Air pre-heater air leakage rate analysis method based on multi-working-condition simulation

Abstract

The invention relates to the technical field of boiler regulation, and discloses an air pre-heater air leakage rate analysis method based on multi-working-condition simulation, which comprises the following steps: dividing boiler operation working conditions based on historical boiler operation data to generate a plurality of boiler operation working conditions, and obtaining inlet and outlet data of an air pre-heater of a boiler and combustion state data of the boiler; classifying the inlet and outlet data of the air pre-heater and the combustion state data of the boiler based on various boiler operation conditions, and generating a thermal efficiency prediction model of the boiler by combining the inlet and outlet data of the air pre-heater of the boiler and the combustion state data of the boiler under all the working conditions; combining the inlet and outlet data of the air pre-heater at the current monitoring time node and the thermal efficiency prediction model to generate a thermal efficiency prediction value RA at the current monitoring time node; and generating a correction instruction of the boiler based on the thermal efficiency prediction value. According to the method, the inlet and outlet data of the air pre-heater and the combustion state data of the boiler are utilized, and the operation state of the boiler is reflected more comprehensively through comprehensive utilization of cross-type data.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler regulation, and in particular to an air preheater air leakage rate analysis method based on multi-operating condition simulation. Background Art

[0002] Air preheaters are key auxiliary equipment for large power plant boilers. By utilizing waste heat from flue gases to heat the air, they improve overall boiler efficiency. However, air leakage in air preheaters can cause changes in the unit's thermal operating conditions, lowering the hot air temperature and reducing the unit's thermal efficiency. This can also increase the power consumption of the forced draft and induced draft fans, leading to higher coal consumption for both power generation and electricity supply.

[0003] Therefore, studying the air preheater leakage performance and taking measures to reduce the leakage rate are crucial to improving boiler efficiency. In recent years, environmental protection requirements have been increasingly stringent. After power plants complete denitrification renovations, the flue gas contains ammonium bisulfate, which is prone to condensation and adhesion, and can easily cause air preheater blockage. Summary of the Invention

[0004] The purpose of the present invention is to comprehensively consider the inlet and outlet data of the air preheater and the combustion status data of the boiler, construct a corresponding thermal efficiency prediction sub-model for the current operating conditions, and then combine the sub-models under all operating conditions to generate a final thermal efficiency prediction model, and generate correction instructions based on the air preheater leakage rate simulated under multiple operating conditions to adjust the boiler operation.

[0005] In order to achieve the above object, the present invention provides an air preheater leakage rate analysis method based on multi-operating condition simulation, comprising: Based on historical boiler operation data, the boiler operating conditions are divided into multiple boiler operating conditions, and the inlet and outlet data of the boiler's air preheater and the boiler's combustion status data are obtained; Based on various boiler operating conditions, the air preheater inlet and outlet data and the boiler combustion status data are classified, and the boiler thermal efficiency prediction model is generated by combining the air preheater inlet and outlet data and the boiler combustion status data under all operating conditions; Combine the import and export data of the air preheater at the current monitoring time node and the thermal efficiency prediction model to generate the thermal efficiency prediction value RA at the current monitoring time node; A correction command for the boiler is generated based on the predicted thermal efficiency value.

[0006] In some embodiments of the present invention, the generating of multiple boiler operating conditions further includes: Obtain historical power generation P and rated power generation P0; Calculate a first reference value a of the boiler operating condition based on the historical power generation P and the rated power generation P0; a=P / P0; Generate a first reference value set A based on a first reference value a in historical boiler operation data, where A={a1, a2…ai…an}; Wherein, ai is the i-th first reference value in the historical boiler operation data, and n is the total number of first reference values ​​in the historical boiler operation data; Obtaining an average value E(A) and a variance D(A) of a first reference value set A; Based on the average value E(A) and the variance D(A) of the first reference value set A, first reference value intervals of the boiler operating conditions are generated, and each first reference value interval corresponds to a boiler operating condition.

[0007] In some embodiments of the present invention, the generating of the first reference value interval of the boiler operating condition further includes: The multiple boiler operating conditions include: a first-level operating condition, a second-level operating condition, and a third-level operating condition; If ai∈[E(A)+k1*D(A),1], the boiler operating condition is in the first-level condition; If ai∈[E(A)-k1*D(A),E(A)+k1*D(A)], the boiler operating condition is in the secondary condition; If ai∈[0,E(A)-k1*D(A)], the boiler operating condition is in the third level condition; Among them, k1 is a constant, and the value of k1 is determined by analyzing and statistics of historical data.

