Prediction model determination method and device, storage medium and electronic device

By using data acquisition components and machine learning models in urban central heating systems, a prediction model for heating load scheduling evaluation is established, which solves the problem of inaccurate heating load prediction in traditional methods and achieves more accurate heating regulation.

CN119721359APending Publication Date: 2025-03-28HUANENG SONGYUAN THERMAL POWER CO LTD +1
0 Cites 0 Cited by

Patent Information

Application Number
CN202411802639.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-09
Publication Date
2025-03-28

AI Technical Summary

Technical Problem

In the existing urban central heating systems, the traditional heating load prediction and scheduling methods have problems with deviations from the prediction results and actual demand, resulting in inaccurate heating regulation.

Method used

The heating data of different heating equipment is obtained through the data acquisition component, target processing and normalization processing are carried out, the target relationship between independent variables and dependent variables is identified, the data is divided into sub-data sets, and the machine learning model is used for training, and a prediction model for heating load scheduling evaluation is established.

Benefits of technology

It improves the accuracy of heating load prediction, achieves more accurate heating regulation, and meets complex and changing heating needs.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a prediction model determination method and device, a storage medium and an electronic device.The method comprises the steps that heat supply data on different heat supply devices are obtained through data collection assemblies, the data collection assemblies are of multiple types, and the data collection assemblies of each type are arranged on the heat supply devices of the same type; executing the target processing mode on the heat supply data to obtain to-be-analyzed target data; identifying a target relationship between multiple groups of independent variables and dependent variables in to-be-analyzed target data, and dividing the target data into multiple sub-data sets based on the target relationship; and training a preset original machine learning model by using the plurality of sub-data sets to obtain a target prediction model for heat supply load scheduling evaluation, the input quantity of the original machine learning model during training being an independent variable, and the corresponding output quantity being a dependent variable having a target relationship with the independent variable. The problem that load prediction is inaccurate due to the fact that the correlation degree between a prediction model and an actual application scene is low is solved.
Need to check novelty before this filing date? Find Prior Art