Atmospheric and vacuum product property prediction method and system, equipment and storage medium
A prediction method, atmospheric and decompression technology, applied in neural learning methods, design optimization/simulation, biological neural network models, etc., can solve problems such as poor model generalization ability, single basic data, and insufficient data feature learning, etc. The effect of fast computation, improved generalization performance and robustness
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0049] Such as figure 1 As shown, a method for predicting properties of atmospheric and vacuum products provided in this embodiment, specifically, includes the following steps:
[0050] 1) Based on the pre-obtained sample data set, train the pre-built machine learning models, and fuse the trained machine learning models to obtain the final atmospheric tower fusion model;
[0051] 2) Using the final atmospheric tower fusion model to predict the properties of the atmospheric and vacuum products, and obtain the prediction results of the properties of the atmospheric and vacuum products.
[0052]Preferably, in the above-mentioned step 1), the method for obtaining the final atmospheric column fusion model comprises the following steps:
[0053] 1.1) Data set acquisition: Collect on-site production data and process simulation data for different typical working conditions, and aggregate them to form a sample data set.
[0054] 1.2) Machine learning model selection: select several m...
Embodiment 2
[0077] This embodiment provides a method for machine learning atmospheric tower prediction model based on model fusion, including:
[0078] 1) Sample data set generation, according to the actual process flow of a refinery, establish process simulation process models of different typical working conditions, and use these typical working conditions as the fulcrum, apply process simulation technology to generate 1000 sets of different Process simulation data, and then aggregate the process simulation data and 1000 sets of data actually produced on site to form a sample data set.
[0079] 2) Data preprocessing, the obtained sample data set is randomly divided into two parts of training set and test set according to 4:1, and the input and output variables of each machine learning model are determined, wherein:
[0080] Input variables: naphtha flow at the top of the tower, flow in the middle of Chang No. 1, heat load in the middle of Chang No. 1, flow in the middle of No. 2, heat l...
Embodiment 3
[0094] Embodiment 1 above provides a method for predicting properties of atmospheric and vacuum products. Correspondingly, this embodiment provides a system for predicting properties of atmospheric and vacuum products. The recognition system provided in this embodiment can implement a method for predicting properties of atmospheric and vacuum products in Embodiment 1, and the recognition system can be realized by software, hardware or a combination of software and hardware. For example, the system may include integrated or separate functional modules or functional units to execute corresponding steps in the methods of Embodiment 1. Since the identification system of this embodiment is basically similar to the method embodiment, the description process of this embodiment is relatively simple. For relevant parts, please refer to the part of the description of Embodiment 1. The embodiment of the system of this embodiment is only schematic .
[0095] Such as image 3 As shown, a...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


