State monitoring method and system for boiler burner

Abstract

The present invention relates to the technical field of boiler burners. Specifically, disclosed are a state monitoring method and system for a boiler burner. The method comprises: acquiring boiler burner fault events regarding occurrences of historical boiler burner faults, and on the basis of the boiler burner fault events, establishing a burner fault feature set; on the basis of an association rule algorithm, performing association rule mining on the burner fault feature set, so as to obtain strong association rules of the burner fault feature set; and on the basis of the strong association rules of the burner fault feature set, determining a combustion state coefficient of a current burner, and on the basis of the combustion state coefficient of the current burner, performing burner state early warning. In the present invention, association rule mining is performed on multi-source data of a boiler burner, such that a fault state of the burner can be monitored in real time, and a fault type can be analyzed, thereby effectively improving the fault state monitoring level for the burner.

Description

A method and system for monitoring the condition of a boiler burner Technical Field

[0001] This application relates to the field of boiler burner technology, and more specifically, to a method and system for monitoring the condition of a boiler burner. Background Technology

[0002] The pulverized coal burner is a major component of the combustion equipment in a coal-fired boiler. Its functions are: (1) to deliver fuel and air into the furnace; (2) to organize the timely and sufficient mixing of fuel and air; and (3) to ensure that the fuel ignites quickly and stably after entering the furnace and burns completely. In the combustion of pulverized coal, in order to reduce the heat required for ignition, the pulverized coal is heated quickly so that it reaches the ignition temperature as soon as possible to achieve rapid ignition. Therefore, the amount of air required for pulverized coal combustion is divided into primary air and secondary air. The function of primary air is to send pulverized coal into the furnace and supply the amount of oxygen required for the combustion of volatiles in the initial ignition stage of pulverized coal. Secondary air is mixed in after the pulverized coal airflow is ignited and supplies the amount of oxygen required for the complete combustion of coke and residual volatiles in the coal to ensure the complete combustion of pulverized coal.

[0003] The combustion process of pulverized coal in the furnace is complex and influenced by numerous factors. The combined effect of multiple variables makes boiler combustion fault diagnosis very difficult. In the process of diagnosing boiler combustion, there is too much reliance on the on-site experience of professionals and limited testing methods. There is a lack of reliable detection and diagnostic devices or methods to help technicians perform quantitative analysis, which makes it impossible to monitor the condition of the burner in a timely manner, resulting in safety hazards and economic losses to boiler operation. Summary of the Invention

[0004] This invention provides a method and system for monitoring the condition of a boiler burner, to solve the problem of low fault condition monitoring level in existing boiler burners, including:

[0005] Obtain historical boiler burner failure events when they occur, and establish a burner failure feature set based on these events.

[0006] Based on the association rule algorithm, association rules are mined from the burner fault feature set to obtain strong association rules for the burner fault feature set;

[0007] The combustion state coefficient of the current burner is determined based on the strong correlation rules of the burner fault feature set, and the burner status is warned based on the current combustion state coefficient.

[0008] Furthermore, the step of establishing a burner fault feature set based on boiler burner fault events includes:

[0009] Acquire multi-source indicator data of the boiler burner for boiler burner failure events. The multi-source indicator data of the burner includes operating sound data, combustion image data and operating temperature data.

[0010] Obtain the preset standard indicator data corresponding to each multi-source indicator data, and calculate the distance between the multi-source indicator data and the corresponding preset standard indicator data.

[0011] Based on the distance value changes between multi-source index data and corresponding preset standard index data, a time series data of distance value changes is established, and a burner fault feature set is established based on the time series data of distance value changes corresponding to each multi-source index data.

[0012] Further, the calculation of the distance between the multi-source indicator data and the corresponding preset standard indicator data includes:

[0013] Calculate the difference between the multi-source indicator data and the corresponding preset standard indicator data, and determine the initial distance value based on the difference between the multi-source indicator data and the corresponding preset standard indicator data;

[0014] Obtain the changes in multi-source indicator data, and plot the multi-source indicator data change curves based on the changes in multi-source indicator data.

