CN-122018337-A - Intelligent analysis method and system for coal quality detection data

CN122018337ACN 122018337 ACN122018337 ACN 122018337ACN-122018337-A

Abstract

The application relates to the technical field of coal-fired power generation, and discloses an intelligent analysis method and system for coal quality detection data, wherein the intelligent analysis method comprises the steps of performing cyclic sampling operation on a plurality of spatial positions of an outlet pipeline of a coal mill to obtain coal dust samples; the method comprises the steps of quantitatively intercepting a coal dust sample, applying preset pressure to form a to-be-detected entity, carrying out multi-physical-quantity signal acquisition on the to-be-detected entity to obtain a dielectric characteristic signal reflecting moisture content, a molecular vibration spectrum signal reflecting organic components and an atomic characteristic spectrum signal reflecting mineral elements, extracting characteristic values in the dielectric characteristic signal, the molecular vibration spectrum signal and the atomic characteristic spectrum signal, substituting the characteristic values into a preset quantitative association equation set, synchronously solving to obtain a moisture index, an ash index and a heat value index, comparing the moisture index, the ash index and the heat value index with corresponding preset reference values, generating and executing a coal mill regulating instruction, and providing an accurate data basis for real-time fine regulation of boiler combustion through fusion analysis of the multi-physical-quantity signals.

Inventors

LI ANMIN
LI YAJUN
ZHANG XIAOXIANG
YANG JIN
LI LE

Assignees

南京中宇自动化有限公司

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (10)

