Boiler combustion intelligent early warning method and system based on multi-source information fusion

Abstract

The invention discloses a boiler combustion intelligent early warning method and system based on multi-source information fusion, and belongs to the technical field of boiler combustion monitoring. Layering the hearth along the height direction, further dividing the hearth into a plurality of combustion sub-layers, and arranging temperature, flue gas oxygen content and hearth pressure acquisition points in each combustion sub-layer. And multi-source operation parameters of all the combustion sublayers are synchronously collected in the data sampling time period, and the instantaneous combustion deviation amount of all the combustion sublayers is calculated. And fusing the instantaneous combustion deviations of the combustion sub-layers in the lower combustion layer and the upper combustion layer, constructing corresponding combustion layer state characteristic quantities, and calculating a comprehensive combustion state evaluation value of the whole boiler according to the corresponding combustion layer state characteristic quantities. According to the method, the threshold value interval is set, grading judgment and dynamic early warning of the boiler combustion state are achieved, combustion characteristic changes of different height areas of the boiler can be reflected more accurately, the sensitivity of combustion abnormity recognition and the perspectiveness of early warning are improved, and the method has high engineering application value and operation safety guarantee effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of boiler combustion monitoring, in particular to a boiler combustion intelligent early warning method and system based on multi-source information fusion. BACKGROUND

[0002] As the core heat energy conversion equipment in the fields of power generation, chemical industry, metallurgy and central heating, the stability, safety and economy of the combustion process of the boiler are directly related to the energy utilization efficiency and operation safety. With the wide application of large-scale and high-parameter boilers, the spatial scale of the furnace is continuously increasing, and the combustion process presents significant spatial non-uniformity and dynamic coupling characteristics. In recent years, with the development of industrial automation and information technology, the monitoring means of the boiler has gradually evolved from traditional manual experience judgment to online monitoring and intelligent analysis. In the prior art, the furnace temperature, flue gas oxygen content, furnace pressure and other operating parameters are generally collected through the boiler information management system, and the combustion regulation and safety interlocking are realized in combination with the DCS system. At the same time, multi-sensor information fusion, state evaluation and early warning analysis have gradually become a research hotspot, and some methods try to introduce multivariate analysis, statistical feature extraction or model-based diagnosis ideas to identify and warn the abnormal combustion of the boiler. However, in actual engineering applications, due to the influence of fuel quality, air-coal ratio, load fluctuation and other factors on the combustion process inside the boiler furnace, the combustion state has significant differences in space and time dimensions, and a single monitoring parameter or overall average index cannot accurately describe the complex combustion state, and the sensitivity and discrimination accuracy of early abnormalities are still limited.

[0003] From the deficiencies of the prior art, most of the current boiler combustion monitoring and early warning schemes still take the overall furnace or a small number of key measuring points as the analysis object, lack of fine description of the combustion distribution characteristics along the height direction of the furnace, and are difficult to reflect the evolution trend of local combustion deviation in time. On the one hand, the existing methods often simply equivalent the combustion behaviors of different height regions, ignoring the essential differences in fuel ignition, volatile analysis, burnout and flue gas remixing between the lower combustion zone and the upper combustion zone, resulting in that the evaluation results are not sensitive enough to local abnormalities; on the other hand, although multi-source operating parameters are collected at the same time, isolated thresholds or empirical rules are often used for judgment in the analysis process, lacking a unified deviation quantification mechanism, and it is difficult to comprehensively reflect the coupling relationship between temperature, oxygen content and pressure and other parameters. In addition, some existing early warning methods focus on post-accident identification or obvious abnormality determination, and lack effective identification means for slight deviation of the combustion state from stable to abnormal transition stage, the early warning level is rough, and it is difficult to support fine operation regulation and control and forward-looking safety prevention and control. SUMMARY

[0004] The present application aims to provide a boiler combustion intelligent early warning method and system based on multi-source information fusion to solve the problems in the background art.

[0005] To solve the above technical problems, the present application provides the following technical solutions:

[0006] A boiler combustion intelligent early warning method based on multi-source information fusion, the method comprising the following steps: step S1: obtaining basic parameters of a boiler from a boiler information management system background, layering a boiler hearth along a height direction, and arranging temperature collection points, flue gas collection points, and pressure collection points; step S2: constructing a data sampling time period, collecting hearth temperature data, flue gas oxygen content data, and hearth pressure data of lower and upper combustion sublayers at a data sampling time point; step S3: calculating the instantaneous combustion deviation of the lower and upper combustion sublayers at the data sampling time point; step S4: obtaining the instantaneous combustion deviation of all combustion sublayers in the lower and upper combustion layers at the data sampling time point, and constructing lower and upper combustion layer state characteristic quantities; calculating a comprehensive combustion state evaluation value of the boiler at the data sampling time point; step S5: presetting a comprehensive combustion state evaluation value threshold interval, and performing analysis and hierarchical early warning.