[0015] Calculate the absolute value of the slope between the current multi-source indicator data and the multi-source indicator data of the previous preset period based on the change curve of the multi-source indicator data. Normalize the absolute value of the slope between the current multi-source indicator data and the multi-source indicator data of the previous preset period to obtain the distance correction coefficient.

[0016] Multiply the distance correction factor by the initial distance value to obtain the distance between the current multi-source indicator data and the corresponding preset standard indicator data.

[0017] Furthermore, the association rule-based algorithm is used to mine association rules in the burner fault feature set to obtain strong association rules for the burner fault feature set, including:

[0018] The time-series data of distance value changes and corresponding fault types in the burner fault feature set are standardized to obtain the standardized burner fault feature set.

[0019] Obtain the preset minimum support threshold and preset minimum confidence threshold, and mine the burner fault feature set based on the preset minimum support threshold and preset minimum confidence threshold using the association rule algorithm to obtain multiple strong association rules.

[0020] Furthermore, the burner fault feature set is mined based on an association rule algorithm according to a preset minimum support threshold and a preset minimum confidence threshold to obtain multiple strong association rules, including:

[0021] Obtain the feature itemset for each fault type in the burner fault feature set, and divide the feature itemset into a first leading itemset and a first successor itemset;

[0022] Calculate the target support of the first leading itemset and the first successor itemset, and compare the target support with the preset minimum support threshold.

[0023] When the target support is greater than or equal to the preset minimum support threshold, the first leading itemset and the first successor itemset corresponding to the target support are taken as frequent itemsets.

[0024] The association rule set is determined based on the frequent itemsets, and the strong association rules for the burner fault feature set are determined based on the association rule set.

[0025] Furthermore, the strong association rules for determining the burner fault feature set based on the association rule set include:

[0026] The association rule set is determined based on the frequent itemsets, and the association rule set is divided into the second leading itemset and the second successor itemset.

[0027] Calculate the target confidence and target lift of the second leading itemset and the second successor itemset;

[0028] The target confidence level is compared with the preset minimum confidence threshold, and the target lift is compared with 1.

[0029] When the target confidence level is greater than or equal to the minimum confidence threshold and the target lift is greater than 1, the corresponding association rule is used as the strong association rule corresponding to the burner fault feature set.

[0030] Furthermore, determining the current combustion state coefficient of the burner based on the strong correlation rules of the burner fault feature set includes:

[0031] Obtain the time series data of distance value changes corresponding to the current multi-source index data of the burner, establish a set of association rules based on strong association rules, and calculate the correlation coefficient between the time series data of distance value changes corresponding to the current multi-source index data and the time series data of distance value changes in the set of association rules.

[0032] The state weights and corresponding fault probabilities are determined based on the correlation coefficient between the time series data of distance value changes corresponding to the current multi-source index data and the time series data of distance value changes with strong correlation rules. The burner state coefficients are then determined based on the state weights and corresponding fault probabilities.

[0033] Further, determining the burner state coefficient based on the state weights and corresponding failure probabilities includes:

[0034] The burner state coefficient is determined based on the state coefficient calculation formula, according to the state weights and corresponding failure probabilities. The specific formula for calculating the state coefficient is as follows:

[0035]

[0036] in, This is the burner condition factor. Let be the state weight corresponding to the time series data of the distance value change of the i-th strongly correlated rule. The number of time-series data points showing changes in distance values ​​where strong association rules exist. This represents the probability of failure. To preset the allowable failure probability, For the preset range coefficient, It is a natural exponential function.

[0037] Furthermore, the step of providing burner status early warning based on the current burner combustion state coefficient includes:

[0038] Obtain preset standard burner state parameters and calculate the difference between the current burner state parameters and the preset standard burner state parameters;

[0039] Determine whether the difference between the current burner status parameters and the preset standard burner status parameters is greater than a first preset threshold. If the difference between the current burner status parameters and the preset standard burner status parameters is greater than the first preset threshold, then set the first level as the burner's warning level.

[0040] If the difference between the current burner status parameter and the preset standard burner status parameter is less than or equal to the first preset threshold, then determine whether the difference between the current burner status parameter and the preset standard burner status parameter is greater than the second preset threshold.

