Chapter 932 No Way, Is This Really a Steal?

Although there are still some things that are not understood, since Chang Haonan, as the general manager, has made a decision, then what Wei Yongming and Song Jingming have to do next. Naturally, it is to execute according to the original plan.

After sending the two people away, the former sat back at his desk again, staring at the content of the third chapter of the paper on the screen, and read it for a long time-

In fact, so far, the problem of computational divergence caused by high computational weights has not been solved, and even the cause has not been widely recognized.

Chang Haonan also realized the existence of this problem through a lot of practice in the past few years.

But when he summarized the problem to the second point just now, a flash of inspiration suddenly flashed in his mind.

The development of the stress field is mainly affected by the convection term and the deformation term, and evolves rapidly in the form of exponential growth, but the traditional numerical discretization format based on polynomial interpolation cannot accurately capture the law of this exponential growth, which makes the contribution of the convection term and the deformation term of the constitutive equation to the stress evolution cannot be balanced, resulting in computational divergence.

Therefore, theoretically speaking, as long as the exponential growth rate of stress is slowed down in the area with large stress gradient changes during the solution process of the finite volume method, the stability of numerical calculation can be guaranteed.

So, after about a quarter of an hour of brainstorming, Chang Haonan stood up suddenly.

The headquarters of the Aviation Power Group is a relatively simple administrative agency and does not have research functions.

Therefore, even if you have an idea, you can't get the result by sitting in front of the computer in the office.

He first picked up the phone and contacted Zhu Yadan who was waiting on the first floor.

Then he changed his clothes and went downstairs to take a car to the new site of the Torch Laboratory, which has been completed.

There is a supercomputer with better performance than the Beijing Aviation Computing Center, which is enough for Chang Haonan to test whether his idea is correct.

In fact, this work is almost entirely out of Chang Haonan's interest.

Because the theoretical principles and technical routes adopted by the TORCH Multiphysics software are different, it does not face the problem of high calculation weights.

But as a scholar who studies numerical calculation and takes solving the N-S equation problem as a long-term goal, Chang Haonan is not particularly utilitarian in his research in this field.

After all, many seemingly earth-shattering achievements were originally just originated from some seemingly useless curiosity.

After arriving at the laboratory, he went straight into the computing center located in the basement of Building No. 2...

In the computer room, Wang Jie and Qin Haiyang were buried in front of the computer, optimizing the design of the engine nacelle for the Falcon Z project.

At this time, there was a sudden noise at the door.

The two looked up and saw their mentors pushing the door in a hurry.

"?"

Looking at Chang Haonan who suddenly appeared, Qin Haiyang, who was not familiar with the situation of the research group, naturally thought that he was here to check on him.

So, he stood up directly, took out a work log from his schoolbag, sat upright, and prepared to report on the progress of his work.

But Wang Jie, who had been studying for a doctorate in the Torch Laboratory for a year, knew very well that Chang Haonan basically would not do this kind of formalistic work inspection.

Many mentors are keen on this because they often let students do research fields that they don’t know much about, and only through regular reports can they grasp the progress of the project in a timely manner.

And here...

Although the business scope of the Torch Laboratory is wider than that of most research groups, each research direction must be familiar to the teachers themselves.

Especially Chang Haonan, basically only needs to take a look at the calculation program and the calculation results, and then ask a few simple questions, and he can almost judge the current situation of the project.

Wang Jie had already experienced this when he was in Shanghai.

Therefore, he first patted Qin Haiyang on the shoulder to signal the latter to relax, and at the same time took two steps forward to briefly explain the work they were currently doing.

Chang Haonan had no intention of checking the work, and Wang Jie's work efficiency and attitude were very reassuring, so he just nodded and sat in front of another computer terminal next to the computer room, ready to conduct his own research.

However, while waiting for the computer to start up, he still inadvertently glanced at the screen next to him.

"Eh?"

It doesn't matter if you don't look at it, but this look really attracted his attention:

"You are... doing the flow field simulation of the engine nacelle, right?"

In theory, the optimization of the aerodynamic shape of the aircraft is not yet so detailed that it is necessary to focus on the design details of the engine nacelle.

Therefore, this work should be a relatively simple structural optimization problem.