[0007] As a preferred scheme of the boiler combustion intelligent early warning method based on multi-source information fusion, the basic parameters of the boiler are obtained from the boiler information management system background, and the basic parameters include the geometric height parameter and the primary air nozzle height parameter of the boiler; the boiler hearth is layered along the height direction based on the geometric height parameter and the primary air nozzle height parameter of the boiler, and the specific steps are as follows:

[0008] The region from the bottom of the boiler hearth to the primary air nozzle height is defined as the lower combustion layer, and is denoted as ;

[0009] The region from the primary air nozzle height to the geometric height of the boiler hearth is defined as the upper combustion layer, and is denoted as ;

[0010] Based on the height ranges of the lower combustion layer and the upper combustion layer , the lower combustion layer and the upper combustion layer are divided into a plurality of sublayers along the hearth height direction, and a lower combustion sublayer set and an upper combustion sublayer set are constructed.

[0011] As a preferred scheme of the boiler combustion intelligent early warning method based on multi-source information fusion, the lower combustion sublayer set is denoted as , wherein Indicates the lower combustion layer The a-th lower combustion sublayer in the diagram, where A represents the lower combustion layer. The total number of lower combustion sublayers in the middle;

[0012] Let the set of upper combustion sublayers be denoted as ,in, Indicates the upper combustion layer The b-th upper combustion sublayer in the structure, where B represents the upper combustion layer. The total number of upper combustion sublayers in the middle;

[0013] Temperature acquisition points, flue gas acquisition points, and pressure acquisition points are set up in each sub-layer. The temperature acquisition points, flue gas acquisition points, and pressure acquisition points are used to collect furnace temperature data, flue gas oxygen content data, and furnace pressure data in the sub-layer, respectively.

[0014] As a preferred embodiment of the intelligent early warning method for boiler combustion based on multi-source information fusion described in this invention, a data sampling time period is constructed, denoted as... ,in, This represents the i-th data sampling time point, where I represents the total number of data sampling time points; at the data sampling time point The lower combustion sublayer was collected separately. and the upper combustion sublayer The furnace temperature data, flue gas oxygen content data, and furnace pressure data are recorded as follows: , , , , , .

[0015] As a preferred embodiment of the intelligent early warning method for boiler combustion based on multi-source information fusion described in this invention, based on data sampling time points... Lower combustion sublayer collected from the bottom Furnace temperature data Flue gas oxygen content data and furnace pressure data Calculate the data sampling time point Lower combustion sublayer The instantaneous combustion deviation is calculated using the following formula:

[0016] ;

[0017] in, Indicates the data sampling time point Lower combustion sublayer Instantaneous combustion deviation, This indicates the preset furnace temperature data. Influence factors This indicates the lower combustion sublayer of the preset stable phase. The average temperature reference value. This indicates the preset oxygen content data in the flue gas. Influence factors This indicates the lower combustion sublayer of the preset stable phase. The reference average oxygen content in flue gas. This indicates the preset furnace pressure data. Influence factors This indicates the lower combustion sublayer of the preset stable phase. Reference average value for furnace pressure data;

[0018] Based on data sampling time points Upper combustion sublayer collected from the lower part Furnace temperature data Flue gas oxygen content data and furnace pressure data Calculate the data sampling time point Lower and upper combustion sublayer The instantaneous combustion deviation is calculated using the following formula:

[0019] ;

[0020] in, Indicates the data sampling time point Lower and upper combustion sublayer Instantaneous combustion deviation, This indicates the preset furnace temperature data. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The average temperature reference value. This indicates the preset oxygen content data in the flue gas. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The reference average oxygen content in flue gas. This indicates the preset furnace pressure data. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The reference average value for furnace pressure data.

[0021] As a preferred embodiment of the intelligent early warning method for boiler combustion based on multi-source information fusion described in this invention, the data sampling time point is obtained. The lower combustion layer and upper combustion layer The instantaneous combustion deviation of all combustion sublayers is calculated, and the state characteristics of the lower combustion layer and the upper combustion layer are constructed, as follows:

[0022] ;

[0023] in, Indicates the data sampling time point Lower combustion layer The characteristic quantities of the lower combustion layer state, Indicates the data sampling time point Lower and upper combustion layers The state characteristics of the upper combustion layer, and These represent the preset lower combustion sublayers. and the upper combustion sublayer The height weighting coefficient;

[0024] Based on data sampling time points Lower combustion layer state characteristics and upper combustion layer state characteristics Calculate the data sampling time point The comprehensive combustion status assessment value of the boiler is calculated using the following formula:

[0025] ;

[0026] in, Indicates the data sampling time point The comprehensive combustion status assessment value of the boiler. and These represent the preset lower combustion layer state characteristic quantities. and upper combustion layer state characteristics Influencing factors.

[0027] As a preferred embodiment of the intelligent early warning method for boiler combustion based on multi-source information fusion described in this invention, a threshold range for the comprehensive combustion state assessment value is preset, and if the data sampling time point Comprehensive combustion status assessment value of the boiler If the value is less than the minimum value within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. The boiler combustion is normal.