[0041] If the difference between the current burner status parameters and the preset standard burner status parameters is greater than the second preset threshold, then the second level is set as the burner's warning level.

[0042] If the difference between the current burner status parameters and the preset standard burner status parameters is less than or equal to the second preset threshold, then the third level is set as the burner's warning level.

[0043] To achieve the above objectives, the present invention also provides a boiler burner condition monitoring system, comprising:

[0044] A module is established to acquire boiler burner failure events that occurred in history, and to establish a burner failure feature set based on the boiler burner failure events.

[0045] The association module is used to mine association rules for the burner fault feature set based on the association rule algorithm, and obtain strong association rules for the burner fault feature set.

[0046] The monitoring module is used to determine the current combustion state coefficient of the burner based on the strong correlation rules of the burner fault feature set, and to provide burner status warnings based on the current combustion state coefficient.

[0047] The beneficial effects of this invention are as follows:

[0048] By applying the above technical solutions, this invention performs association rule mining on multi-source data of boiler burners, enabling timely monitoring of changes in multi-source data before a fault occurs, real-time monitoring of burner fault status and analysis of fault types, and timely early warning based on fault status level, effectively improving the level of burner fault status monitoring. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 shows an overall flowchart of a boiler burner condition monitoring method proposed in an embodiment of the present invention;

[0051] Figure 2 shows a schematic diagram of the structure of a boiler burner condition monitoring system proposed in an embodiment of the present invention. Embodiments of the present invention

[0052] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0053] This application provides a method for monitoring the condition of a boiler burner, as shown in Figure 1, including:

[0054] S101, Obtain boiler burner fault events when historical boiler burner faults occur, and establish a burner fault feature set based on the boiler burner fault events;

[0055] In some embodiments of this application, the step of establishing a burner fault feature set based on boiler burner fault events includes: acquiring multi-source indicator data of the burner for the boiler burner fault events, wherein the multi-source indicator data includes operating sound data, combustion image data, and operating temperature data; acquiring preset standard indicator data corresponding to each multi-source indicator data, and calculating the distance value between the multi-source indicator data and the corresponding preset standard indicator data; establishing time-series data of distance value changes based on the changes in the distance values ​​between the multi-source indicator data and the corresponding preset standard indicator data; and establishing a burner fault feature set based on the time-series data of distance value changes corresponding to each multi-source indicator data.

[0056] In this embodiment, multi-source indicator data of the burner is collected by fault events when historical boiler burner failures occur. The multi-source indicator data is preprocessed to obtain operating sound data, combustion image data and operating temperature data. The distance value between each multi-source indicator data and the standard data is calculated by pre-setting standard indicator data for each multi-source indicator data. Thus, a burner fault feature set is established based on the time series data of distance value changes. By combining the multi-source indicator data of historical burner failures to establish a fault feature set, the real-time operating status of the burner can be analyzed more accurately.

[0057] In some embodiments of this application, calculating the distance between multi-source indicator data and corresponding preset standard indicator data includes: calculating the difference between the multi-source indicator data and the corresponding preset standard indicator data; determining an initial distance value based on the difference between the multi-source indicator data and the corresponding preset standard indicator data; acquiring the changes in the multi-source indicator data; plotting a multi-source indicator data change curve based on the changes in the multi-source indicator data; calculating the absolute value of the slope between the current multi-source indicator data and the multi-source indicator data of the previous preset time period based on the multi-source indicator data change curve; normalizing the absolute value of the slope between the current multi-source indicator data and the multi-source indicator data of the previous preset time period to obtain a distance correction coefficient; and multiplying the distance correction coefficient by the initial distance value to obtain the distance between the current multi-source indicator data and the corresponding preset standard indicator data.

[0058] In this embodiment, the absolute value of the slope between the current multi-source indicator data and the multi-source indicator data of the previous preset period is calculated by the change curve of multi-source indicator data, and the absolute value of the slope is normalized to obtain the distance correction coefficient, so as to prevent misjudgment of faults due to data fluctuations.