But it is obvious that Wang Jie and Qin Haiyang are not satisfied with this.

"That's right."

Wang Jie scratched his head and replied:

"We have carefully looked at the overall design plan of the Falcon Z project during this period, and found that no matter how much the structure of the engine nacelle is optimized, especially the weight, the impact on the performance of the entire aircraft is negligible."

Chang Haonan thought that it was indeed negligible, otherwise I would not dare to bring it over for you two to practice...

Of course, this must not be said, so he still pretended to listen carefully.

"So, we thought, in addition to the structure, can we make some optimizations in other places."

"Like aerodynamic drag?"

Chang Haonan looked at the nacelle model that had been completed with high-precision meshing, and in his heart he had given Wang Jie a higher evaluation.

However, the latter shook his head:

"At the beginning, I did think about starting from the drag reduction level, but the tail-mounted engine layout has an inherent disadvantage in terms of drag, and the optimization range is really small, and it still does not help the overall performance too much."

"So in the end, we decided to learn from your previous ideas, use the calculation form of flow field combined with sound field, and optimize the cabin by improving the internal structure of the engine nacelle, especially the noise level at the rear of the cabin."

It must be said that although Wang Jie is only in the second year of his doctoral studies, he does have the level of being able to take charge of his own affairs.

It's not just a matter of ability, but also subjective initiative.

Just at this time, the computer in front of Chang Haonan had also entered the supercomputer system, so he did not continue to ask other details, and instead devoted himself to his work...

...

A week passed quickly.

Chang Haonan's research can be regarded as beginning to show signs.

He has found a constitutive equation that does not involve diffusion terms, and based on it, he has achieved convergence processing for convection terms and deformation terms.

It can be said that there is only one step left to solve the problem that is not faced by Torch Group.

But Chang Haonan's doubts have not been alleviated.

Instead, they have become stronger.

You know, he applied a lot of special means in the process of solving this problem.

For example, from a calculation log with divergent results, useful data is extracted through data dimensionality reduction algorithm, and then iterated on this basis to optimize the form of constitutive equation.

In terms of theoretical research, this is a naked unorthodox violent cheat.

It is almost equivalent to doing the geometry multiple-choice questions in the college entrance examination, while others are deducing step by step, and you just take a ruler to measure.

However!

It is not that Chang Haonan has a prejudice against Professor Anthony Jameson.

But this method, I am afraid that no one else in the world can master it except him.

So here comes the question.

How did COMSOL solve this problem?

Could it be that Professor Jameson has thought of a very clever calculation method that even Chang Haonan himself has not discovered?

Or is the former simply lucky and achieved the same effect as high-order data dimensionality reduction through blind guessing?

It seems unreasonable...

But I can't think of any other possibilities.

In short, Chang Haonan began to look forward to the day when the new version of COMSOL Multiphysics was officially released.

However, plans can't keep up with changes.

Before the new product of the friendly company was released, something unexpected happened, interrupting Chang Haonan's work arrangements...

The trouble came from the Hudong factory.

On a Friday afternoon with heavy rain, Lei Zhixing suddenly took a train from Shanghai to Beijing overnight.

"Mr. Chang, there is an emergency."

Lei Zhixing took off his wet raincoat and put it outside the door, then shook off the water drops on his body, and then walked into Chang Haonan's office.

While talking, the former put a sealed file bag on the desk.

Unlike the seal they had pasted last time, this time the seal had a big seal on it, indicating that it contained a formal factory document.

While Chang Haonan was unsealing the seal, Lei Zhixing briefly introduced the situation:

"After Samsung Offshore Engineering snatched the order for 8 LNG ships from us, it acted very quickly and finalized the specific technical plan not long ago."

"Then, we obtained the design draft used by the Koreans for bidding through public channels."

When he said this, Chang Haonan had already taken out the contents of the file bag.

It was a design drawing.

Or strictly speaking, it was a design plan.

It's just that a lot of the content was blacked out, resulting in only the design drawing being relatively complete.

But this was enough.

Because almost at a glance, Chang Haonan saw a set of structures around the liquid cargo tank that he was very familiar with...

"Oh my..."

He was so shocked that his eyes almost popped out:

"Isn't this our BOG reliquefaction system?"

"How dare you steal it?"