Method for controlling the pressure of the heat transfer fluid in a heating system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing heating systems face issues with excessive pressure drops in heat transfer fluids, leading to blocking conditions and maintenance challenges, even when pressure values are within acceptable ranges but slowly decreasing over time.
Innovation Solution
A control method using a machine learning classification model, trained on historical data, predicts the likelihood of boiler blockage due to pressure drops within a future time window. This method processes pressure values and additional parameters from existing components in heating systems, providing notifications for maintenance interventions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the boiler operates with pressure values below the standard range, then the system can continue running without immediate blocking, but progressive pressure reduction leads to frequent blocking errors and maintenance issues
Solution Approach 1:
The system performs preliminary actions by monitoring pressure trends and predicting future blockages before they occur. The machine learning model analyzes historical pressure data to identify progressive pressure reduction patterns, enabling early warning notifications that prompt preventive maintenance interventions, thus avoiding frequent blocking errors and ensuring continuous operation
Solution Approach 2:
The system implements feedback by continuously monitoring pressure values and using machine learning algorithms to analyze trends. The model processes historical data and provides feedback through predictive warnings about potential blockages, allowing the system to adjust maintenance schedules and prevent operational disruptions before they happen
2Measurement precision
If traditional pressure monitoring is used with simple threshold warnings, then the system responds to critical low pressure, but it fails to predict progressive pressure drops that occur before critical thresholds are reached
Solution Approach 1:
The system changes the monitoring approach from static threshold detection to dynamic trend analysis. The machine learning model processes multiple pressure parameters over time, analyzing rate of change, acceleration of pressure drop, and other temporal patterns to predict future blockages, thereby recovering the lost trend prediction information while maintaining precise threshold detection
Solution Approach 2:
The system performs preliminary prediction actions by analyzing pressure trends before critical thresholds are reached. The machine learning model forecasts future pressure values and issues early warnings, enabling preventive maintenance before the system enters blocking conditions, thus addressing both precise detection and trend prediction requirements
3Reliability
If additional devices are added to monitor and predict pressure issues, then prediction accuracy improves, but system cost and complexity increase
Solution Approach 1:
The system implements self-service by using its existing pressure monitoring infrastructure and control unit to perform predictive analysis. The machine learning model processes data already being collected by the boiler's standard sensors and control systems, eliminating the need for additional monitoring devices while maintaining high prediction accuracy and reducing system complexity
Solution Approach 2:
The system applies multi-functionality by enabling the existing control unit to perform both traditional pressure threshold monitoring and advanced predictive analysis. The same hardware infrastructure serves multiple purposes: real-time pressure detection, historical data storage, trend analysis, and predictive warning generation, thereby avoiding additional devices while improving reliability
Data Source
AI summary
A method for controlling the pressure of heat transfer fluid in a heating system including a thermo-sanitary appliance, in particular a gas boiler, the method including the following steps. Extrapolating minimum values of the fluid pressure from the original minimum values, within a time observation window corresponding to a given number of previous days. Identifying one or more ramps of the minimum values of the pressure of the fluid, within the time observation window. Calculating a plurality of characteristics relating to the trend of the pressure of the fluid on the basis of said one or more ramps of the minimum values. Providing the calculated characteristics to a machine learning classification model, adapted to produce a three-class classification, each of which is representative of the more or less high probability that in a future time window the thermo-sanitary appliance will block due to an excessively low value of fluid pressure.