体极限问题。
Moreover, in Chapter 4, we also consider the non-relativistic global limits of entropy solutions to the extremely relativistic Euler equations.
然后在此基础上，我们讨论了上述熵解同拟一维流问题的解在L2意义下的相近程度。
Then based on it, we give an estimation on the difference between this local entropy solution and the solution of the quasi-one-dimensional problem in L2 norm.
本文证明了带有L1资料和对以散度形式的低阶项没作任何增长假设的某类抛物问题熵解的存在性。
In this paper, an existence result of entropy solutions to some parabolic problems is established. The data belongs to L1 and no growth assumption is made on the lower order term in divergence form.
介绍了逼近理想解排序法—TOPSIS算法，同时利用熵值法确定指标的权重，并详细描述了该方法在优选配煤方案中的应用。
The application of TOPSIS method in seeking the optimum mixture ratio of blended coal is described in much detail by confirming the weight of index with entropy.
提出了一类解极小极大问题的熵函数法，这种方法也可用来解线性或约束优化问题。
A method for solving minimax problem is presented, which also can be used to solve linear or constrained optimization problems.
本文导出的熵变量形式QPNS方程具有对称性和自动满足热力学第二定律，这将提高解的稳定性。
The QPNS equations in entropy variables derived in the present paper have the symmetrization and satisfy the second law of thermodynamics automatically that can improve the stability of the solution.
并且给出了一个新的以数值解的二阶差分作为基本模块的熵耗散函数。
A new entropy dissipator, based on the second-order difference of the numerical solution is given.
该算法直接在基因的层面上进行优化，能学习劣解的基因，并用信息熵作为结束条件的判据。
GOA optimizes directly at the gene level and can learn from the gene of bad individuals. The entropy is used for the terminal criterion of the algorithm.
该算法直接在基因的层面上进行优化，能学习劣解的基因，并用信息熵作为结束条件的判据。
GOA optimizes directly at the gene level and can learn from the gene of bad individuals. The entropy is used for the terminal criterion of the algorithm.
应用推荐