[0008] In some embodiments of the present invention, generating a boiler thermal efficiency prediction model further includes: Generate a corresponding boiler thermal efficiency prediction sub-model based on the inlet and outlet data of the air preheater of the corresponding boiler under the current operating conditions and the combustion state data of the boiler; Generate an air leakage rate prediction model based on the inlet and outlet data of the air preheater; Generate a thermal efficiency reference value R of the boiler based on the combustion state data of the boiler; Combine the inlet and outlet data of the air preheater and the thermal efficiency reference value R of the boiler to generate a thermal efficiency prediction sub-model of the boiler under the current operating conditions; The thermal efficiency prediction model of the boiler is generated by combining the thermal efficiency prediction sub-models under all operating conditions.

[0009] In some embodiments of the present invention, generating an air leakage rate prediction model includes: Classify the import and export data of the air preheater to generate first-category data and second-category data; The second type of data includes: air flow at inlets and outlets; Generate an air preheater state evaluation value Q based on the first type of data; Generate an air leakage rate reference value E based on the air flow rate at the inlet and outlet; Based on the historical import and export data of the air preheater under the current operating conditions, a leakage rate sample set D is constructed, D = {(q1, e1), (q2, e2)…(qi, ei)…(qm, em)}; Wherein, (qi, ei) represents the i-th air leakage rate sample, qi represents the air preheater status evaluation value in the i-th air leakage rate sample, and ei represents the air leakage rate reference value in the i-th air leakage rate sample; Calculate a first proportional coefficient f1 based on the air leakage rate sample set D;

[0010] Calculate a first correction parameter b1 by combining the first proportional coefficient f1 and the air leakage rate sample set D;

[0011] Generate an air leakage rate prediction model by combining the first proportional coefficient f1 and the first correction parameter b1; .

[0012] In some embodiments of the present invention, when generating the air preheater state evaluation value Q based on the first type of data, the method further includes: The first type of data includes: air duct temperature, air duct pressure, and seal aging degree; Get the mean and variance of each type of first-class data in turn; Get the historical value range [L1, L2] of the current first-category data based on the mean and variance of the current first-category data; Setting the reference value X of the current first category data based on the historical value range of the current first category data;

[0013] Acquire sediment data inside the air preheater to generate a sediment reference value Y inside the air preheater; The air preheater state evaluation value Q is generated by combining the reference value X of the first type of data and the reference value Y of the sediment inside the air preheater;

[0014] Among them, x i is the reference value of the first type of data of type i, is the coefficient of the first type of data of type i, x1 is the reference value of the duct temperature, x2 is the reference value of the duct pressure, x3 is the reference value of the seal aging degree, is the weight of the first category of data, is the weight of the sediment inside the air preheater, Reference value of sediment inside the air preheater.

[0015] In some embodiments of the present invention, the generating of the boiler thermal efficiency reference value R further includes: The combustion state feature set S is constructed based on the combustion state data of the boiler, S={s1, s2…s i …s v1}; Among them, s i is the i-th combustion state feature in the combustion state feature set S, v1 is the total number of combustion state features in the combustion state feature set S; Calculating the correlation between the combustion state characteristics to generate combustion characteristic correlation values, and setting combustion characteristic preset values; By comparing the combustion feature related values ​​with the combustion feature preset values, the combustion state features in the combustion state feature set S are screened to obtain the second combustion state feature set St, st={st1, st2…st i …st v2}; Among them, st i is the i-th second combustion state feature in the second combustion state feature set St, v2 is the total number of second combustion state features in the second combustion state feature set St, v2 <v1; Generate a thermal efficiency reference value R of the boiler based on the second combustion state feature set St;

[0016] in, The second combustion state characteristic st i The weight of The second combustion state characteristic st i The fixed coefficient of Second correction parameter; For the second combustion state characteristic st i Converted to characteristic reference values ​​with the same range.

[0017] In some embodiments of the present invention, generating a thermal efficiency prediction sub-model for the boiler under the current operating conditions further includes: Obtain the air leakage rate reference value E and the corresponding thermal efficiency reference value R to generate a thermal efficiency prediction sample; Generate a thermal efficiency prediction sub-model for the boiler under current operating conditions based on the thermal efficiency prediction sample;

[0018] in, is the coefficient of the air leakage rate reference value E, is the third correction parameter.

[0019] In some embodiments of the present invention, the step of generating the predicted thermal efficiency value RA at the current monitoring time node further includes: Get the boiler operating conditions at the current monitoring time node; Select the thermal efficiency prediction sub-model based on the boiler operating conditions at the current monitoring time node;

[0020] in, represents the predicted value of thermal efficiency of the i-th boiler operating condition, R1 is the predicted value of thermal efficiency of the first-level operating condition, R2 is the predicted value of thermal efficiency of the second-level operating condition, and R3 is the predicted value of thermal efficiency of the third-level operating condition; The thermal efficiency prediction value RA of the current monitoring time node is generated by combining the import and export data of the air preheater at the current monitoring time node and the selected thermal efficiency prediction sub-model.

