世界上第一颗氢弹来自数学建模——仿真

一本《Principles Of Mathematical Modeling Ideas, Methods, Examples》，下载，来头不小。作者A.A. Samarskii ，苏联院士！

上来就说数学建模复兴的第二个原因是爆了核弹！The second was an unprecedented social problem - the Soviet and American national nuclear arms programs, which could not be realized by traditional methods. Mathematical modeling solved that problem: both nuclear explosions and flights of rockets and satellites were previously tested within the computers using mathematical models, and only later were realized in practice.

再看作者，Founder of the Institute of Mathematical Modeling, Moscow, Russia。俄罗斯数学建模中心的founder——铁帽子王。伟大苏联的院士，优秀的我党成员，拿过列宁导师的奖，社会主义劳动英雄，列宁勋章、国家勋章和罗蒙诺索夫奖章获得者，荣获三枚列宁勋章、十月革命勋章、红旗勋章、俄罗斯人民友谊勋章、一级光荣勋章和卫国战争勋章，并荣获“伟大卫国战争中战胜德国”勋章、“保卫莫斯科”勋章等多项勋章。真正的铁帽子王！

更牛逼的是扛过枪，打过二战，踩过地雷，受过重伤，1941年。On December 12, I was on the scout and got at the land mine. They pulled out more than thirty fragments from me - it was a lot of operations. However, the eight fragments were left in me, the surgeons could not get them out. 身体里取了30多块弹片，到那时还留了8块取不出来的弹片。

一篇访谈里写他介绍苏联核弹的历史。媒体是这家：Literary Gazette, 2001

1948年6月10日，伟大导师斯大林亲自签字：铁帽子王A.A. Samarskii跟未来的苏联氢弹之父萨哈罗夫要关禁闭，搞核武器。- could only claim the rooms that they have received.

然后A.A. Samarskii的任务是建立一个原子弹的数学模型，没用计算机，找了几个妇女同志进行运算，获得计算结果，跟最后爆了的炸弹比较起来结果差异是30%！30%好像差得很远，不是的，准确度不低了，但是竞争对手美帝的误差从来没在30%以内过！

为啥干得比美帝漂亮？米帝的计算工作是物理学家操刀的，这是根本区别。大佬请了30个妹子来并行解题。两个月算出结果——原子弹研制第一年最重要的成就。这是1953年。

两年后，大佬就提出了一个更精确的数学模型，研究数值方法的理论，比美帝领先。美帝的资料也承认，在这些复杂的物理过程的数学计算上面，苏修干得更漂亮。

大佬从50年代开始做这个工作，到80年代就把核弹相关的问题基本上搞清楚了。

diss氢弹之父萨哈罗夫。牛逼但是认他是氢弹之父不服，许多与萨哈罗夫共事的物理学家的功劳就都要XXX拉！

关于科学、数学问题：

世界是非线性的，非常难准确预测的！对基础研究的认同度不高，应该先有需求，再有研究。不是先研究再找应用。

重点：。在工程技术领域应用数学建模及其相关的信息化工具需要审慎的智力思考和严谨的组织工作。跟发达国家比起来，俄罗斯在这方面的滞后现象或许比基础科学领域更令人担忧。

西方国家正在大规模部署数学建模和计算机仿真技术。汽车公司购买超级计算机来对车辆的整体设计进行计算，尤其是汽车事故的分析计算，基本上已成为一种常态。这是一项非常有价值的业务，因为发生“事故”的只是数学模型，而不是昂贵的真实的车辆。没有相应的仿真能力的公司将失去竞争力……欧洲“工业数学”联盟应运而生。其核心目的是行业内正确有效地应用数学建模并知道相应的任务清单。

在此背景下，俄罗斯的专家在微电子、仪器仪表、激光和材料热处理等技术领域积累的独特数学建模经验几乎没有得到应用。

大学里要重视数学建模，国家更加要重视数学建模

On June 10, 1948, the Decision of the Council of Ministers of the USSR №1990-774ss/op "On additional work plan of special research for 1948" was issued and was signed by Stalin and stamped "Top Secret (Special Files)". In the 9th paragraph it refers to the provision in "a matter of urgency" of apartments and rooms to several scientists. At that time, the candidate of geophysical sciences A.A. Samarskii and the candidate of physical and mathematical sciences A.D. Sakharov could only claim the rooms that they have received. But most importantly, according to this resolution, they entered the very narrow circle of people who were entrusted to "perform calculations on "PO" for RDS-1, RDS-2, RDS-3, 4-RDS, RDS-5 with different variants of the equation of state ..."

