CN-122015114-A - Method and device for early warning combustion state of opposed firing boiler and computer equipment

CN122015114ACN 122015114 ACN122015114 ACN 122015114ACN-122015114-A

Abstract

The application discloses a method and a device for early warning of the combustion state of a opposed firing boiler and computer equipment. The method comprises the steps of obtaining multi-source temperature data in the opposite-flow combustion boiler and parameter characteristics of left and right side flues of the opposite-flow combustion boiler, determining wall heating characteristics based on wall temperature distribution data, clustering target monitoring points in the opposite-flow combustion boiler by adopting a DBSCAN spatial clustering algorithm based on the temperature distribution data in the boiler, positioning flame centers of flame areas, determining behavior characteristics of flame center movement tracks based on attribute characteristics of the flame centers in continuous time frames, matching the behavior characteristics, the wall heating characteristics, the parameter characteristics of the left and right side flues of the opposite-flow combustion boiler with early warning conditions, and carrying out slagging, partial burning or overtemperature early warning according to matching results. Not only can automatically identify a plurality of flame centers and reduce the misjudgment rate of the number of the flame centers, but also can solve the problem of insufficient early warning accuracy and improve the operation safety of the boiler.

Inventors

WANG YINGMIN
ZHANG WENJING
LI BO
GAO CHUNYU
YIN XIAOYAN
ZHANG ZHIGANG
HU JIAMING

Assignees

中国大唐集团科学技术研究总院有限公司
中国大唐集团科学技术研究总院有限公司华北电力试验研究院

Dates

Publication Date: 20260512
Application Date: 20260129

Claims (10)