[0028] If the data sampling time point Comprehensive combustion status assessment value of the boiler If the data sampling time point is within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. The boiler combustion status is slightly abnormal, triggering a Level 1 warning, reminding relevant personnel to pay attention to the combustion status;

[0029] If the data sampling time point Comprehensive combustion status assessment value of the boiler If the value is greater than the maximum value within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. If the combustion status of the boiler is abnormal, a level two early warning will be triggered, and an alarm signal will be output to remind relevant personnel to make combustion adjustments or check the relevant equipment.

[0030] The system acquires the comprehensive combustion status assessment value of the boiler at the data sampling time point in real time and provides dynamic intelligent early warning.

[0031] A boiler combustion intelligent early warning system based on multi-source information fusion. The system includes: a layered and data collection point layout module, a data acquisition module, a combustion deviation calculation module, a feature quantity construction and evaluation value calculation module, and an analysis and early warning module.

[0032] The layering and data collection point deployment module: obtains the basic parameters of the boiler from the boiler information management system backend, divides the boiler furnace into layers along the height direction, and deploys temperature data collection points, flue gas data collection points, and pressure data collection points.

[0033] The data acquisition module: constructs a data sampling time period, and collects furnace temperature data, flue gas oxygen content data, and furnace pressure data for the lower combustion sub-layer and the upper combustion sub-layer respectively at the data sampling time point;

[0034] The combustion deviation calculation module calculates the instantaneous combustion deviation of the lower and upper combustion sub-layers at the data sampling time points, respectively.

[0035] The feature quantity construction and evaluation value calculation module: obtains the instantaneous combustion deviation of all combustion sub-layers in the lower and upper combustion layers at the data sampling time point, and constructs the state feature quantity of the lower combustion layer and the state feature quantity of the upper combustion layer; calculates the comprehensive combustion state evaluation value of the boiler at the data sampling time point;

[0036] The analysis and early warning module: presets a threshold range for comprehensive combustion status assessment values, analyzes and provides graded early warnings.

[0037] Furthermore, the combustion deviation calculation module includes a combustion deviation calculation unit;

[0038] The combustion deviation calculation unit calculates the instantaneous combustion deviation of the lower and upper combustion sub-layers based on the furnace temperature data, flue gas oxygen content data, and furnace pressure data collected at the data sampling time point.

[0039] Furthermore, the feature quantity construction and evaluation value calculation module includes a feature quantity construction unit and an evaluation value calculation unit;

[0040] The feature quantity construction unit: obtains the instantaneous combustion deviation of all combustion sub-layers in the lower combustion layer and the upper combustion layer at the data sampling time point, and constructs the state feature quantity of the lower combustion layer and the state feature quantity of the upper combustion layer;

[0041] The evaluation value calculation unit calculates the comprehensive combustion state evaluation value of the boiler at the data sampling time point based on the state characteristic quantities of the lower combustion layer and the upper combustion layer at the data sampling time point.

[0042] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: This invention provides a boiler combustion intelligent early warning method and system based on multi-source information fusion. By structurally layering the boiler furnace along the height direction and further refining it into multiple combustion sub-layers, a spatial combustion model matching the actual combustion mechanism of the boiler is established. On this basis, multi-source operating parameters such as furnace temperature, flue gas oxygen content, and furnace pressure of each combustion sub-layer are synchronously collected within a unified data sampling time period, making the data of different height regions and different physical quantities comparable in time and space dimensions. Subsequently, by normalizing and calculating the deviation between the multi-source parameters of each combustion sub-layer at the current sampling time point and the corresponding stable operating condition reference value, a combustion deviation quantity reflecting the instantaneous combustion state of a single sub-layer is constructed, thereby integrating complex and multi-dimensional combustion information. The information is transformed into quantifiable and fusionable evaluation indicators. Furthermore, the sub-layer height weight is introduced to fuse the combustion deviation of each sub-layer, forming state characteristic quantities of the lower combustion layer and the upper combustion layer respectively. The overall combustion state evaluation value of the boiler is obtained at the hierarchical scale, so that local combustion anomalies can be transmitted level by level and reflected in the overall operating state. Finally, by setting threshold ranges for the comprehensive combustion state evaluation value and performing graded judgment, dynamic intelligent early warning of boiler combustion state is realized. It can not only distinguish different operating states such as normal, mild abnormality and severe abnormality, but also issue early warning signals in time before the combustion state deteriorates significantly. This improves the precision of boiler combustion process monitoring and the foresight of early warning, effectively reduces combustion anomalies and operational risks, and enhances the safety and operational reliability of the boiler system. Attached Figure Description

[0043] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0044] Figure 1 This is a schematic diagram illustrating the steps of an intelligent early warning method for boiler combustion based on multi-source information fusion according to the present invention.