[0059] S102, Based on the association rule algorithm, association rule mining is performed on the burner fault feature set to obtain strong association rules for the burner fault feature set;

[0060] In some embodiments of this application, the step of mining association rules for the burner fault feature set based on the association rule algorithm to obtain strong association rules for the burner fault feature set includes: standardizing the time series data of distance value changes and the corresponding fault types in the burner fault feature set to obtain a standardized burner fault feature set; obtaining a preset minimum support threshold and a preset minimum confidence threshold; and mining the burner fault feature set based on the association rule algorithm according to the preset minimum support threshold and the preset minimum confidence threshold to obtain multiple strong association rules.

[0061] In some embodiments of this application, the step of mining the burner fault feature set based on an association rule algorithm according to a preset minimum support threshold and a preset minimum confidence threshold to obtain multiple strong association rules includes: obtaining feature itemsets for each fault type in the burner fault feature set, and dividing the feature itemsets into a first leading itemset and a first successor itemset; calculating the target support of the first leading itemset and the first successor itemset, and comparing the target support with a preset minimum support threshold; when the target support is greater than or equal to the preset minimum support threshold, taking the first leading itemset and the first successor itemset corresponding to the target support as frequent itemsets; determining an association rule set based on the frequent itemsets, and determining strong association rules for the burner fault feature set based on the association rule set.

[0062] In some embodiments of this application, determining the strong association rules for the burner fault feature set based on the association rule set includes: determining the association rule set based on frequent itemsets, dividing the association rule set into a second leading itemset and a second successor itemset; calculating the target confidence and target lift of the second leading itemset and the second successor itemset; comparing the target confidence with a preset minimum confidence threshold, and comparing the target lift with 1; when the target confidence is greater than or equal to the minimum confidence threshold and the target lift is greater than 1, the corresponding association rule is taken as the strong association rule corresponding to the burner fault feature set.

[0063] In this embodiment, the distance value change time series data and corresponding strong association rules of the fault type in the burner fault feature set are mined based on the Apriori association rule algorithm. By mining a large amount of data in the burner fault feature set, strong association rules are extracted to improve the fault monitoring efficiency of the burner.

[0064] S103, determine the current combustion state coefficient of the burner based on the strong correlation rules of the burner fault feature set, and perform burner state warning based on the current combustion state coefficient of the burner.

[0065] In some embodiments of this application, determining the combustion state coefficient of the current burner based on strong correlation rules of the burner fault feature set includes: acquiring time-series data of distance value changes corresponding to the current multi-source index data of the burner; establishing a set of association rules based on strong correlation rules; calculating the correlation coefficient between the time-series data of distance value changes corresponding to the current multi-source index data and the time-series data of distance value changes in the set of association rules; determining the state weight and the corresponding fault probability based on the correlation coefficient between the time-series data of distance value changes corresponding to the current multi-source index data and the time-series data of distance value changes with strong correlation rules; and determining the burner state coefficient based on the state weight and the corresponding fault probability.

[0066] In some embodiments of this application, determining the burner state coefficient based on state weights and corresponding failure probabilities includes: determining the burner state coefficient based on a state coefficient calculation formula using state weights and corresponding failure probabilities, wherein the state coefficient calculation formula is specifically as follows:

[0067]

[0068] in, This is the burner condition factor. Let be the state weight corresponding to the time series data of the distance value change of the i-th strongly correlated rule. The number of time-series data points showing changes in distance values ​​where strong association rules exist. This represents the probability of failure. To preset the allowable failure probability, For the preset range coefficient, It is a natural exponential function.

[0069] In this embodiment, the corresponding fault probability is obtained through the confidence of strong association rules, and the state weight is obtained through the correlation coefficient between the time series data of distance value change corresponding to the current multi-source index data and the time series data of distance value change with strong association rules. Then, the state coefficient of the burner is determined according to the state weight and the fault probability, so as to achieve accurate monitoring of the burner fault state.