1. An intelligent analysis method for coal quality detection data is characterized by comprising the following steps: performing cyclic sampling operation on a plurality of spatial positions of an outlet pipeline of the coal mill to obtain coal dust samples; Quantitatively intercepting the pulverized coal sample, and applying preset pressure to form a to-be-detected entity; Performing multi-physical-quantity signal acquisition on the entity to be detected to obtain a dielectric characteristic signal reflecting the moisture content, a molecular vibration spectrum signal reflecting the organic component and an atomic characteristic spectrum signal reflecting the mineral elements; Extracting characteristic values in the dielectric characteristic signals, the molecular vibration spectrum signals and the atomic characteristic spectrum signals, substituting the characteristic values into a preset quantitative association equation set, and synchronously solving to obtain a moisture index, an ash index and a heat value index; And comparing the solved moisture index, ash index and heat value index with corresponding preset reference values to generate and execute a coal mill adjusting instruction.
2. The intelligent analysis method of coal quality detection data according to claim 1, wherein the step of performing a cyclic sampling operation at a plurality of spatial positions of the coal mill outlet to obtain the coal dust sample comprises: Sampling channels distributed at a plurality of space positions of an outlet pipeline of the coal mill are alternately opened according to preset time intervals; And establishing a negative pressure environment in the sampling channel so as to guide the pulverized coal airflow in the outlet pipeline of the coal mill to a separation path, and realizing physical separation of pulverized coal particles in the pulverized coal airflow and carrier gas by centrifugal force so as to obtain the pulverized coal sample.
3. The intelligent analysis method of coal quality detection data according to claim 2, wherein the step of establishing a negative pressure environment in the sampling channel to guide the pulverized coal airflow in the outlet pipe of the coal mill to a separation path and to realize physical separation of pulverized coal particles in the pulverized coal airflow from carrier gas by centrifugal force to obtain the pulverized coal sample comprises: the pulverized coal airflow is guided to a primary separation area, and a first centrifugal force field is utilized to trap a first part of pulverized coal particles and enable the first part of pulverized coal particles to fall into a collection area; Guiding the pulverized coal airflow passing through the primary separation area to a secondary separation area for secondary separation, and capturing fine dust by using a second centrifugal force field and enabling the fine dust to fall into the collection area; and combining the pulverized coal particles of the first part in the collecting area with the fine dust to obtain the pulverized coal sample.
4. The intelligent analysis method of coal quality detection data according to claim 1, wherein the step of quantitatively intercepting the pulverized coal sample and applying a preset pressure to form an entity to be detected comprises: intercepting a pulverized coal sample with preset weight, applying preset pressure and keeping for preset time to obtain the entity to be detected with preset saturation density.
5. The intelligent analysis method of coal quality detection data according to claim 4, wherein the step of applying a preset pressure and maintaining the preset time period comprises: monitoring a correlation curve of pressure intensity and displacement variation in the pressure applying process in real time; judging the elastic characteristics of the pulverized coal sample according to the correlation curve; And when the elastic characteristic meets a preset rebound condition, prolonging the preset time to eliminate the internal stress.
6. The intelligent analysis method of coal quality detection data according to claim 1, comprising, before the step of performing multi-physical-quantity signal acquisition on the entity to be detected: performing morphology scanning on the surface of the entity to be detected to obtain surface integrity characteristics; when the surface integrity characteristics are lower than a preset threshold value, generating a morphology correction coefficient according to the distribution density of the surface defects of the entity to be detected; The method comprises the following steps of acquiring dielectric characteristic signals reflecting the moisture content, molecular vibration spectrum signals reflecting the organic components and atomic characteristic spectrum signals reflecting the mineral elements: and carrying out numerical compensation on the collected molecular vibration spectrum signals by using the morphology correction coefficient so as to offset diffuse reflection loss.
7. The intelligent analysis method of coal quality detection data according to claim 1, wherein the step of performing multi-physical-quantity signal acquisition on the entity to be detected in the same processing cycle comprises: acquiring the initial surface temperature, the ambient temperature and the ambient humidity of the entity to be detected; calculating the water evaporation intensity according to the difference value between the initial surface temperature and the ambient humidity; and calling a preset attenuation compensation function based on the water evaporation intensity, and carrying out real-time adjustment and increase on the water response weight in the dielectric characteristic signal to obtain the compensated dielectric characteristic signal.
8. The intelligent analysis method of coal quality detection data according to claim 1, wherein the steps of extracting the characteristic values in the dielectric characteristic signal, the molecular vibration spectrum signal and the atomic characteristic spectrum signal, substituting the characteristic values into a preset quantitative association equation set, and synchronously solving to obtain a moisture index, an ash index and a heat value index comprise: Extracting a resonant frequency offset and a quality factor from the dielectric characteristic signal as dielectric characteristic values; Extracting absorbance or reflectivity at a preset wavelength position from the molecular vibration spectrum signal as a molecular vibration characteristic value; Extracting the characteristic peak intensity of a preset mineral element from the atomic characteristic spectrum signal as an atomic characteristic spectrum characteristic value; identifying the coal classification information of the current coal powder sample according to the characteristic peak intensity of the preset mineral elements; Retrieving a target quantitative association equation set matched with the coal classification information from a preset equation database; Substituting the extracted resonance frequency offset, the extracted quality factor, the extracted absorbance or reflectivity and the extracted characteristic peak intensity into the target quantitative association equation set for solving so as to obtain the moisture index, the extracted ash index and the extracted heat value index.
9. The intelligent analysis method of coal quality detection data according to claim 1, wherein the step of comparing the solved moisture index, ash index and heat value index with corresponding preset reference values to generate and execute a coal mill adjustment command comprises: When the water index is higher than a preset water reference value, executing an adjusting instruction for reducing the coal feeding amount of the coal mill and increasing the drying air quantity; when the ash index is higher than a preset ash reference value, executing a regulating instruction of soot blowing on a heating surface of the boiler; And when the heat value index is lower than a preset heat value reference value, executing a regulating command for increasing the output of the coal mill or increasing the primary air temperature.
10. A coal quality detection data intelligent analysis system for performing the coal quality detection data intelligent analysis method according to any one of claims 1 to 9, characterized in that the system comprises: The sampling module is used for performing cyclic sampling operation on a plurality of spatial positions of the outlet pipeline of the coal mill to obtain coal dust samples; The sample preparation module is used for quantitatively intercepting the pulverized coal sample and applying preset pressure to form a to-be-detected entity; The signal acquisition module is used for performing multi-physical-quantity signal acquisition on the entity to be detected to obtain a dielectric characteristic signal reflecting the moisture content, a molecular vibration spectrum signal reflecting the organic component and an atomic characteristic spectrum signal reflecting the mineral elements; The index solving module is used for extracting characteristic values in the dielectric characteristic signals, the molecular vibration spectrum signals and the atomic characteristic spectrum signals, substituting the characteristic values into a preset quantitative association equation set, and synchronously solving to obtain a moisture index, an ash index and a heat value index; And the control execution module is used for comparing the solved moisture index, ash index and heat value index with corresponding preset reference values to generate and execute a coal mill adjusting instruction.