[0021] In some embodiments of the present invention, generating a correction instruction for the boiler further includes: Setting a plurality of thermal efficiency preset values ​​based on historical thermal efficiency reference values; By comparing the thermal efficiency prediction value RA of the current monitoring time node with the thermal efficiency preset value, the corresponding alarm level is generated; Generate corresponding correction instructions to the air preheater based on the current alarm level; And by monitoring the air preheater leakage rate, the effect of the correction instruction is predicted and the correction instruction feedback value is generated.

[0022] Compared with the prior art, the method for analyzing the air preheater leakage rate based on multi-operating condition simulation provided by the embodiment of the present invention has the following advantages: By constructing a first reference value set based on the ratio of historical power generation to rated power generation and dividing different operating condition intervals, the boiler operating conditions can be classified more accurately, which can reflect the differences in the operating status of the boiler at different power generation levels, and help to carry out subsequent refined analysis and control of different operating conditions.

[0023] When generating the thermal efficiency prediction model, the inlet and outlet data of the air preheater and the combustion status data of the boiler are comprehensively considered. The hierarchical model construction method can make full use of the data characteristics under different working conditions and improve the accuracy of thermal efficiency prediction.

[0024] The combustion state feature set is screened and features with high correlation are removed, which reduces the interference of redundant information, thereby improving the accuracy of the thermal efficiency reference value calculation, and further helps to improve the accuracy of the thermal efficiency prediction model.

[0025] The thermal efficiency prediction value RA is generated based on the air preheater inlet and outlet data of the current monitoring time node and the selected thermal efficiency prediction sub-model. The thermal efficiency can be predicted according to the real-time air preheater operation data, making the prediction result dynamically adaptable.

[0026] By classifying the air preheater inlet and outlet data, taking into account various status factors of the air preheater and the characteristics of historical data samples, the air leakage rate can be evaluated more accurately.

[0027] The model parameters mined from historical data can better reflect the air leakage rate law of the air preheater under different conditions, thereby improving the accuracy of air leakage rate prediction.

[0028] By setting multiple thermal efficiency preset values ​​based on historical thermal efficiency reference values, abnormal thermal efficiency situations can be discovered in a timely manner and corresponding corrective measures can be taken to ensure the safety and efficiency of boiler operation.

[0029] The effect of the correction instruction is predicted by monitoring the air preheater leakage rate to generate a correction instruction feedback value. This feedback mechanism can evaluate the effectiveness of the correction instruction. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of an air preheater leakage rate analysis method based on multi-operating condition simulation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following embodiments are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0032] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0033] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0034] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0035] Example 1: The embodiment of the present invention provides an air preheater leakage rate analysis method based on multi-condition simulation, such as Figure 1 Shown, including: Based on historical boiler operation data, the boiler operating conditions are divided into multiple boiler operating conditions, and the inlet and outlet data of the boiler's air preheater and the boiler's combustion status data are obtained; Based on various boiler operating conditions, the air preheater inlet and outlet data and the boiler combustion status data are classified, and the boiler thermal efficiency prediction model is generated by combining the air preheater inlet and outlet data and the boiler combustion status data under all operating conditions; Combine the import and export data of the air preheater at the current monitoring time node and the thermal efficiency prediction model to generate the thermal efficiency prediction value RA at the current monitoring time node; A correction command for the boiler is generated based on the predicted thermal efficiency value.

[0036] Example 2: When generating multiple boiler operating conditions, the method further includes: Obtain historical power generation P and rated power generation P0; Calculate a first reference value a of the boiler operating condition based on the historical power generation P and the rated power generation P0; a=P / P0; Generate a first reference value set A based on a first reference value a in historical boiler operation data, where A={a1, a2…ai…an}; Wherein, ai is the i-th first reference value in the historical boiler operation data, and n is the total number of first reference values ​​in the historical boiler operation data; Obtaining an average value E(A) and a variance D(A) of a first reference value set A; Based on the average value E(A) and the variance D(A) of the first reference value set A, first reference value intervals of the boiler operating conditions are generated, and each first reference value interval corresponds to a boiler operating condition.

[0037] Embodiment 3: When generating the first reference value interval of the boiler operating condition, the method further includes: The multiple boiler operating conditions include: a first-level operating condition, a second-level operating condition, and a third-level operating condition; If ai∈[E(A)+k1*D(A),1], the boiler operating condition is in the first-level condition; If ai∈[E(A)-k1*D(A),E(A)+k1*D(A)], the boiler operating condition is in the secondary condition; If ai∈[0,E(A)-k1*D(A)], the boiler operating condition is in the third level condition; Among them, k1 is a constant, and the value of k1 is determined by analyzing and statistics of historical data.