[0070] In some embodiments of this application, the step of providing burner status early warning based on the current burner combustion state coefficient includes: obtaining preset standard burner status parameters; calculating the difference between the current burner status parameters and the preset standard burner status parameters; determining whether the difference between the current burner status parameters and the preset standard burner status parameters is greater than a first preset threshold; if the difference between the current burner status parameters and the preset standard burner status parameters is greater than the first preset threshold, then setting the first level as the burner's early warning level; if the difference between the current burner status parameters and the preset standard burner status parameters is less than or equal to the first preset threshold, then determining whether the difference between the current burner status parameters and the preset standard burner status parameters is greater than a second preset threshold; if the difference between the current burner status parameters and the preset standard burner status parameters is greater than the second preset threshold, then setting the second level as the burner's early warning level; if the difference between the current burner status parameters and the preset standard burner status parameters is less than or equal to the second preset threshold, then setting the third level as the burner's early warning level.

[0071] Based on the same technical concept, as shown in Figure 2, the present invention also provides a boiler burner condition monitoring system, comprising:

[0072] The system includes a feature set module for acquiring historical boiler burner failure events and establishing a burner failure feature set based on these events; an association module for mining association rules on the burner failure feature set using an association rule algorithm to obtain strong association rules; and a monitoring module for determining the current burner combustion state coefficient based on the strong association rules of the burner failure feature set and providing burner status warnings based on the current burner combustion state coefficient.

[0073] By applying the above technical solutions, this invention acquires historical boiler burner failure events at the time of failure and establishes a burner failure feature set based on these events. It then mines association rules within the feature set using an association rule algorithm to obtain strong association rules. Finally, it determines the current burner's combustion state coefficient based on these strong association rules and provides a burner status early warning based on this coefficient. This invention utilizes association rule mining on multi-source boiler burner data, enabling real-time monitoring of burner failure status and analysis of failure types, effectively improving the level of burner failure status monitoring.

[0074] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for monitoring the condition of a boiler burner, characterized in that, include: Obtain historical boiler burner failure events when they occur, and establish a burner failure feature set based on these events. Based on the association rule algorithm, association rules are mined from the burner fault feature set to obtain strong association rules for the burner fault feature set; The combustion state coefficient of the current burner is determined based on the strong correlation rules of the burner fault feature set, and the burner status is warned based on the current combustion state coefficient.

2. The boiler burner condition monitoring method according to claim 1, characterized in that, The establishment of a burner fault feature set based on boiler burner fault events includes: Acquire multi-source indicator data of the boiler burner for boiler burner failure events. The multi-source indicator data of the burner includes operating sound data, combustion image data and operating temperature data. Obtain the preset standard indicator data corresponding to each multi-source indicator data, and calculate the distance between the multi-source indicator data and the corresponding preset standard indicator data. Based on the distance value changes between multi-source index data and corresponding preset standard index data, a time series data of distance value changes is established, and a burner fault feature set is established based on the time series data of distance value changes corresponding to each multi-source index data.

3. The boiler burner condition monitoring method according to claim 2, characterized in that, The calculation of the distance between multi-source indicator data and corresponding preset standard indicator data includes: Calculate the difference between the multi-source indicator data and the corresponding preset standard indicator data, and determine the initial distance value based on the difference between the multi-source indicator data and the corresponding preset standard indicator data; Obtain the changes in multi-source indicator data, and plot the multi-source indicator data change curves based on the changes in multi-source indicator data; Calculate the absolute value of the slope between the current multi-source indicator data and the multi-source indicator data of the previous preset period based on the change curve of the multi-source indicator data. Normalize the absolute value of the slope between the current multi-source indicator data and the multi-source indicator data of the previous preset period to obtain the distance correction coefficient. Multiply the distance correction factor by the initial distance value to obtain the distance between the current multi-source indicator data and the corresponding preset standard indicator data.

4. The boiler burner condition monitoring method according to claim 1, characterized in that, The association rule-based algorithm is used to mine association rules in the burner fault feature set to obtain strong association rules for the burner fault feature set, including: The time-series data of distance value changes and corresponding fault types in the burner fault feature set are standardized to obtain the standardized burner fault feature set. Obtain the preset minimum support threshold and preset minimum confidence threshold, and mine the burner fault feature set based on the preset minimum support threshold and preset minimum confidence threshold using the association rule algorithm to obtain multiple strong association rules.