[0045] Figure 2 This is a schematic diagram of the structure of an intelligent early warning system for boiler combustion based on multi-source information fusion according to the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Please see Figure 1 In this first embodiment, a method for intelligent early warning of boiler combustion based on multi-source information fusion is provided. The method includes the following steps:

[0048] Step S1: Obtain the basic parameters of the boiler from the boiler information management system backend, divide the boiler furnace into layers along the height direction, and set up temperature acquisition points, flue gas acquisition points, and pressure acquisition points.

[0049] Specifically, the basic parameters of the boiler are obtained from the boiler information management system backend. These basic parameters include the boiler's geometric height parameters and the primary air nozzle height parameters. Based on these parameters, the boiler furnace is divided into layers along the height direction, as follows:

[0050] The area from the bottom of the boiler furnace to the height of the primary air nozzle is defined as the lower combustion layer, and denoted as... ;

[0051] The area from the height of the primary air nozzle in the boiler furnace to its geometric height is defined as the upper combustion layer, and denoted as... ;

[0052] Based on the lower combustion layer and upper combustion layer The height range, along the height direction of the furnace, is used to separate the lower combustion layers. and upper combustion layer It is divided into several sub-layers, and a lower combustion sub-layer set and an upper combustion sub-layer set are constructed respectively.

[0053] Furthermore, the lower combustion sublayer set is denoted as... ,in, Indicates the lower combustion layer The a-th lower combustion sublayer in the diagram, where A represents the lower combustion layer. The total number of lower combustion sublayers in the middle;

[0054] Let the set of upper combustion sublayers be denoted as ,in, Indicates the upper combustion layer The b-th upper combustion sublayer in the structure, where B represents the upper combustion layer. The total number of upper combustion sublayers in the middle;

[0055] Temperature acquisition points, flue gas acquisition points, and pressure acquisition points are set up in each sub-layer. The temperature acquisition points, flue gas acquisition points, and pressure acquisition points are used to collect furnace temperature data, flue gas oxygen content data, and furnace pressure data in the sub-layer, respectively.

[0056] In this invention, the furnace is divided into upper and lower combustion layers along the height direction based on the boiler's geometric height parameters and the primary air nozzle height parameters. These layers are further refined into several sub-layers, establishing a clear spatial combustion structure model. This achieves the following technical effects: boiler combustion exhibits significantly different physical and chemical characteristics in different height regions. The lower combustion layer is dominated by fuel ignition and initial combustion, while the upper combustion layer is dominated by burnout and heat release. This layering method uses the primary air nozzle as the key physical boundary point, ensuring that the division has clear engineering significance and avoiding modeling distortion caused by arbitrary partitioning.

[0057] Sub-level modeling gives the subsequently collected data a natural spatial label, laying the foundation for the calculation of instantaneous combustion deviation, weight allocation, and hierarchical fusion. This is a prerequisite for subsequent steps to achieve accurate early warning. Once an anomaly occurs, it can not only determine whether there is an anomaly, but also trace it back to a specific height range, providing directional basis for operation adjustment and fault diagnosis.

[0058] Step S2: Establish a data sampling time period, and collect furnace temperature data, flue gas oxygen content data, and furnace pressure data for the lower combustion sub-layer and the upper combustion sub-layer respectively at the data sampling time points.

[0059] Specifically, the data sampling time period is constructed, denoted as . ,in, This represents the i-th data sampling time point, where I represents the total number of data sampling time points; at the data sampling time point The lower combustion sublayer was collected separately. and the upper combustion sublayer The furnace temperature data, flue gas oxygen content data, and furnace pressure data are recorded as follows: , , , , , .

[0060] In this invention, a unified data sampling time period is constructed, and temperature, flue gas oxygen content and furnace pressure data of each combustion sublayer are collected synchronously at each data sampling time point.

[0061] Temperature, oxygen content, and pressure are strongly coupled. A unified sampling time point avoids judgment errors introduced by time misalignment of different parameters, giving subsequent deviation calculations a true physical meaning. Constructing a time series through continuous sampling time points allows combustion state assessment to move beyond a single static moment and acquire dynamic evolution analysis capabilities, facilitating the early detection of combustion degradation trends. Furthermore, acquiring multi-source data under the same time reference reduces uncertainty in information fusion, making the fusion results more reliable for engineering applications.

[0062] Step S3: Calculate the instantaneous combustion deviation of the lower and upper combustion sublayers at the data sampling time points respectively.

