Chapter 134 The core of the problem

Author: September Sauce

Chapter 134 The core of the problem

Chen Nian followed Xu Yuan into the laboratory. Personnel with insufficient confidentiality levels had been taken away, and several core researchers who had been waiting for a long time immediately surrounded them.

"Thank you for your hard work, Mr. Chen!"

"Gong Chen, I'm sorry to trouble you."

"Thank you for your hard work. I even asked you to work overtime at night, but we are really a little confused."

"Yes, I originally planned to find you again tomorrow, but..."

The group of people were chatting away, but no one came to question the accuracy of Chen Nian's model.

Because they knew very well that being able to calculate this model proved that Chen Nian's level was definitely not lower than anyone present.

If there is indeed a problem, it is sporadic.

It may be that the parameters of the model itself are not considered enough, or there may be small deviations in the variable design, but it is definitely not a problem with Chen Nian's logic and philosophy.

This is the way of communication between real technicians. It is extremely simple and pure. The strong will be respected unconditionally.

Chen Nian greeted them one by one and did not bother to exchange too many pleasantries with them. Instead, he directly asked for the simulation data and began to analyze and compare them one by one.

Others stood beside him and explained a few words to him from time to time.

"What we do here is to adjust the injection angle. Under the premise of a fixed 45 degrees and the same outlet cross-sectional area, we used three different groove types: straight groove, tapering type, and gradually expanding type to conduct experiments. Here are the results

It’s still consistent with the model predictions.”

"The problem is this one. After the anti-cross-flow structure was introduced, the data became completely different."

"The impact of the rotation effect on the flow heat transfer in the channel exceeds the expected value, and the final results are very different from the model results."

While listening to other people's explanations, Chen Nian looked at the data carefully.

His expression became more solemn.

This isn't right.

There is no problem with the experimental process, and there is no problem with using the model, but the results really don’t match up!

What's going on?

Is it really a problem with the model itself?

If that's the case, it would be great fun.

After all, the main purpose of this model is to analyze the cooling mechanism and cooling effect of turbine blades under various structural designs.

The data output based on the model will be used as a reference for final R&D and manufacturing.

And once the model is wrong, the direction it leads will definitely be wrong.

So what is the use of this model?

After finally analyzing the technology, doesn’t it mean that we have to go back to the steps of doing physical experiments manually step by step?

Chen Nian put down the data and couldn't help but reach out and scratch his head.

After a while, he reluctantly turned the document to the first page again and looked at it one by one from the beginning.

This time, he didn't even miss every detail in the experimental records.

And just when he saw it, his eyes suddenly lit up.

The researcher on the side noticed his expression and asked:

"How about it, Mr. Chen, can you find the problem?"

Chen Nian shook his head and replied:

"It can be said that it was found, or it can be said that it was not found."

"At present, there is no problem with either the model itself or your experimental process, but the difference between the two exists, which is a difficult thing to deal with."

"I may need more data to verify and troubleshoot the model to confirm whether there are any flaws in the model itself."

Hearing this, everyone's faces showed troubled expressions.

Want data?

Where is the data?

Originally, the biggest problem faced by China's gas turbine design work was the lack of data. Dongqi wanted to cooperate with Motor Sich partly to obtain the historical data they had accumulated.

Now that the cooperation has not been completed and the things have not been obtained, it is really difficult to solve the problem in this way.

Thinking of this, one of the researchers said:

"Gong Chen, this method is indeed feasible, but the problem is that we really have no data."

"Otherwise, we should redo the experiment a few more times?"

"Perhaps there was something wrong with the operation during our experiment, and some uncontrolled variables affected the final results."

Chen Nian shook his head and replied:

"It's no use doing this."

"Even if the same results are measured in the end, such results are random and cannot guide our subsequent design work at all."

“To fundamentally solve the problem, we have to start with the model.”

"In this way, you guys give me the data first, and I will study it when I go back. There is really nothing much that can be done in such a short time."

"If there is any progress, I will notify you as soon as possible."

In fact, at this time, Chen Nian had already vaguely realized what the problem was.

No, there is nothing wrong with the system and there is nothing wrong with the model.

On the contrary, the model output by the system is extremely accurate.

However, this model is a theoretical model output under absolutely ideal conditions.

It will not consider the influence of various conditions in physical experiments, nor the influence of any human factors.

Or use a classic joke to express it:

This thing is a spherical chicken in a vacuum.

Publish a paper? That’s enough.

Theoretical experiments? Enough!

However, if it is used as a guide for physical experiments, its accuracy will inevitably have problems.

Therefore, what Chen Nian wants to do is to complete this incomplete model and add the fault tolerance factors it should have after considering various environmental factors.

This job is not simple.

According to the normal process, it requires a large amount of historical data to support it, which is why Chen Nian made that request in the first place.

But after hearing what the researcher said, he suddenly reacted.

Maybe you don't need this kind of thing.

Because, he has a problem!

I still remember that when he was designing a single crystal furnace, he used constant trial and error to reduce the analysis cost of the single crystal furnace time and time again, and finally reduced the "price" of the system to an extremely low level.

Now, in fact, he can use the same method to find the correct fault tolerance factors!

Thinking of this, he quietly opened the system panel, selected an unfamiliar LM-6000 gas turbine, and then selected the turbine blade module of this gas turbine.

Since he had already completed the analysis of AGT110 before, the analysis consumption displayed by the system at this time was only 0.12 points.

But, just have it!

With only one number, the increment of knowledge can be reflected in the number.

Models are readily available, but models applicable to reality do not yet exist.

That is to say, as long as you fill in the fault tolerance factor, you can definitely accommodate changes in the numbers.

This is considered opportunistic, but...

Isn’t it opportunistic to use a large amount of data to correct the model?

He just moved the process to a virtual system.

Thinking of this, Chen Nian became energetic.

Afterwards, he packed up the data files on the desktop and ran back to his residence without looking back.

The researchers behind him froze on the spot, looking at each other.

"Is this okay?"

One of them asked slightly anxiously.
To be continued...