5. The boiler burner condition monitoring method according to claim 4, characterized in that, The burner fault feature set is mined based on a preset minimum support threshold and a preset minimum confidence threshold using an association rule algorithm to obtain multiple strong association rules, including: Obtain the feature itemset for each fault type in the burner fault feature set, and divide the feature itemset into a first leading itemset and a first successor itemset; Calculate the target support of the first leading itemset and the first successor itemset, and compare the target support with the preset minimum support threshold. When the target support is greater than or equal to the preset minimum support threshold, the first leading itemset and the first successor itemset corresponding to the target support are taken as frequent itemsets. The association rule set is determined based on the frequent itemsets, and the strong association rules for the burner fault feature set are determined based on the association rule set.

6. The boiler burner condition monitoring method according to claim 5, characterized in that, The strong association rules for determining the burner fault feature set based on the association rule set include: The association rule set is determined based on the frequent itemsets, and the association rule set is divided into the second leading itemset and the second successor itemset. Calculate the target confidence and target lift of the second leading itemset and the second successor itemset; The target confidence level is compared with the preset minimum confidence threshold, and the target lift is compared with 1. When the target confidence level is greater than or equal to the minimum confidence threshold and the target lift is greater than 1, the corresponding association rule is used as the strong association rule corresponding to the burner fault feature set.

7. The boiler burner condition monitoring method according to claim 1, characterized in that, The determination of the current burner's combustion state coefficient based on the strong correlation rules of the burner fault feature set includes: Obtain the time series data of distance value changes corresponding to the current multi-source index data of the burner, establish a set of association rules based on strong association rules, and calculate the correlation coefficient between the time series data of distance value changes corresponding to the current multi-source index data and the time series data of distance value changes in the set of association rules. The state weights and corresponding fault probabilities are determined based on the correlation coefficient between the time series data of distance value changes corresponding to the current multi-source index data and the time series data of distance value changes with strong correlation rules. The burner state coefficients are then determined based on the state weights and corresponding fault probabilities.

8. The boiler burner condition monitoring method according to claim 7, characterized in that, The step of determining the burner state coefficient based on state weights and corresponding failure probabilities includes: The burner state coefficient is determined based on the state coefficient calculation formula, according to the state weights and corresponding failure probabilities. The specific formula for calculating the state coefficient is as follows: in, This is the burner condition factor. Let be the state weight corresponding to the time series data of the distance value change of the i-th strongly correlated rule. The number of time-series data points showing changes in distance values ​​where strong association rules exist. This represents the probability of failure. To preset the allowable failure probability, For the preset range coefficient, It is a natural exponential function.

9. The boiler burner condition monitoring method according to claim 8, characterized in that, The method of providing burner status early warning based on the current burner combustion state coefficient includes: Obtain preset standard burner state parameters and calculate the difference between the current burner state parameters and the preset standard burner state parameters; Determine whether the difference between the current burner status parameters and the preset standard burner status parameters is greater than a first preset threshold. If the difference between the current burner status parameters and the preset standard burner status parameters is greater than the first preset threshold, then set the first level as the burner's warning level. If the difference between the current burner status parameter and the preset standard burner status parameter is less than or equal to the first preset threshold, then determine whether the difference between the current burner status parameter and the preset standard burner status parameter is greater than the second preset threshold. If the difference between the current burner status parameters and the preset standard burner status parameters is greater than the second preset threshold, then the second level is set as the burner's warning level. If the difference between the current burner status parameters and the preset standard burner status parameters is less than or equal to the second preset threshold, then the third level is set as the burner's warning level.

10. A condition monitoring system for a boiler burner, characterized in that, include: A module is established to acquire boiler burner failure events that occurred in history, and to establish a burner failure feature set based on the boiler burner failure events. The association module is used to mine association rules for the burner fault feature set based on the association rule algorithm, and obtain strong association rules for the burner fault feature set. The monitoring module is used to determine the current combustion state coefficient of the burner based on the strong correlation rules of the burner fault feature set, and to provide burner status warnings based on the current combustion state coefficient.

Images