[0038] In this embodiment, determining the k1 value is crucial for accurately classifying boiler operating conditions. Different k1 values ​​can lead to different operating condition intervals, which in turn affects the determination of the boiler's operating status. If the k1 value is too large, the operating condition intervals may be illogical. For example, if the first-level operating condition interval is too large, some conditions that do not originally belong to high-load operation may also be classified as first-level operating conditions. Conversely, if the k1 value is too small, the operating condition intervals may be overly detailed, adding unnecessary complexity.

[0039] First, set an initial k1 value based on experience. Then, by backtesting historical data, divide the operating conditions based on this k1 value and check whether the divided operating conditions are consistent with the actual operating conditions (such as high, medium, and low load conditions determined by other relevant parameters). If not, adjust the k1 value until a satisfactory operating condition division is achieved.

[0040] Example 4: When generating the boiler thermal efficiency prediction model, the method further includes: Generate a corresponding boiler thermal efficiency prediction sub-model based on the inlet and outlet data of the air preheater of the corresponding boiler under the current operating conditions and the combustion state data of the boiler; Generate an air leakage rate prediction model based on the inlet and outlet data of the air preheater; Generate a thermal efficiency reference value R of the boiler based on the combustion state data of the boiler; Combine the inlet and outlet data of the air preheater and the thermal efficiency reference value R of the boiler to generate a thermal efficiency prediction sub-model of the boiler under the current operating conditions; The thermal efficiency prediction model of the boiler is generated by combining the thermal efficiency prediction sub-models under all operating conditions.

[0041] Example 5: Generating the air leakage rate prediction model includes: Classify the import and export data of the air preheater to generate first-category data and second-category data; The second type of data includes: air flow at inlets and outlets; Generate an air preheater state evaluation value Q based on the first type of data; Generate an air leakage rate reference value E based on the air flow rate at the inlet and outlet; Based on the historical import and export data of the air preheater under the current operating conditions, a leakage rate sample set D is constructed, D = {(q1, e1), (q2, e2)…(qi, ei)…(qm, em)}; Wherein, (qi, ei) represents the i-th air leakage rate sample, qi represents the air preheater status evaluation value in the i-th air leakage rate sample, and ei represents the air leakage rate reference value in the i-th air leakage rate sample; Calculate a first proportional coefficient f1 based on the air leakage rate sample set D;

[0042] Calculate a first correction parameter b1 by combining the first proportional coefficient f1 and the air leakage rate sample set D;

[0043] Generate an air leakage rate prediction model by combining the first proportional coefficient f1 and the first correction parameter b1; .

[0044] Example 6: When the air preheater state evaluation value Q is generated based on the first type of data, the method further includes: The first type of data includes: air duct temperature, air duct pressure, and seal aging degree; Get the mean and variance of each type of first-class data in turn; Get the historical value range [L1, L2] of the current first-category data based on the mean and variance of the current first-category data; Setting the reference value X of the current first category data based on the historical value range of the current first category data;

[0045] Acquire sediment data inside the air preheater to generate a sediment reference value Y inside the air preheater; The air preheater state evaluation value Q is generated by combining the reference value X of the first type of data and the reference value Y of the sediment inside the air preheater;

[0046] Among them, x i is the reference value of the first type of data of type i, is the coefficient of the first type of data of type i, x1 is the reference value of the duct temperature, x2 is the reference value of the duct pressure, x3 is the reference value of the seal aging degree, is the weight of the first category of data, is the weight of the sediment inside the air preheater, Reference value of sediment inside the air preheater.

[0047] In this embodiment, for the three first-category data, namely, the air duct temperature, the air duct pressure, and the sealing component aging degree, their average values ​​and variances are obtained in sequence.

[0048] The mean reflects the central tendency of the data. For example, the mean of the duct temperature tells us the approximate level of the duct temperature over a period of time. The variance measures the dispersion of the data. A high variance means that the data fluctuates greatly, while a low variance indicates that the data is relatively stable.

[0049] Based on the mean and variance of the current first-category data, the historical value range [L1, L2] of the current first-category data is obtained. This value range is determined based on the statistical characteristics of the historical data. The reference value X of the current first-category data is set based on the historical value range of the current first-category data. The reference value X is a quantitative representation of the first-category data under normal operating conditions.

[0050] The sediment data inside the air preheater is obtained to generate a sediment reference value Y inside the air preheater. The presence of sediment will affect the performance of the air preheater, such as reducing heat transfer efficiency and increasing air flow resistance.

[0051] The generation of the sediment reference value Y can be based on factors such as sediment thickness, density, and composition. For example, if the thickness of the sediment inside the air preheater is found to be h mm through testing, according to the pre-set standard, when h = 1 mm, the corresponding sediment reference value Y = 0.5 (this is just an example; the actual conversion relationship needs to be determined according to the specific evaluation standard).