For the uninitiated, I will explain what the abbreviation "RDS" is hiding an atomic bomb, and the "PO" - was the central part of it made of Pu-239 ...

It is how it would happen, but this time a secret decree of the Central Committee came out on the establishment of a mathematical laboratory for solving problems related to the creation of the atomic bomb. It was some meeting, "on the Higher Levels", where Tikhonov proposed a calculation of the atomic bomb ... By the way, the meeting was attended by Landau, who said that "if this can be done, it will be a scientific feat!" Nevertheless, Tikhonov's offer was accepted, and there appeared a tiny laboratory in which there were only a few mathematicians. And about thirty girls-solvers who had graduated from the geodesic institute.

They were instead of computer?

Yes... And we were given a task: create a "numerical model of atomic bomb".

– It was then when the Special Committee had given for you and for candidate Sakharov a room for each?

– That's right, because I had nowhere to live ... However, the decision was carried out only at the end of 1950. Since I was single, not very legitimately, I continued to live in the dormitory of the University, and then began to rent an apartment ... However, I spent at work most of the time – the time period for us to make the job was allotted very small: only about a year! And this problem was of the highest level of difficulty, and besides, the physicists had quite inaccurate data... Their models were very rough, approximate... They worked only with such models... And together with Tikhonov, we agreed that I will work with the exact models.

– Apparently, you have come to the finish line at the same time?

– Yes, at the time of the first test of our bombs, we already had the first results... The error was only 30 percent...

– Only?!

– It's a great result! I do not know what it is now, but the Americans had never had errors less than 30 percent. Thus, our calculations were very accurate ... In the future, we have reduced the error to 10 percent...

– How did you manage it?! I think the Americans had always better computer technology?

– Our initial data was taken correctly. We tried to maintain the correct mathematical description of the physical process, and in this, I am convinced, I was helped by the fact that my first education was physical.

– So, math is in the first place after all?

– And that is what has defined our success. While in Los Alamos, the calculations were made by physicists. This is a fundamental difference ... But how to solve the resulting equations? I am proud that I came up with "parallelization". There were 30 girls-solvers. There were several hundred equations. That is about ten equations for each girl ... They calculated as if independently, but passing their data to each other ... Of course, I explain it simplified, but the idea of the method, it seems to me, is clear ... The "parallelization of calculations" enabled us to make the calculations for two months, we have accelerated the process of work by about 15 times ... I consider it was the most important achievement in the first year of work on the atomic bomb.

– Everyone who started working on the bomb, were young, so there were a lot of such "nuclear families" ... That time, of course, comes to mind with good feelings, although it was very difficult, since at the first stage, we worked with primitive computer technology... But it was very interesting, it was a creative work. Numerical methods were improving rapidly: just two years later, I proposed a more accurate mathematical model... Until 1953, we used manual technique and we advanced quite far in this area... I immediately realized that it is necessary to work on the theory of numerical methods, and it was rightly so, as it was possible to promote special calculation methods. By the way, the Americans have lagged behind in this area - they were hoping for the computer equipment and failed.

– They recognized it: they later confirmed that, despite the strong backlog in computers, we have not conceded to them in the main point: in the calculation of complex physical processes that occur in the explosions of atomic and thermonuclear bombs ... You named the date: 1953. Really you were not dealing with arms after that?!

– I deal with it all my life... The Institute of Applied Mathematics was created in 1953 and our laboratory became a department.

