权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Simulation Technology: The Next Generation

仿真技术：下一代

基本信息

批准号：
EP/D502357/1
负责人：
Richard Wheatley
金额：
$ 18.61万
依托单位：
University of Nottingham
依托单位国家：
英国
项目类别：
Research Grant
财政年份：
2006
资助国家：
英国
起止时间：
2006 至无数据
项目状态：
已结题

来源：
https://gtr.ukri.org/projects?ref=EP%2FD502357%2F1
关键词：
Simulation Technology Next Generation

项目摘要

All the change and activity that we see around us daily, both natural and man-made, is caused by the movement and Interactions of molecules. Molecules are so small (about one millionth of a millimetre across) that we cannot see them, and in order to understand why they behave as they do, we have to use computer simulations: virtual reality experiments in which the activity of individual molecules is predicted using computers, and then magnified millions of times for us to see.Computer simulations are not just interesting visual insights into the molecular world, they are also very useful to scientists. Computer simulations are capable of predicting what will happen in many important situations; for example, whether building materials will fail under extreme temperatures or loads, how poisons will spread, or how effective a new medicine is likely to be. In chemistry, computer simulations are replacing expensive, polluting and dangerous laboratory experiments, and it is also hoped that they will help to reduce the need to test chemicals on animals.When we use computers, we have to remember that they follow our Instructions exactly and pedantically; they cannot interpret what we mean by the instructions, and they cannot judge the results that they produce. This is summed up by the GIGO principle of computing: 'garbage in, garbage out'. If the instructions that we give a computer are defective in some way, the computer will not realise this, but will operate as usual and produce some meaningless output. The danger is that we do not realise that the output is meaningless when we use it; the computer cannot warn us.The instructions which we give to a computer in order to make it perform a simulation Include information about the molecules, such as their sizes and shapes, and also information about the forces between the molecules. These forces can pull molecules together, push them apart, twist them and turn them. The forces are the most important part of the simulation. If the Information about the forces that we give to a computer is 'garbage', so will be the resulting simulation, and any predictions that we make from it.How do we know the forces between molecules? The answer is that we do not, but again computers can come to the rescue. It is, in theory, possible to calculate forces between molecules using computers. Unfortunately, even with the thousands of millions of calculations per second performed by modem computers, our ability to calculate forces which are not 'garbage' is limited to extremely small molecules, without much practical use. In part, the difficulty of calculating forces between molecules is the fault of the method that is often used. The forces between molecules are so weak that the molecules are not even held together firmly at normal temperatures, yet the usual method starts by calculating the forces needed to completely smash the molecules into subatomic particles (something which would only happen at millions of degrees), then to rebuild them in a different way!I shall take an Indirect approach to calculating forces between molecules, by looking at the electric field which is produced by the positive and negative electric charges. in each molecule. This electric field affects the surrounding molecules, and it should be possible to calculate the forces based on this effect. Fortunately, the electric field around a molecule can be calculated quite easily using modem computers. I hope that by improving the forces between molecules which are fed in to computer simulations in this way, the simulations will be more accurate, and the predictions made from them will be more useful to scientists who use molecular simulations In their work

我们每天看到的所有变化和活动，无论是自然的还是人为的，都是由分子的运动和相互作用引起的。分子是如此之小（大约百万分之一毫米宽），我们看不到它们，为了理解它们为什么会这样做，我们必须使用计算机模拟：虚拟现实实验中，单个分子的活动是用计算机预测的，然后放大数百万倍给我们看。计算机模拟不仅仅是对分子世界的有趣的视觉见解，对科学家也很有用。计算机模拟能够预测在许多重要情况下会发生什么;例如，建筑材料是否会在极端温度或负载下失效，毒药如何传播，或者新药可能有多有效。在化学方面，计算机模拟正在取代昂贵、污染和危险的实验室实验，人们也希望计算机模拟有助于减少在动物身上进行化学试验的需要。当我们使用计算机时，我们必须记住，它们严格而迂腐地遵循我们的指令;它们不能解释我们的指令是什么意思，也不能判断它们产生的结果。这可以用GIGO的计算原则来概括：“垃圾进，垃圾出”。如果我们给计算机的指令在某些方面有缺陷，计算机不会意识到这一点，但会照常运行并产生一些无意义的输出。危险的是，我们在使用它的时候没有意识到输出是没有意义的;计算机不能警告我们。我们给计算机的指令是为了让它进行模拟，包括分子的信息，如它们的大小和形状，以及分子之间的力的信息。这些力可以把分子拉在一起，把它们分开，扭曲它们，转动它们。力是模拟中最重要的部分。如果我们提供给计算机的力的信息是"垃圾"，那么由此产生的模拟以及我们从中做出的任何预测也将是"垃圾"。我们如何知道分子之间的力？答案是我们不需要，但计算机可以再次拯救我们。从理论上讲，用计算机计算分子之间的力是可能的。不幸的是，即使现代计算机每秒进行数千万次计算，我们计算力的能力也仅限于非常小的分子，没有太多的实际用途。在某种程度上，计算分子间力的困难是由于常用方法的错误。分子之间的力是如此之弱，以至于分子在正常温度下甚至不能牢固地结合在一起，然而通常的方法是从计算将分子完全粉碎成亚原子粒子所需的力开始（这只有在数百万度的情况下才会发生），然后以不同的方式重建它们!我将采取一种间接的方法来计算分子间的力，通过观察由正负电荷产生的电场。在每个分子中。这个电场会影响周围的分子，应该可以根据这个效应来计算力。幸运的是，用现代计算机可以很容易地计算出分子周围的电场。我希望通过改进分子间的作用力，以这种方式输入计算机模拟，模拟将更加准确，由此做出的预测将对使用分子模拟的科学家更有用。