1. The method for early warning the combustion state of the opposed firing boiler is characterized by comprising the following steps: Acquiring multi-source temperature data in a opposed firing boiler and parameter characteristics of left and right side flues of the opposed firing boiler, wherein the multi-source temperature data comprises temperature distribution data in the boiler and wall temperature distribution data; Determining a wall heating characteristic based on the wall temperature distribution data; Clustering each target monitoring point in the hedging combustion boiler by adopting a DBSCAN spatial clustering algorithm based on the temperature distribution data in the boiler, positioning the flame center of a flame area, and determining the behavior characteristics of the flame center movement track based on the attribute characteristics of the flame center in a continuous time frame; And matching the behavior characteristics, the wall heating characteristics, the parameter characteristics of the left and right side flues of the opposite firing boiler with early warning conditions, and carrying out slag formation, partial firing or over-temperature early warning according to the matching results.
2. The method for pre-warning the combustion state of a hedging combustion boiler according to claim 1, wherein the clustering of each target monitoring point in the hedging combustion boiler by using a DBSCAN spatial clustering algorithm based on the temperature distribution data in the boiler to locate the flame center of a flame region comprises: Acquiring three-dimensional time sequence data of a flame area in the opposed firing boiler, and screening out the target monitoring points with the temperature higher than a first temperature threshold value; Clustering the two-dimensional coordinates of the target monitoring points by adopting a DBSCAN spatial clustering algorithm, and taking an effective cluster obtained by clustering as the flame center; And calculating a weighted center coordinate by taking the temperature value of the target monitoring point in the effective cluster as a weight, and taking the weighted center coordinate and the highest temperature of the target monitoring point in the effective cluster as attribute characteristics of the flame center.
3. The method of claim 1, wherein determining the behavior feature of the flame center motion profile based on the attribute feature of the flame center in the continuous time frame comprises: constructing a time sequence data set of attribute characteristics of the flame center under continuous time frames; Calculating the total similarity between flame centers in adjacent time frames through a space-temperature weighting algorithm based on the time sequence data set; A cost matrix is constructed based on the total similarity, the cost matrix is solved to obtain matching pairs of flame centers in adjacent time frames, and effective matching pairs of which the total similarity meets a similarity threshold are screened out from the matching pairs of flame centers in the adjacent time frames; creating and marking a motion trail by taking a flame center in a first time frame in the time sequence data set or a flame center which does not form the matched pair as a starting point of the motion trail; linking the flame center of the current time frame in the effective matching pair to a motion track corresponding to the flame center of the previous time frame so as to update the motion track; Extracting the behavior characteristics of the motion trail after all flame centers are traversed; The behavior characteristics comprise total track displacement, unit time displacement, average temperature, highest temperature, temperature fluctuation variance and continuous track time frame number.
4. The method of claim 3, wherein the attribute features include weighted center coordinates of flame centers and a maximum temperature, and the calculating the total similarity between flame centers in adjacent time frames by a space-temperature weighting algorithm based on the time series data set includes: calculating the space Euclidean distance of the flame center in the adjacent time frame based on the weighted center coordinates; Normalizing the space Euclidean distance by taking the maximum dimension of the hearth as a reference to obtain a space distance similarity; Calculating the temperature relative deviation of flame centers in adjacent time frames based on the highest temperature to obtain the temperature similarity; And carrying out weighted summation on the spatial distance similarity and the temperature similarity to obtain the total similarity.
5. The method for pre-warning the combustion state of a opposed firing boiler according to claim 1, characterized in that, The wall heating characteristic comprises at least one of a temperature difference between a left water cooling wall and a right water cooling wall of the opposed firing boiler, a temperature difference between flue gas at the left side and the right side of the low-temperature superheater, a temperature difference between flue gas at the left side and the right side of the high-temperature superheater and a temperature difference between flue gas at the left side and the right side of the high-temperature reheater; The parameter characteristics of the left and right side flues of the opposed firing boiler comprise at least one of carbon monoxide concentration deviation, oxygen content difference, flue gas flow and flue negative pressure.
6. The method for pre-warning the combustion state of a opposed firing boiler according to claim 1, characterized in that, The early warning condition for triggering the slag formation early warning comprises at least one of the following: S1, the distance between the flame center and the geometric center of the opposed firing boiler is smaller than a safe distance threshold, and the difference value of the temperature of the inner wall surface of the preset duration is larger than or equal to a second temperature threshold; S2, the maximum value of the wall surface temperature is within the slagging temperature range; S3, the fluctuation amplitude and frequency of the flue negative pressure are larger than corresponding threshold values; the early warning condition for triggering the partial burn early warning comprises at least one of the following: S4, the average offset distance of the motion trail to the single side direction of the hearth is larger than or equal to a distance threshold value, and the duration of continuous offset is larger than or equal to a duration threshold value; s5, the temperature difference of the water cooling wall surfaces at the left side and the right side of the opposed firing boiler is larger than or equal to a third temperature threshold value; S6, the temperature difference of the flue gas at the left side and the right side of the low-temperature superheater is larger than or equal to a fourth temperature threshold value; S7, the difference of the oxygen content of the left flue and the right flue of the opposed firing boiler is larger than or equal to an oxygen content threshold value; S8, carbon monoxide concentration deviation is larger than or equal to a carbon monoxide concentration threshold value; The early warning condition for triggering the overtemperature early warning comprises at least one of the following: s9, the occupation ratio of the overtemperature point of the regional water wall is larger than or equal to a proportional threshold; S10, the temperature difference of the flue gas at the left side and the right side of the high-temperature superheater and the temperature difference of the flue gas at the left side and the right side of the high-temperature reheater are both larger than or equal to a fifth temperature threshold value.
7. The method for pre-warning of the combustion state of a opposed firing boiler according to claim 1, further comprising: if slag bonding, partial burning or over-temperature early warning is triggered, based on an early warning result and corresponding behavior characteristics, wall heating characteristics and parameter characteristics of left and right side flues of the opposed firing boiler, the combustion parameters and soot blowing operation parameters of the opposed firing boiler are adjusted to correct a combustion field.
8. The method for pre-warning the combustion state of the opposed firing boiler according to claim 1, wherein the acquiring the multi-source temperature data in the opposed firing boiler and the parameter characteristics of the left and right side flues of the opposed firing boiler comprises: Acquiring temperature distribution data in the furnace through acoustic wave propagation signals acquired by an acoustic sensor array, wherein the acoustic sensor array is symmetrically arranged along the circumferential direction of a hearth; acquiring wall temperature distribution data through a fiber bragg grating wall temperature sensor, wherein the fiber bragg grating wall temperature sensor is arranged on a key heating surface, and the key heating surface comprises a water cooling wall, a low-temperature superheater, a high-temperature superheater and a high-temperature reheater; And respectively collecting left and right gas content and flue negative pressure of the opposed firing boiler through a gas sensor and a pressure sensor which are arranged in the flue, and respectively calculating carbon monoxide concentration deviation and oxygen content difference based on the gas content.
9. An early warning device for the combustion state of a opposed firing boiler, the device comprising: The acquisition module is used for acquiring multi-source temperature data in the opposed firing boiler and parameter characteristics of left and right side flues of the opposed firing boiler, wherein the multi-source temperature data comprises temperature distribution data in the boiler and wall temperature distribution data; A data processing module for determining the heating characteristics of the wall surface based on the wall surface temperature distribution data, and Clustering each target monitoring point in the hedging combustion boiler by adopting a DBSCAN spatial clustering algorithm based on the temperature distribution data in the boiler, positioning the flame center of a flame area, and determining the behavior characteristics of the flame center movement track based on the attribute characteristics of the flame center in a continuous time frame; and the monitoring and early warning module is used for matching the behavior characteristics, the wall heating characteristics and the parameter characteristics of the left and right side flues of the opposite firing boiler with early warning conditions and carrying out slag formation, partial firing or over-temperature early warning according to the matching results.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method for pre-warning of the combustion state of a hedging boiler according to any one of claims 1 to 8 when executing the program.

