Predicting SCR Catalyst Resonance for Ammonia Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current selective catalytic reduction (SCR) systems face challenges in accurately predicting the resonant frequency of catalysts, which represents the concentration of reducing agents, leading to ammonia slip and respiratory issues due to over-abundance of ammonia in exhaust emissions.
Innovation Solution
A trained machine-learning model is developed to predict the future resonant frequency of SCR catalysts by acquiring and processing characteristics of internal combustion engines and SCR systems, using machine-learning algorithms such as random-forest or neural-network methods to correlate resonant frequency with reducing agent concentration, enabling precise control of reducing agent injection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If continuously large amount of reducing agent is injected into SCR, then NOx reduction efficiency is improved, but ammonia slip increases causing harmful emissions
Solution Approach 1:
The patent implements a feedback control system that continuously monitors the resonant frequency of the SCR catalyst and uses this information to adjust the reducing agent injection rate. The machine learning model predicts future resonant frequency based on current and historical data, enabling proactive adjustment of injection timing and dosage to maintain optimal ammonia concentration and prevent ammonia slip while ensuring effective NOx reduction
Solution Approach 2:
The system dynamically changes the operational parameters of the SCR system by adjusting the reducing agent injection rate based on predicted resonant frequency. By modifying the injection timing, dosage, and rate as parameters, the system optimizes the balance between NOx reduction efficiency and ammonia slip prevention according to real-time catalyst state
2Object-generated harmful factors
If reducing agent injection is precisely controlled to prevent ammonia slip, then harmful emissions are reduced, but NOx reduction efficiency may deteriorate
Solution Approach 1:
The system performs preliminary action by predicting the future resonant frequency of the SCR catalyst using machine learning models before ammonia slip occurs. Based on these predictions, the control system proactively adjusts the reducing agent injection parameters in advance, preparing the system to maintain optimal operation and prevent both ammonia slip and NOx emission without waiting for actual deviations to occur
Solution Approach 2:
The resonant frequency measurement serves as an intermediary parameter that indirectly reflects the ammonia concentration and catalyst state within the SCR. Instead of directly measuring harmful ammonia slip or NOx emissions, the system uses resonant frequency as a mediator to infer catalyst condition and optimize reducing agent injection, enabling precise control without direct measurement of harmful substances
3Ease of operation
If machine learning model is used to predict resonant frequency, then reducing agent injection is optimized, but system complexity increases
Solution Approach 1:
The machine learning model is trained using historical data from the same SCR system it later controls, enabling the system to self-learn and adapt to its specific operational characteristics. The model uses readily available sensor data (resonant frequency, injection parameters, operating conditions) that the system already collects for other purposes, eliminating the need for additional complex sensing infrastructure while achieving precise prediction and control
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The model effectively predicts the resonant frequency and reducing agent concentration, allowing for optimized SCR system operation that minimizes ammonia slip and reduces harmful emissions, thereby improving air quality and safety.
Implementation Method 1
SCR is a technique for after-treatment of exhaust gases that consists in selectively reducing the NOx into nitrogen via continuous injection of a specific reducing agent in the presence of a catalyst
Implementation Method 2
the reducing agent used consists of an aqueous urea solution that, when it undergoes thermolysis followed by hydrolysis, decomposes into ammonia and carbon dioxide
Implementation Method 3
the reducing agent used consists of an aqueous urea solution that, when it undergoes thermolysis followed by hydrolysis, decomposes into ammonia and carbon dioxide
Implementation Method 4
predicting a resonant frequency of a catalyst for selective reduction of nitrogen oxides (SCR), the resonant frequency being representative of a concentration of a reducing agent within the SCR
Data Source
AI summary
The subject matter of the present invention relates to trained machine-learning models (300), methods (200, 400) and apparatuses (500) allowing a future resonant frequency of a catalyst for selective reduction of nitrogen oxides (SCR) to be predicted, the resonant frequency being representative of a concentration of a reducing agent within the SCR. The SCR forms part of a system for after-treatment of a flow of exhaust gases of an internal combustion engine with which a motor vehicle is provided. The general principle of the invention is based on the observation of correlations between the resonant frequency of an SCR and the concentration of ammonia present within the SCR. This observation led the inventor to envision using machine learning to create a trained machine-learning model in order to predict the resonant frequency of an SCR. In the invention, the trained machine-learning model is a so-called predictive model in which significant correlations are discovered in a set of past observations and in which it is sought to generalize these correlations to cases that have not yet been observed.