[0063] Specifically, based on the data sampling time point Lower combustion sublayer collected from the bottom Furnace temperature data Flue gas oxygen content data and furnace pressure data Calculate the data sampling time point Lower combustion sublayer The instantaneous combustion deviation is calculated using the following formula:

[0064] ;

[0065] in, Indicates the data sampling time point Lower combustion sublayer Instantaneous combustion deviation, This indicates the preset furnace temperature data. Influence factors This indicates the lower combustion sublayer of the preset stable phase. The average temperature reference value. This indicates the preset oxygen content data in the flue gas. Influence factors This indicates the lower combustion sublayer of the preset stable phase. The reference average oxygen content in flue gas. This indicates the preset furnace pressure data. Influence factors This indicates the lower combustion sublayer of the preset stable phase. Reference average value for furnace pressure data;

[0066] Based on data sampling time points Upper combustion sublayer collected from the lower part Furnace temperature data Flue gas oxygen content data and furnace pressure data Calculate the data sampling time point Lower and upper combustion sublayer The instantaneous combustion deviation is calculated using the following formula:

[0067] ;

[0068] in, Indicates the data sampling time point Lower and upper combustion sublayer Instantaneous combustion deviation, This indicates the preset furnace temperature data. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The average temperature reference value. This indicates the preset oxygen content data in the flue gas. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The reference average oxygen content in flue gas. This indicates the preset furnace pressure data. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The reference average value for furnace pressure data.

[0069] It should be noted that in the instantaneous combustion deviation formula, the numerator is the absolute value of the measured value minus the reference mean. This is used to calculate the absolute deviation between the current sampling point data and the stable operating condition benchmark. When the boiler is in stable combustion, the temperature, oxygen content, and pressure of each sub-layer will be within a relatively fixed range (as shown by the reference mean temperature of the lower combustion layer). Typically 800-1200℃, depending on the fuel type), a larger deviation indicates less stable combustion. (A certain lower combustion sublayer...) =1000℃, current measured temperature =1100℃, then the absolute temperature deviation is 100℃, reflecting that the combustion of this sublayer is too strong.

[0070] The magnitudes of different parameters vary greatly (e.g., temperature is measured in °C, oxygen content in %), making direct addition meaningless. By dividing by the steady-state reference mean, the deviation is converted into a relative deviation rate (e.g., temperature deviation rate of 10%, oxygen content deviation rate of 20%), thus achieving comparability among multiple parameters.

[0071] In this invention, the instantaneous combustion deviation of each combustion sublayer is constructed by normalizing the temperature, oxygen content and pressure data of each combustion sublayer at the current sampling time point with the reference mean of the corresponding stable stage, and by introducing an influencing factor for weighting.

[0072] The multi-parameter combustion state is transformed into a single comparable quantitative indicator: different physical quantities have different dimensions and different directions of change. They are processed through relative deviation, so that the three types of parameters are integrated under the same evaluation scale, avoiding the distortion problem caused by direct superposition; using stable operating conditions as a reference benchmark is more conducive to identifying early and gradual combustion anomalies, rather than waiting until the absolute value exceeds the limit to trigger an early warning; different influencing factors are set for the upper and lower combustion sub-layers, so that the model can reflect the sensitivity of different height regions to each parameter, avoiding the judgment bias caused by uniform parameter weights.

[0073] Step S4: Obtain the instantaneous combustion deviation of all combustion sub-layers in the lower and upper combustion layers at the data sampling time point, and construct the state characteristic quantities of the lower combustion layer and the upper combustion layer; calculate the comprehensive combustion state evaluation value of the boiler at the data sampling time point.

[0074] Specifically, obtain the data sampling time point. The lower combustion layer and upper combustion layer The instantaneous combustion deviation of all combustion sublayers is calculated, and the state characteristics of the lower combustion layer and the upper combustion layer are constructed, as follows:

[0075] ;

[0076] in, Indicates the data sampling time point Lower combustion layer The characteristic quantities of the lower combustion layer state, Indicates the data sampling time point Lower and upper combustion layers The state characteristics of the upper combustion layer, and These represent the preset lower combustion sublayers. and the upper combustion sublayer The height weighting coefficient;

[0077] It should be noted that in the lower combustion layer, the sublayer (fuel injection zone) closest to the primary air nozzle has the greatest impact on combustion and ignition. The weight is set at 0.7-0.8; the weight of the sublayer farther from the nozzle is lower (0.2-0.3). In the upper combustion layer, the middle and upper sublayers are the key areas for flue gas mixing and burnout. A weight of 0.6-0.7 is used, with a lower weight for the top sub-layer.

[0078] The lower combustion layer has three sub-layers, near the nozzle. Weight =0.6, adjacent sub-layers =0.2、 =0.2, if If the deviation D = 0.3 and the deviation D = 0.1 for other sub-layers, then the characteristic value of this layer is... =0.073;

[0079] Output (Characteristics of the lower combustion layer) and (Characteristics of the upper combustion layer) reflect the overall combustion state of the entire combustion layer. For example, if A significant increase indicates that the lower fire zone is generally unstable, possibly due to an imbalance in the coal-air mixture ratio; if An increase in temperature indicates an abnormality in the upper burnout zone (such as insufficient oxygen content leading to incomplete combustion).

[0080] Based on data sampling time points Lower combustion layer state characteristics and upper combustion layer state characteristics Calculate the data sampling time point The comprehensive combustion status assessment value of the boiler is calculated using the following formula:

[0081] ;

[0082] in, Indicates the data sampling time point The comprehensive combustion status assessment value of the boiler. and These represent the preset lower combustion layer state characteristic quantities. and upper combustion layer state characteristics Influencing factors.