Description

Intelligent analysis method and system for coal quality detection data Technical Field The application relates to the technical field of coal-fired power generation, in particular to an intelligent analysis method and system for coal quality detection data. Background Coal-fired power generation is taken as an important component of energy supply, and the economical and environmental protection performance of the operation of the coal-fired power generation is closely related to the stability of coal quality entering a furnace. With the increasing demand of the power grid for peak shaving capability, the generator set needs to frequently carry out deep peak shaving and low-load flexible operation. Under such conditions, small fluctuations in the quality of the coal may have a significant impact on the combustion stability and safety of the boiler. However, the coal quality detection means that are commonly relied upon in the current industry are still traditional off-line laboratory assays. This process typically involves manual sampling, sample feeding, sample preparation, and lengthy chemical analysis, with the entire cycle potentially being as long as several hours or even longer. This severe data hysteresis constitutes a critical technical problem in that when the test results are finally obtained, the batch of coal represented by it is already burnt out, the results being historical data and not reflecting the actual quality of the coal fines currently being fed into the coal mill and boiler. Therefore, the combustion control system of the boiler cannot perform feedforward adjustment according to real-time and accurate coal quality parameters, and only can perform feedback control with hysteresis, which results in low combustion efficiency and increased pollutant emission, and is extremely easy to cause risks of unstable combustion and even flameout when the load changes rapidly. In view of the above, there is a need in the art for improvements. Disclosure of Invention In order to solve the defects of the prior art, the application provides an intelligent analysis method and system for coal quality detection data, which can solve the technical problems that the traditional coal quality detection method has data lag and can not provide real-time effective data support for boiler combustion. In a first aspect, the application provides an intelligent analysis method for coal quality detection data, comprising the following steps: performing cyclic sampling operation on a plurality of spatial positions of an outlet pipeline of the coal mill to obtain coal dust samples; Quantitatively intercepting a coal dust sample, and applying preset pressure to form a to-be-detected entity; Performing multi-physical-quantity signal acquisition on an entity to be detected to obtain a dielectric characteristic signal reflecting the moisture content, a molecular vibration spectrum signal reflecting the organic component and an atomic characteristic spectrum signal reflecting the mineral element; extracting characteristic values in dielectric characteristic signals, molecular vibration spectrum signals and atomic characteristic spectrum signals, substituting the characteristic values into a preset quantitative association equation set, and synchronously solving to obtain a moisture index, an ash index and a heat value index; and comparing the solved moisture index, ash index and heat value index with corresponding preset reference values to generate and execute a coal mill adjusting instruction. A closed loop online analysis flow from sampling, sample preparation, detection solving and control is constructed, and a plurality of key coal quality indexes such as moisture, ash content, heat value and the like can be rapidly and synchronously acquired through fusion analysis of multiple physical quantity signals, so that the defects of long time consumption and data lag of a traditional test method are overcome, and an accurate data basis is provided for real-time fine adjustment of boiler combustion. Further, the step of performing a cyclic sampling operation at a plurality of spatial locations of the coal mill outlet to obtain a pulverized coal sample includes: Sampling channels distributed at a plurality of space positions of an outlet pipeline of the coal mill are alternately opened according to preset time intervals; And establishing a negative pressure environment in the sampling channel so as to guide the pulverized coal airflow in the outlet pipeline of the coal mill to a separation path, and realizing physical separation of pulverized coal particles in the pulverized coal airflow and carrier gas by centrifugal force so as to obtain a pulverized coal sample. By adopting a multi-point circulating sampling mode, the method effectively solves the problem of the representative deficiency of single-point sampling possibly caused by uneven distribution of pulverized coal in a pipeline, and simultaneously realizes full-automat