[0052] Embodiment 7: When generating the boiler thermal efficiency reference value R, the method further includes: The combustion state feature set S is constructed based on the combustion state data of the boiler, S={s1, s2…s i …s v1}; Among them, s i is the i-th combustion state feature in the combustion state feature set S, v1 is the total number of combustion state features in the combustion state feature set S; Calculating the correlation between the combustion state characteristics to generate combustion characteristic correlation values, and setting combustion characteristic preset values; By comparing the combustion feature related values ​​with the combustion feature preset values, the combustion state features in the combustion state feature set S are screened to obtain the second combustion state feature set St, st={st1, st2…st i …st v2}; Among them, st i is the i-th second combustion state feature in the second combustion state feature set St, v2 is the total number of second combustion state features in the second combustion state feature set St, v2 <v1; Generate a thermal efficiency reference value R of the boiler based on the second combustion state feature set St;

[0053] in, The second combustion state characteristic st i The weight of The second combustion state characteristic st i The fixed coefficient of Second correction parameter; For the second combustion state characteristic st i Converted to characteristic reference values ​​with the same range.

[0054] In this embodiment, the boiler's combustion state is a key factor affecting thermal efficiency. The combustion state feature set S includes various combustion-related features, such as fuel type (s1 may indicate coal, natural gas, or other fuels), fuel supply (s2), air-to-fuel ratio (s3), temperature distribution within the combustion chamber (s4), and combustion stability (s5).

[0055] These characteristics comprehensively describe various aspects of the boiler combustion process, and different characteristics have different degrees of impact on thermal efficiency.

[0056] Combustion state characteristics may be interrelated. For example, the fuel supply and the air-fuel ratio influence each other. When the fuel supply increases, if the air-fuel ratio is not adjusted accordingly, combustion performance will be affected. By calculating the correlation between combustion state characteristics and generating a combustion characteristic correlation value, we can understand the degree of interrelationship between these characteristics.

[0057] By setting combustion characteristic preset values, we can determine which correlations between characteristics require special attention and which can be ignored. This preset value can be set based on an understanding of boiler combustion principles, past experience data, or statistical analysis of a large amount of historical data.

[0058] Assume that we find through calculation that the correlation value between the fuel supply (s2) and the temperature distribution in the combustion chamber (s4) is r. If we set the combustion characteristic preset value to r0, when r>r0, it means that the correlation between the two features is strong and requires special attention in subsequent analysis.

[0059] Since the number of features v1 in the combustion state feature set S may be large, some of these features may be redundant or have little impact on thermal efficiency. By comparing the combustion feature correlation values ​​with the preset combustion feature values, the combustion state features in the combustion state feature set S are screened to obtain a more streamlined and targeted second combustion state feature set St.

[0060] For example, if a feature is found to be highly correlated with other features and its impact on thermal efficiency can be indirectly reflected through other features, then this feature can be screened out, thereby reducing the amount of calculation and improving the accuracy of subsequent calculations of thermal efficiency reference values.

[0061] The number of features v2 in the second combustion state feature set St is smaller than v1. These features are selected to have a more direct and significant impact on boiler thermal efficiency. For example, after screening, key features such as fuel type, optimal air-fuel ratio, and combustion stability may remain.

[0062] Based on the screened key combustion state feature set St, the boiler thermal efficiency reference value R can be generated more accurately. This value will play an important role in the subsequent thermal efficiency prediction model construction and other processes.

[0063] Example 8: When generating the thermal efficiency prediction sub-model of the boiler under the current operating conditions, the method further includes: Obtain the air leakage rate reference value E and the corresponding thermal efficiency reference value R to generate a thermal efficiency prediction sample; Generate a thermal efficiency prediction sub-model for the boiler under current operating conditions based on the thermal efficiency prediction sample;

[0064] in, is the coefficient of the air leakage rate reference value E, is the third correction parameter.

[0065] In this embodiment, if a nonlinear method such as a neural network is used, the thermal efficiency prediction sample will serve as the input and output data of the neural network. By training the weights and bias of the neural network, it can accurately predict the thermal efficiency reference value R based on the air leakage rate reference value E.

[0066] The impact of air leakage on thermal efficiency may vary under different operating conditions. For example, under high-load conditions, where the boiler's combustion intensity is high, air leakage may be more sensitive to the impact on thermal efficiency. Under low-load conditions, while air leakage also affects thermal efficiency, the degree and manner of the impact may differ from those under high-load conditions.

[0067] Therefore, the thermal efficiency prediction sub-model constructed for the current working condition can more accurately reflect the relationship between the air leakage rate and thermal efficiency under the working condition, thereby improving the accuracy of thermal efficiency prediction.