[0083] It should be noted that the lower combustion layer is the foundation of combustion. If the lower ignition is unstable (such as flameout or coking), it will directly lead to boiler shutdown or explosion risk. Typically, a value of 0.6-0.7 is used; anomalies in the upper combustion layer (such as incomplete combustion) mainly affect efficiency and environmental emissions, but pose a relatively low risk. Take 0.3-0.4.

[0084] Transform the upper and lower layer state characteristics into a single overall evaluation value. To avoid misjudging the overall state due to anomalies in a single layer, such as a slight anomaly in the upper combustion layer ( (Rising) but the lower part is stable ( (Normal), then It will not increase significantly, only triggering a mild warning; if the lower combustion layer is severely abnormal ( (a significant increase), then Rapidly exceeding the threshold triggers an emergency warning.

[0085] In this invention, by introducing height weights to the instantaneous combustion deviation of sub-layers for fusion, state characteristic quantities of upper and lower combustion layers are constructed respectively, and further, an overall comprehensive combustion state evaluation value of the boiler is formed. This structure avoids the complexity of directly making global judgments on a large amount of sub-layer data, making the combustion state evaluation logic clear, hierarchical, and with good interpretability. The influence of sub-layers at different heights on overall combustion safety varies, and this difference is reflected by weights, making the evaluation results more consistent with actual operating characteristics. The complex spatial and multi-parameter combustion state is ultimately condensed into a single comprehensive evaluation value, which retains the precision of front-end modeling and reduces the complexity of back-end decision-making, making it suitable for real-time online applications.

[0086] Step S5: Preset the threshold range of the comprehensive combustion status assessment value, analyze and issue graded warnings.

[0087] Specifically, a preset threshold range for the comprehensive combustion status assessment value is defined, based on the data sampling time point. Comprehensive combustion status assessment value of the boiler If the value is less than the minimum value within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. The boiler combustion is normal.

[0088] If the data sampling time point Comprehensive combustion status assessment value of the boiler If the data sampling time point is within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. The boiler combustion status is slightly abnormal, triggering a Level 1 warning, reminding relevant personnel to pay attention to the combustion status;

[0089] If the data sampling time point Comprehensive combustion status assessment value of the boiler If the value is greater than the maximum value within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. If the combustion status of the boiler is abnormal, a level two early warning will be triggered, and an alarm signal will be output to remind relevant personnel to make combustion adjustments or check the relevant equipment.

[0090] The system acquires the comprehensive combustion status assessment value of the boiler at the data sampling time point in real time and provides dynamic intelligent early warning.

[0091] In this invention, by pre-setting a threshold range for the comprehensive combustion status assessment value, and by classifying and outputting alarms based on the range in which the assessment value falls, the distinction between normal and abnormal is no longer simply made, but rather between mild and severe abnormalities is introduced. This facilitates operators in taking tiered and differentiated control measures. Through the threshold range design, frequent alarms caused by short-term fluctuations are avoided, thereby improving the stability and acceptability of the system in actual operation. The comprehensive combustion status assessment value is updated in real time, so that the warning results are dynamically adjusted according to changes in the combustion status, meeting the monitoring needs of long-term continuous boiler operation.

[0092] Please see Figure 2 In this second embodiment: a boiler combustion intelligent early warning system based on multi-source information fusion is provided. The system includes: a layered and collection point layout module, a data acquisition module, a combustion deviation calculation module, a feature quantity construction and evaluation value calculation module, and an analysis and early warning module.

[0093] The layering and data collection point deployment module: obtains the basic parameters of the boiler from the boiler information management system backend, divides the boiler furnace into layers along the height direction, and deploys temperature data collection points, flue gas data collection points, and pressure data collection points.

[0094] The data acquisition module: constructs a data sampling time period, and collects furnace temperature data, flue gas oxygen content data, and furnace pressure data for the lower combustion sub-layer and the upper combustion sub-layer respectively at the data sampling time point;

[0095] The combustion deviation calculation module calculates the instantaneous combustion deviation of the lower and upper combustion sub-layers at the data sampling time points, respectively.

[0096] The feature quantity construction and evaluation value calculation module: obtains the instantaneous combustion deviation of all combustion sub-layers in the lower and upper combustion layers at the data sampling time point, and constructs the state feature quantity of the lower combustion layer and the state feature quantity of the upper combustion layer; calculates the comprehensive combustion state evaluation value of the boiler at the data sampling time point;

[0097] The analysis and early warning module: presets a threshold range for comprehensive combustion status assessment values, analyzes and provides graded early warnings.

[0098] Furthermore, the combustion deviation calculation module includes a combustion deviation calculation unit;

[0099] The combustion deviation calculation unit calculates the instantaneous combustion deviation of the lower and upper combustion sub-layers based on the furnace temperature data, flue gas oxygen content data, and furnace pressure data collected at the data sampling time point.

