Multi-step prediction method for ultra-supercritical boiler heating surface pipe wall overtemperature early warning

Abstract

The invention relates to a multi-step prediction method for ultra-supercritical boiler heating surface pipe wall overtemperature early warning, which comprises the following steps: step 1, collectingtraining data in a training stage, preprocessing the collected training data, then constructing a time sequence network model, and carrying out offline training; and step 2, acquiring input data as input of the time sequence network model in a use stage to obtain a prediction result of the metal wall temperature data. The method has the beneficial effects that the pipe wall temperature of the heating surface can be predicted in advance, and guidance is provided for over-temperature control; longer time sequence feature information can be captured, and the prediction precision is improved; timeis subjected to sinusoidal encoding, so that transition of each day is smooth, and the actual operation condition is better met; the problem that the sensor cannot work for a long time due to directmeasurement can be solved, and wall temperature prediction at different positions can be realized; latest data can be conveniently added, the model is retrained, and good anti-interference performanceis achieved.

Description

technical field [0001] The invention belongs to the technical field of boiler safety, and in particular relates to a multi-step prediction method for early warning of overtemperature of a tube wall on a heating surface of an ultra-supercritical boiler. Background technique [0002] With the continuous improvement of power demand and energy saving and emission reduction requirements, the steam temperature and pressure of boilers are constantly increasing. Ultra-supercritical boilers and superheaters work in harsh environments, which are more likely to cause over-temperature tube explosion. [0003] The temperature of the tube wall on the heating surface of the ultra-supercritical boiler is high, the flue gas flow rate is large, and the working conditions are harsh, so the temperature sensor cannot measure the temperature of the metal tube wall for a long time, and the temperature change of the tube wall is hysteretic, nonlinear, non-stationary, and strongly coupled. character...

AI Technical Summary

Problems solved by technology

The patent No. 201010183756.6 is equipped with a series of thermocouple sensors on the pipeline to collect the boiler heating surface wall temperature signal, which optimizes the sensor layout; however, the internal environment of the ultra-supercritical boiler is harsh and the temperature is high, and the sensors cannot work for a long time and can only monitor current wall temperature

Patent No. 201611123426.1 collects flue gas and steam-water variables through the boiler DCS system or SIS system, and calculates the metal wall temperature through empirical formulas. This method has a large error in calculating the wall temperature and cannot be predicted in advance

Images