[0068] Embodiment 9: When generating the thermal efficiency prediction value RA at the current monitoring time node, the method further includes: Get the boiler operating conditions at the current monitoring time node; Select the thermal efficiency prediction sub-model based on the boiler operating conditions at the current monitoring time node;

[0069] in, represents the predicted value of thermal efficiency of the i-th boiler operating condition, R1 is the predicted value of thermal efficiency of the first-level operating condition, R2 is the predicted value of thermal efficiency of the second-level operating condition, and R3 is the predicted value of thermal efficiency of the third-level operating condition; The thermal efficiency prediction value RA of the current monitoring time node is generated by combining the import and export data of the air preheater at the current monitoring time node and the selected thermal efficiency prediction sub-model.

[0070] In this embodiment, once an appropriate thermal efficiency prediction sub-model is selected, the air preheater inlet and outlet data at the current monitoring time point is substituted into the sub-model for calculation. The air preheater inlet and outlet data includes information such as air flow, temperature, and pressure, which are closely related to thermal efficiency.

[0071] The thermal efficiency prediction sub-model may contain parameter relationships related to the inlet and outlet air flow of the air preheater. By substituting the actual monitored inlet and outlet air flow data into the model, the thermal efficiency prediction value can be calculated under the current operating conditions after considering the working status of the air preheater.

[0072] RA reflects the thermal efficiency of the boiler under current operating conditions. This value can be compared with a preset thermal efficiency value to determine whether the boiler is operating normally and whether appropriate adjustments (such as adjusting or modifying the air preheater) are needed to improve thermal efficiency or ensure safe and stable operation of the boiler.

[0073] Example 10: The generating of the boiler correction instruction further includes: Setting a plurality of thermal efficiency preset values ​​based on historical thermal efficiency reference values; By comparing the thermal efficiency prediction value RA of the current monitoring time node with the thermal efficiency preset value, the corresponding alarm level is generated; Generate corresponding correction instructions to the air preheater based on the current alarm level; And by monitoring the air preheater leakage rate, the effect of the correction instruction is predicted and the correction instruction feedback value is generated.

[0074] In this embodiment, the historical thermal efficiency reference value data covers the thermal efficiency of the boiler under various conditions, different operating times, etc. Through in-depth analysis of these historical data, the distribution range and common values ​​of the boiler thermal efficiency can be understood.

[0075] Analysis of historical data may reveal that when the boiler is operating normally and stably, the thermal efficiency is usually in a relatively stable range; however, during startup, shutdown or special operating conditions, the thermal efficiency will fluctuate greatly.

[0076] Based on the results of data analysis, multiple thermal efficiency preset values ​​are set. These preset values ​​can be divided into different levels or ranges. A normal operating thermal efficiency preset value range is set. This range is determined based on the common thermal efficiency values ​​of the boiler during stable operation in historical data.

[0077] A lower thermal efficiency warning value can also be set. When the thermal efficiency is lower than this value, it indicates that there may be some problems with the boiler, such as serious air leakage in the air preheater or incomplete combustion. In addition, a critical value may be set. When the thermal efficiency is lower than this critical value, the boiler may face serious safety risks or economic losses.

[0078] At the current monitoring time point, a predicted thermal efficiency value, RA, has been obtained. This predicted value is compared with multiple previously set thermal efficiency preset values. For example, if RA is within the normal operating thermal efficiency preset value range, the boiler's operating status can be considered normal.

[0079] If the RA is below the normal operating range but above the warning level, a low-level alarm may be generated, indicating that the boiler's operating efficiency has decreased but not yet reached a critical level. If the RA is below the warning level, a higher-level alarm will be generated, indicating that timely attention and action are required.

[0080] Alert levels can be divided into different levels, such as level 1 (minor anomaly), level 2 (moderate anomaly), level 3 (serious anomaly), etc. Different alert levels correspond to different handling measures and levels of attention.

[0081] A level one alarm may only require simple data recording and preliminary inspection; a level two alarm may require more detailed inspection and analysis of related equipment (such as air preheaters); a level three alarm may require immediate measures to stop partial operations or perform emergency repairs.

[0082] Once the alarm level is determined, corresponding correction instructions are generated for the air preheater based on the alarm level. For a Level 1 alarm, the correction instructions may involve fine-tuning some of the air preheater's operating parameters. For example, adjusting the air preheater's inlet and outlet damper openings to slightly change the air flow rate in an attempt to improve thermal efficiency.

[0083] A Level 2 alert may require a more in-depth inspection and adjustment of the air preheater. For example, the air preheater seals may be inspected. If minor seal damage is found, a command may be issued to replace some seals, while also adjusting parameters such as the air preheating temperature and pressure.