[0100] Furthermore, the feature quantity construction and evaluation value calculation module includes a feature quantity construction unit and an evaluation value calculation unit;

[0101] The feature quantity construction unit: obtains the instantaneous combustion deviation of all combustion sub-layers in the lower combustion layer and the upper combustion layer at the data sampling time point, and constructs the state feature quantity of the lower combustion layer and the state feature quantity of the upper combustion layer;

[0102] The evaluation value calculation unit calculates the comprehensive combustion state evaluation value of the boiler at the data sampling time point based on the state characteristic quantities of the lower combustion layer and the upper combustion layer at the data sampling time point.

[0103] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0104] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent early warning of boiler combustion based on multi-source information fusion, characterized in that, The method includes the following steps: Step S1: Obtain the basic parameters of the boiler from the boiler information management system backend, divide the boiler furnace into layers along the height direction, and set up temperature acquisition points, flue gas acquisition points and pressure acquisition points. Step S2: Construct a data sampling time period, and collect furnace temperature data, flue gas oxygen content data, and furnace pressure data for the lower combustion sub-layer and the upper combustion sub-layer respectively at the data sampling time points; Step S3: Calculate the instantaneous combustion deviation of the lower and upper combustion sublayers at the data sampling time points respectively; Step S4: Obtain the instantaneous combustion deviation of all combustion sub-layers in the lower and upper combustion layers at the data sampling time point, and construct the state characteristic quantities of the lower combustion layer and the upper combustion layer; calculate the comprehensive combustion state evaluation value of the boiler at the data sampling time point; Step S5: Preset the threshold range of the comprehensive combustion status assessment value, analyze and issue graded warnings.

2. The intelligent early warning method for boiler combustion based on multi-source information fusion according to claim 1, characterized in that, The specific implementation process of step S1 includes: The boiler's basic parameters are obtained from the boiler information management system backend. These basic parameters include the boiler's geometric height parameters and primary air nozzle height parameters. Based on these parameters, the boiler furnace is divided into layers along the height direction, as follows: The area from the bottom of the boiler furnace to the height of the primary air nozzle is defined as the lower combustion layer, and denoted as... ; The area from the height of the primary air nozzle in the boiler furnace to its geometric height is defined as the upper combustion layer, and denoted as... ; Based on the lower combustion layer and upper combustion layer The height range, along the height direction of the furnace, is used to separate the lower combustion layers. and upper combustion layer It is divided into several sub-layers, and a lower combustion sub-layer set and an upper combustion sub-layer set are constructed respectively.

3. The intelligent early warning method for boiler combustion based on multi-source information fusion according to claim 2, characterized in that, The specific implementation process of step S1 also includes: Let the set of lower combustion sublayers be denoted as ,in, Indicates the lower combustion layer The a-th lower combustion sublayer in the diagram, where A represents the lower combustion layer. The total number of lower combustion sublayers in the middle; Let the set of upper combustion sublayers be denoted as ,in, Indicates the upper combustion layer The b-th upper combustion sublayer in the structure, where B represents the upper combustion layer. The total number of upper combustion sublayers in the middle; Temperature acquisition points, flue gas acquisition points, and pressure acquisition points are set up in each sub-layer. The temperature acquisition points, flue gas acquisition points, and pressure acquisition points are used to collect furnace temperature data, flue gas oxygen content data, and furnace pressure data in the sub-layer, respectively.

4. The intelligent early warning method for boiler combustion based on multi-source information fusion according to claim 3, characterized in that, The specific implementation process of step S2 includes: The data sampling time period is defined as follows: ,in, This represents the i-th data sampling time point, where I represents the total number of data sampling time points; at the data sampling time point The lower combustion sublayer was collected separately. and the upper combustion sublayer The furnace temperature data, flue gas oxygen content data, and furnace pressure data are recorded as follows: , , , , , .

5. The intelligent early warning method for boiler combustion based on multi-source information fusion according to claim 4, characterized in that, The specific implementation process of step S3 includes: Based on data sampling time points Lower combustion sublayer collected from the bottom Furnace temperature data Flue gas oxygen content data and furnace pressure data Calculate the data sampling time point Lower combustion sublayer The instantaneous combustion deviation is calculated using the following formula: ； in, Indicates the data sampling time point Lower combustion sublayer Instantaneous combustion deviation, This indicates the preset furnace temperature data. Influence factors This indicates the lower combustion sublayer of the preset stable phase. The average temperature reference value. This indicates the preset oxygen content data in the flue gas. Influence factors This indicates the lower combustion sublayer of the preset stable phase. The reference average oxygen content in flue gas. This indicates the preset furnace pressure data. Influence factors This indicates the lower combustion sublayer of the preset stable phase. Reference average value for furnace pressure data; Based on data sampling time points Upper combustion sublayer collected from the lower part Furnace temperature data Flue gas oxygen content data and furnace pressure data Calculate the data sampling time point Lower and upper combustion sublayer The instantaneous combustion deviation is calculated using the following formula: ； in, Indicates the data sampling time point Lower and upper combustion sublayer Instantaneous combustion deviation, This indicates the preset furnace temperature data. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The average temperature reference value. This indicates the preset oxygen content data in the flue gas. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The reference average oxygen content in flue gas. This indicates the preset furnace pressure data. Influence factors This indicates the upper combustion sublayer of the preset stable phase. The reference average value for furnace pressure data.

