Boiler Combustion Learning Data Generation via Automated Pre-processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating learning data for combustion optimization in coal-fired power plants are cumbersome and require skilled users, leading to increased costs and reduced combustion efficiency due to inefficient exhaust gas treatment and incomplete combustion.
Innovation Solution
An apparatus and method that includes a data pre-processor to restore abnormal signals, filter data, and erase outliers, and a data analyzer to cluster and sample data for deriving learning data, which is then used to generate a combustion model and controller for optimizing boiler operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual data processing operations are performed to generate learning data for combustion optimization, then combustion optimization can be achieved, but the process becomes cumbersome and requires skilled users
Solution Approach 1:
The system performs data processing automatically without requiring skilled user intervention. The automated pipeline includes data collection from multiple sensors, automatic anomaly detection and handling, data cleaning, feature extraction, and model training, all executed autonomously by the system components.
Solution Approach 2:
Manual mechanical data processing operations are replaced with automated computational algorithms. The system uses computer-based anomaly detection algorithms, data cleaning algorithms, feature extraction algorithms, and machine learning models to automatically generate learning data for combustion optimization.
2Reliability
If comprehensive data processing is performed to ensure accurate learning data, then combustion optimization improves, but the complexity of the system increases
Solution Approach 1:
The data processing system is divided into distinct functional modules: data collection module, anomaly detection module, data cleaning module, feature extraction module, and model training module. Each module handles a specific aspect of data processing independently, making the complex system manageable and maintainable.
Solution Approach 2:
The system introduces intermediate processing layers between raw sensor data and the final combustion optimization model. These intermediaries include feature extraction layers that transform raw data into meaningful features, and data cleaning layers that remove anomalies, thereby simplifying the overall system architecture.
3Productivity
If skilled users process data manually, then data can be processed, but operational costs increase
Solution Approach 1:
The system eliminates the need for skilled user labor by performing data processing autonomously. The automated pipeline handles all data processing tasks from collection to model training, reducing operational costs associated with skilled personnel while maintaining high productivity.
Solution Approach 2:
Manual skilled labor is replaced with automated computational systems. The system uses computer algorithms and machine learning models to perform data processing tasks that previously required skilled human operators, thereby reducing operational costs while maintaining or improving productivity.
Data Source
AI summary
An apparatus and method for generating learning data for combustion optimization is provided. The apparatus includes a data pre-processor to collect raw data including currently measured real-time data for boiler combustion and previously measured past data for the boiler combustion, and to perform pre-processing for the collected raw data, and a data analyzer to derive learning data from the raw data by analyzing the raw data. An apparatus for combustion optimization includes a management layer to collect currently measured real-time data for boiler combustion, to determine whether to perform combustion optimization, and to determine whether to tune a combustion model and a combustion controller; a data layer to derive learning data from raw data; a model layer to generate the combustion model/controller through the learning data; and an optimal layer to calculate a target value for combustion optimization and to output a control signal according to the calculated target value.