Quantile regression neural network is used for probability prediction

Hello, brothers

As a beginner in photovoltaic prediction, I used the layers neural network to make probability prediction, which produced the result as shown in the figure. I found that the prediction result fluctuated frequently, as did the generated probability interval. What might be the cause?

Input time step = one day PV

Output time step =1

You did not attach any images, just the “Uploading…” message.

Thank you for reminding me. It is my first time to use the forum and I am not familiar with it. I have uploaded pictures。

And TCN (time convolutional neural networks) why does it work this way, what could be the reason for that。