6. The intelligent early warning method for boiler combustion based on multi-source information fusion according to claim 5, characterized in that, The specific implementation process of step S4 includes: Obtain data sampling time points The lower combustion layer and upper combustion layer The instantaneous combustion deviation of all combustion sublayers is calculated, and the state characteristics of the lower combustion layer and the upper combustion layer are constructed, as follows: ； in, Indicates the data sampling time point Lower combustion layer The characteristic quantities of the lower combustion layer state, Indicates the data sampling time point Lower and upper combustion layers The state characteristics of the upper combustion layer, and These represent the preset lower combustion sublayers. and the upper combustion sublayer The height weighting coefficient; Based on data sampling time points Lower combustion layer state characteristics and upper combustion layer state characteristics Calculate the data sampling time point The comprehensive combustion status assessment value of the boiler is calculated using the following formula: ； in, Indicates the data sampling time point The comprehensive combustion status assessment value of the boiler. and These represent the preset lower combustion layer state characteristic quantities. and upper combustion layer state characteristics Influencing factors.

7. The intelligent early warning method for boiler combustion based on multi-source information fusion according to claim 6, characterized in that, The specific implementation process of step S5 includes: The preset threshold range for comprehensive combustion status assessment values, if the data sampling time point Comprehensive combustion status assessment value of the boiler If the value is less than the minimum value within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. The boiler combustion is normal. If the data sampling time point Comprehensive combustion status assessment value of the boiler If the data sampling time point is within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. The boiler combustion status is slightly abnormal, triggering a Level 1 warning. If the data sampling time point Comprehensive combustion status assessment value of the boiler If the value is greater than the maximum value within the threshold range of the comprehensive combustion state assessment value, then the data sampling time point is determined. An abnormal combustion state in the lower boiler triggers a level-two early warning and outputs an alarm signal; The system acquires the comprehensive combustion status assessment value of the boiler at the data sampling time point in real time and provides dynamic intelligent early warning.

8. A boiler combustion intelligent early warning system based on multi-source information fusion, executing the boiler combustion intelligent early warning method based on multi-source information fusion as described in any one of claims 1-7, characterized in that, The system includes: a layering and collection point layout module, a data acquisition module, a combustion deviation calculation module, a characteristic quantity construction and evaluation value calculation module, and an analysis and early warning module; The layering and data collection point deployment module: obtains the basic parameters of the boiler from the boiler information management system backend, divides the boiler furnace into layers along the height direction, and deploys temperature data collection points, flue gas data collection points, and pressure data collection points. The data acquisition module: constructs a data sampling time period, and collects furnace temperature data, flue gas oxygen content data, and furnace pressure data for the lower combustion sub-layer and the upper combustion sub-layer respectively at the data sampling time point; The combustion deviation calculation module calculates the instantaneous combustion deviation of the lower and upper combustion sub-layers at the data sampling time points, respectively. The feature quantity construction and evaluation value calculation module: obtains the instantaneous combustion deviation of all combustion sub-layers in the lower and upper combustion layers at the data sampling time point, and constructs the state feature quantity of the lower combustion layer and the state feature quantity of the upper combustion layer; calculates the comprehensive combustion state evaluation value of the boiler at the data sampling time point; The analysis and early warning module: presets a threshold range for comprehensive combustion status assessment values, analyzes and provides graded early warnings.

9. A boiler combustion intelligent early warning system based on multi-source information fusion according to claim 8, characterized in that: The combustion deviation calculation module includes a combustion deviation calculation unit; The combustion deviation calculation unit calculates the instantaneous combustion deviation of the lower and upper combustion sub-layers based on the furnace temperature data, flue gas oxygen content data, and furnace pressure data collected at the data sampling time point.

10. A boiler combustion intelligent early warning system based on multi-source information fusion according to claim 9, characterized in that: The feature quantity construction and evaluation value calculation module includes a feature quantity construction unit and an evaluation value calculation unit; The feature quantity construction unit: obtains the instantaneous combustion deviation of all combustion sub-layers in the lower combustion layer and the upper combustion layer at the data sampling time point, and constructs the state feature quantity of the lower combustion layer and the state feature quantity of the upper combustion layer; The evaluation value calculation unit calculates the comprehensive combustion state evaluation value of the boiler at the data sampling time point based on the state characteristic quantities of the lower combustion layer and the upper combustion layer at the data sampling time point.

Images