Building Heating Control With ML Flow Temperature Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing building heating control systems require complex hardware installations due to numerous adjustable parameters and lack efficient energy management, leading to suboptimal energy consumption.
Innovation Solution
A control system that integrates an energy control unit with a predictive model and machine learning instance to optimize energy consumption by using weather data, flow and return temperatures, and consumption data, allowing for more precise regulation of heating output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a control system uses numerous adjustable parameters to regulate heating output, then the precision of temperature control is improved, but the hardware installation complexity increases
Solution Approach 1:
The patent uses a virtual model (building model) that replicates the thermal behavior of the actual building without requiring physical sensors or measurement devices for every parameter. The virtual model is trained to copy the building's response to heating inputs, allowing the system to infer temperatures and thermal states without complex hardware installations.
Solution Approach 2:
The patent replaces physical measurement devices and complex hardware sensing systems with a computational approach. Instead of using numerous temperature sensors and flow meters to precisely measure system states, the invention uses a machine learning model that calculates these values based on readily available data from simple sensors and building characteristics.
2Use of energy by moving object
If the control system regulates heating output based on indoor temperature forecasts, then the energy efficiency is improved, but the system cannot function properly when radiators are closed
Solution Approach 1:
The patent incorporates feedback from actual building responses to heating inputs. The building model is trained using historical data that includes actual temperature measurements and heating outputs, allowing the system to learn the true relationship between heating actions and thermal responses. This feedback mechanism enables the system to adapt to changing conditions, including when radiators are closed or partially closed.
Solution Approach 2:
The system performs preliminary actions by pre-heating the building before extreme cold periods or when forecast conditions indicate upcoming temperature drops. The building model predicts future thermal states and allows the control system to advance heating actions, storing thermal energy in the building's thermal mass before it is needed, thus maintaining comfort even when radiators are later closed.
3Loss of energy
If the control system optimizes energy consumption internally, then the heating output efficiency is improved, but the optimization of additional energy inputs is insufficient
Solution Approach 1:
The patent creates a universal control system that can optimize multiple types of energy inputs simultaneously. The building model and control algorithm are designed to handle not only heating output optimization but also pre-heating energy inputs, thermal energy storage, and coordination with renewable energy sources. This multi-functional approach allows the system to optimize the entire energy ecosystem rather than just isolated heating operations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A control system (100) for the heating system of a building, in which the heating system uses a heat transfer fluid (5) heated by a heat source (30) and the heat transfer fluid (5) can be introduced into a flow circuit via a circulation pump (20), has a controller (100) to regulate the flow temperature (12) of the fluid to a temperature setpoint (92), wherein the controller has a machine learning instance of a building model (60) which receives weather data (72) and heat quantity data (42) as input vectors and supplies a control system (90) for regulating the flow temperature with a modified flow temperature.