[0084] In the case of a level 3 alarm, the correction instruction may be to stop the operation of the air preheater and conduct a comprehensive inspection and maintenance, including checking whether the tube bundle of the air preheater is blocked, whether there are serious air leaks, etc., and thoroughly repairing any problems found.

[0085] In addition to the specific operations performed on the air preheater itself, the correction instructions also specify the order in which these operations should be performed. For example, when performing air preheater maintenance, the relevant auxiliary equipment may be stopped first, and then the air preheater components may be disassembled in a specific order for inspection and repair. After the repairs are completed, the components may be reassembled and commissioned in the reverse order.

[0086] After issuing the correction instruction, the air preheater leakage rate is continuously monitored. The leakage rate data can be obtained by using the previously established leakage rate prediction model or specialized leakage rate monitoring equipment.

[0087] Compare the leakage rate data before and after the correction instruction is executed to evaluate the impact of the correction instruction on the air preheater leakage rate. If the leakage rate decreases significantly after the correction instruction is executed, it means that the correction instruction is effective. If the leakage rate does not change significantly or even increases, it indicates that there may be a problem with the correction instruction.

[0088] Based on the evaluation of the effectiveness of the correction instructions, a correction instruction feedback value is generated. This feedback value can be a quantitative value. For example, if the air leakage rate decreases by a certain percentage, the feedback value can be expressed as this percentage; if the air leakage rate does not change, the feedback value can be 0; if the air leakage rate increases, the feedback value can be a negative number.

[0089] The purpose of the correction command feedback value is to provide a basis for subsequent decision-making. If the feedback value indicates that the correction command is valid, the current strategy can be continued. If the feedback value indicates that the correction command is invalid, it is necessary to re-evaluate the correction command and adjust the strategy, which may require re-checking the status of the air preheater or re-determining the content of the correction command.

[0090] Finally, it should be noted that it is apparent that those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, to the extent such modifications and variations fall within the scope of the present invention and its equivalents, the present invention is intended to include such modifications and variations.

[0091] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A method for analyzing air preheater leakage rate based on multi-operating condition simulation, characterized in that: include: Based on historical boiler operation data, the boiler operating conditions are divided into multiple boiler operating conditions, and the inlet and outlet data of the boiler's air preheater and the boiler's combustion status data are obtained; Based on various boiler operating conditions, the air preheater inlet and outlet data and the boiler's combustion status data are classified, and the boiler's thermal efficiency prediction model is generated by combining the air preheater inlet and outlet data and the boiler's combustion status data under all operating conditions; Combine the import and export data of the air preheater at the current monitoring time node and the thermal efficiency prediction model to generate the thermal efficiency prediction value RA at the current monitoring time node; A correction command for the boiler is generated based on the predicted thermal efficiency value.

2. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 1 is characterized in that: The generation of multiple boiler operating conditions also includes: Obtain historical power generation P and rated power generation P0; Calculate a first reference value a of the boiler operating condition based on the historical power generation P and the rated power generation P0; a=P / P0; Generate a first reference value set A based on a first reference value a in historical boiler operation data, where A={a1, a2…ai…an}; Wherein, ai is the i-th first reference value in the historical boiler operation data, and n is the total number of first reference values ​​in the historical boiler operation data; Obtaining the average value E(A) and variance D(A) of the first reference value set A; Based on the average value E(A) and the variance D(A) of the first reference value set A, first reference value intervals of the boiler operating conditions are generated, and each first reference value interval corresponds to a boiler operating condition.

3. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 2 is characterized in that: When generating the first reference value interval of the boiler operating condition, the method further includes: The multiple boiler operating conditions include: first-level operating condition, second-level operating condition and third-level operating condition; If ai∈[E(A)+k1*D(A),1], the boiler operating condition is in the first-level condition; If ai∈[E(A)-k1*D(A),E(A)+k1*D(A)], the boiler operating condition is in the secondary condition; If ai∈[0,E(A)-k1*D(A)], the boiler operating condition is in the third level condition; Among them, k1 is a constant, and the value of k1 is determined by analyzing and statistics of historical data.

4. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 3 is characterized in that: When generating the thermal efficiency prediction model of the boiler, the method further includes: Generate a corresponding boiler thermal efficiency prediction sub-model based on the inlet and outlet data of the air preheater of the corresponding boiler under the current operating conditions and the combustion state data of the boiler; Generate an air leakage rate prediction model based on the inlet and outlet data of the air preheater; Generate a thermal efficiency reference value R of the boiler based on the combustion state data of the boiler; Combine the inlet and outlet data of the air preheater and the thermal efficiency reference value R of the boiler to generate a thermal efficiency prediction sub-model of the boiler under the current operating conditions; The thermal efficiency prediction model of the boiler is generated by combining the thermal efficiency prediction sub-models under all operating conditions.

5. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 4 is characterized in that: The generation of the air leakage rate prediction model includes: Classify the import and export data of the air preheater to generate first-category data and second-category data; The second type of data includes: air flow at inlets and outlets; Generate an air preheater state evaluation value Q based on the first type of data; Generate an air leakage rate reference value E based on the air flow rate at the inlet and outlet; Based on the historical import and export data of the air preheater under the current operating conditions, a leakage rate sample set D is constructed, D = {(q1, e1), (q2, e2)…(qi, ei)…(qm, em)}; Wherein, (qi, ei) represents the i-th air leakage rate sample, qi represents the air preheater status evaluation value in the i-th air leakage rate sample, and ei represents the air leakage rate reference value in the i-th air leakage rate sample; Calculate a first proportional coefficient f1 based on the air leakage rate sample set D; Calculate a first correction parameter b1 by combining the first proportional coefficient f1 and the air leakage rate sample set D; Generate an air leakage rate prediction model by combining the first proportional coefficient f1 and the first correction parameter b1; 。 6. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 5 is characterized in that: When the air preheater state evaluation value Q is generated based on the first type of data, the method further includes: The first type of data includes: air duct temperature, air duct pressure, and seal aging degree; Get the mean and variance of each type of first-class data in turn; Get the historical value range [L1, L2] of the current first-category data based on the mean and variance of the current first-category data; Setting the reference value X of the current first category data based on the historical value range of the current first category data; Acquire sediment data inside the air preheater to generate a sediment reference value Y inside the air preheater; The air preheater state evaluation value Q is generated by combining the reference value X of the first type of data and the reference value Y of the sediment inside the air preheater; Among them, x i is the reference value of the first type of data of type i, is the coefficient of the first type of data of type i, x1 is the reference value of the duct temperature, x2 is the reference value of the duct pressure, x3 is the reference value of the aging degree of the seal, is the weight of the first category of data, is the weight of the sediment inside the air preheater, Reference value of sediment inside the air preheater.

7. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 6, characterized in that: When generating the thermal efficiency reference value R of the boiler, the method further includes: The combustion state feature set S is constructed based on the combustion state data of the boiler, S={s1, s2…s i …s v1 }; Among them, s i is the i-th combustion state feature in the combustion state feature set S, v1 is the total number of combustion state features in the combustion state feature set S; Calculating the correlation between the combustion state characteristics to generate combustion characteristic correlation values, and setting combustion characteristic preset values; By comparing the combustion feature related values ​​with the combustion feature preset values, the combustion state features in the combustion state feature set S are screened to obtain the second combustion state feature set St, st={st1, st2…st i …st v2 }; Among them, st i is the i-th second combustion state feature in the second combustion state feature set St, v2 is the total number of second combustion state features in the second combustion state feature set St, v2 <v1; Generate a boiler thermal efficiency reference value R based on the second combustion state feature set St; in, The second combustion state characteristic st i The weight of The second combustion state characteristic st i The fixed coefficient of Second correction parameter; For the second combustion state characteristic st i Converted to characteristic reference values ​​with the same range.

8. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 7 is characterized in that: When generating the thermal efficiency prediction sub-model of the boiler under the current operating conditions, the method further includes: Obtain the air leakage rate reference value E and the corresponding thermal efficiency reference value R to generate a thermal efficiency prediction sample; Generate a thermal efficiency prediction sub-model for the boiler under current operating conditions based on the thermal efficiency prediction sample; in, is the coefficient of the air leakage rate reference value E, is the third correction parameter.

9. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 8, characterized in that: When generating the thermal efficiency prediction value RA at the current monitoring time node, the method further includes: Get the boiler operating conditions at the current monitoring time node; Select the thermal efficiency prediction sub-model based on the boiler operating conditions at the current monitoring time node; in, represents the predicted value of thermal efficiency of the i-th boiler operating condition, R1 is the predicted value of thermal efficiency of the first-level operating condition, R2 is the predicted value of thermal efficiency of the second-level operating condition, and R3 is the predicted value of thermal efficiency of the third-level operating condition; The thermal efficiency prediction value RA of the current monitoring time node is generated by combining the import and export data of the air preheater at the current monitoring time node and the selected thermal efficiency prediction sub-model.

10. The air preheater leakage rate analysis method based on multi-operating condition simulation according to claim 9, characterized in that: The generating of the boiler correction instruction further includes: Setting a plurality of thermal efficiency preset values ​​based on historical thermal efficiency reference values; By comparing the thermal efficiency prediction value RA of the current monitoring time node with the thermal efficiency preset value, the corresponding alarm level is generated; Generate corresponding correction instructions to the air preheater based on the current alarm level; And by monitoring the air preheater leakage rate, the effect of the correction instruction is predicted and the correction instruction feedback value is generated.

Images