Description

Method and device for early warning combustion state of opposed firing boiler and computer equipment Technical Field The application relates to the technical field of thermal power generation and monitoring, in particular to a method and a device for early warning of the combustion state of a opposed firing boiler and computer equipment. Background In the field of thermal power generation, the opposed firing boiler is widely used by virtue of the advantages of high combustion efficiency and low pollutant emission, the opposed firing boiler is divided into a plurality of layers of front walls and rear walls which are arranged on a water cooling wall of a hearth, and pulverized coal air flows are sprayed out from front wall burners and rear wall burners and form main upward air flows in opposed firing. The working condition characteristics of multi-flame opposite firing of the opposite firing boiler determine the quantity, the position and the temperature distribution of flame centers to directly influence the combustion efficiency, the load adjusting precision and the pollutant emission level of the unit. In the prior art, focusing on a single flame center device, the flame field of the single-core, high-temperature region concentrated features are presented. The multi-flame opposite firing of the opposite firing boiler can possibly form a plurality of independent flame cores, high-temperature flue gas interference exists among flames, the space competition and interference characteristics of the multi-flame cores are not considered in the existing clustering model, and when the clustering model is directly applied to the opposite firing boiler, misjudgment of multi-flame merging identification is easy to occur, and the working condition requirement of optimizing and adjusting a single burner of the opposite firing boiler cannot be adapted. Disclosure of Invention In view of the above, the application provides a method, a device and computer equipment for early warning the combustion state of a opposed firing boiler, which realize comprehensive improvement of flame center recognition precision, anti-interference capability and industrial suitability through scene adaptation, algorithm dimension separation and weighted model optimization. According to a first aspect of the present application, there is provided a method for early warning of a combustion state of a opposed firing boiler, the method comprising: Acquiring multi-source temperature data in a opposed firing boiler and parameter characteristics of left and right side flues of the opposed firing boiler, wherein the multi-source temperature data comprises temperature distribution data in the boiler and wall temperature distribution data; Determining a wall heating characteristic based on the wall temperature distribution data; Clustering each target monitoring point in the hedging combustion boiler by adopting a DBSCAN spatial clustering algorithm based on the temperature distribution data in the boiler, positioning the flame center of a flame area, and determining the behavior characteristics of the flame center movement track based on the attribute characteristics of the flame center in a continuous time frame; And matching the behavior characteristics, the wall heating characteristics, the parameter characteristics of the left and right side flues of the opposite firing boiler with early warning conditions, and carrying out slag formation, partial firing or over-temperature early warning according to the matching results. Optionally, clustering each target monitoring point in the opposed firing boiler by using a DBSCAN spatial clustering algorithm based on the temperature distribution data in the boiler, and positioning a flame center of a flame area, including: Acquiring three-dimensional time sequence data of a flame area in the opposed firing boiler, and screening out the target monitoring points with the temperature higher than a first temperature threshold value; Clustering the two-dimensional coordinates of the target monitoring points by adopting a DBSCAN spatial clustering algorithm, and taking an effective cluster obtained by clustering as the flame center; And calculating a weighted center coordinate by taking the temperature value of the target monitoring point in the effective cluster as a weight, and taking the weighted center coordinate and the highest temperature of the target monitoring point in the effective cluster as attribute characteristics of the flame center. Optionally, the determining the behavior feature of the flame center motion profile based on the attribute feature of the flame center in the continuous time frame includes: constructing a time sequence data set of attribute characteristics of the flame center under continuous time frames; Calculating the total similarity between flame centers in adjacent time frames through a space-temperature weighting algorithm based on the time sequence data set; A cost matrix is con