– The director was Mstislav Keldysh, wasn't he?

– He was the director, Tikhonov - the deputy director, me - the head of a department. Our department was the largest in the institute. A little later, Okhotsimsky was made the head of another department. That is cosmos...

– And it was one of the achievements of which I am proud. If we turn to the same bomb, the scheme looked something like this. There was some separation of work between the calculation groups. First, they calculated the initial compression process - a kind of preparation for the explosion, and then the data and the calculations were passed to our department, where we calculated all the processes associated with the explosion... It is curious that the task was written right in my office. For example, Sakharov came and then on my desk gave us a task... By the way, I passed recently to Sarov, i.e. to Arzamas-16, my notebook in which Sakharov and Zeldovich and Babaev kept records.

– And how long did you work for the "atomic project"?

– Very actively - up to about 1980. Then we cooperated only sporadically, when there was a need ... In fact, by that time all the fundamental problems were solved.

– You said: "The problems have been solved"... That is what determined the fact that eminent scientists - Sakharov and Zeldovich - left Arzamas-16. By the way, what is your impression of them?

– Sakharov, no doubt, a remarkable man, devoid of any complexes. He combined talents of visionary physicist and mathematician. But he was credited with the definition of "father of the hydrogen bomb," and this is not true, as it offended many physicists who worked with Sakharov. And there were no need for this, because the Sakharov did not need exaltation - by himself he was an outstanding scientist and a man.

– And what about Zeldovich?

– He was unique in his own way ... "reactive type", he was ready to pounce on any problem, but he lacked a general mathematical culture. He grasped the idea quickly, but he was still scattered in his character... And at the same time, was easier to communicate with him than with Sakharov. I would like to say about the theorist Yuri Romanov. He had long since earned the right to be a member of the Academy, but he was not even elected into corresponding members. It's not fair! .. Unfortunately, this applies to many. For example, to Feoktistov, Shelkin...

Reflections on science. The world is nonlinear, that is, the basic laws of development of animate and inanimate nature (from the micro to the macro world), including the social and economic structures, are nonlinear. This means in particular that there are several possible ways of evolution of a complex object, that is the future is non-uniquely determined by the preset time (by the initial conditions), and it can not be predicted based only on prior experience. The optimal path of the evolution should be chosen based on the knowledge of the laws of its development, it is necessary to calculate and control it. This is a complex and difficult task, but the life requires its solution.

Reflections on science. We should not base on the belief that science will evolve spontaneously to meet its internal needs of self-development and self-organization. Science must perform urgent social order, promoting the scientific and technological progress not in the distant future, but today. We should not use such model (which has a considerable number of supporters): first conduct basic research, and then look for where they can be used. We need to find ways of development of science in a given direction, associated with the solution of certain large problems. Apparently, the methods of resource management (material and human) can be applied to this. It is important to remember that all problems should be resolved quickly and at a high scientific level. The required level of applied research is only possible on the basis of fundamental research, which is oriented in character. In connection with the development and application of computer technology, a special responsibility is on the math. Modern applied mathematics must, performing the social order, to decide "what is necessary" and "how it is necessary".

Reflections on science. Using the benefits of mathematical modeling and the informatics tools that are based on it in the technological applications requires serious intellectual and organizational effort. Symptoms of our backlog in this area from the developed countries, perhaps are more alarming than in the basic sciences. In the West, there has been a move to mass deployment of mathematical modeling and computer simulations in technology. The purchases of supercomputers by automotive companies become typical to calculate the total vehicle design, particularly in the situations of accidents. This is a very profitable business, as "accidents" involve mathematical models and not expensive vehicles. The companies that do not have corresponding calculating techniques become uncompetitive... European consortium "Mathematics in Industry" was created. Its goal - the effective use of mathematical modeling in the industry and the development of corresponding directory of tasks. Against this background, there are almost no use of the accumulated unique mathematical modeling experience of our experts in some of the technologies of microelectronics, instrumentation, laser and thermal processing of materials.