A method for modeling variable working condition of waste heat boiler evaporator based on linear regression

Abstract

The application discloses a kind of linear regression-based waste heat boiler evaporator variable condition modeling method, comprising: step 1, using the boundary parameter of traditional waste heat boiler evaporator variable condition thermodynamic calculation model to comb the influence of waste heat boiler evaporator variable condition thermodynamic performance;Step 2, in the range of variation of the boundary parameter determined, waste heat boiler evaporator variable condition thermodynamic calculation is carried out, a series of calculation results are obtained;Step 3, the feature of multiple linear regression algorithm and the prediction result of multiple linear regression are obtained;Step 4, the feature and prediction result of multiple linear regression algorithm determined in step 3 are used as the data set of multiple linear regression algorithm, and the data set is divided into training set and test set, the training of multiple linear regression algorithm is carried out, and the simplified calculation model of waste heat boiler evaporator variable condition thermodynamic calculation is constructed based on the training result.The application is used for waste heat boiler variable condition thermodynamic performance analysis, online performance monitoring and the like.

Description

Technical Field

[0001] This invention belongs to the field of gas-steam combined cycle units, specifically involving a variable operating condition modeling method for waste heat boiler evaporators based on linear regression. Background Technology

[0002] Waste heat boilers are an important heat exchange device in gas-steam combined cycle units. They recover the exhaust waste heat of the gas turbine and use it to heat the feedwater of the steam turbine system, generating high-temperature and high-pressure steam, which is then sent to the steam turbine to do work and generate some mechanical work. This not only increases the power of the combined cycle unit, but also improves the conversion efficiency between the chemical energy and mechanical energy of the fuel.

[0003] Variations in load, environmental meteorological conditions, and fuel composition in gas-fired combined cycle units often cause gas turbines to operate outside their design conditions. Consequently, the temperature, flow rate, and composition of the flue gas entering the waste heat boiler will frequently change. Furthermore, changes in environmental conditions will affect the turbine exhaust back pressure, thus impacting the thermodynamic performance of both the turbine and the waste heat boiler. Additionally, for combined cycle units with cogeneration, variations in heat supply directly affect the thermodynamic performance of the waste heat boiler. All these factors contribute to the waste heat boiler frequently operating under variable conditions. Therefore, researching methods for calculating the thermodynamic performance of waste heat boilers under variable conditions is crucial for predicting and analyzing the thermodynamic performance of waste heat boilers and even combined cycle units under these conditions.

[0004] Waste heat boilers consist of an economizer, an evaporator, and a superheater. Therefore, the thermal calculation of a waste heat boiler under varying operating conditions is a combined process of calculating the thermal performance of the economizer, evaporator, and superheater under these conditions. In traditional thermal calculations of waste heat boiler evaporators under varying operating conditions, the heat transfer coefficients on the flue gas and water sides require complex and tedious iterative calculations. This makes traditional methods unsuitable for rapid thermal performance analysis scenarios, such as online performance monitoring of waste heat boilers. Therefore, there is an urgent need to propose a method for quickly, easily, and accurately establishing a thermal calculation model for waste heat boiler evaporators under varying operating conditions. Summary of the Invention

[0005] The technical problem to be solved by this invention is that, in the traditional thermal calculation model of waste heat boiler evaporator under variable operating conditions, the heat transfer coefficients on the flue gas side and the water side need to be calculated through complex iterative calculations, which makes the thermal calculation process of waste heat boiler evaporator under variable operating conditions quite cumbersome. This invention provides a modeling method for waste heat boiler evaporator under variable operating conditions based on linear regression.

[0006] The present invention is achieved using the following technical solution:

[0007] A method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression includes the following steps:

[0008] Step 1: Use the traditional waste heat boiler evaporator variable operating condition thermodynamic calculation model to identify the boundary parameters affecting the variable operating condition thermodynamic performance of the waste heat boiler evaporator.

[0009] Step 2: Using the traditional waste heat boiler evaporator variable operating condition thermodynamic calculation model, and within the range of boundary parameter variation determined in Step 1, perform variable operating condition thermodynamic calculations of the waste heat boiler evaporator to obtain a series of calculation results;

[0010] Step 3: Take the thermodynamic calculation results of the waste heat boiler evaporator under the design operating condition as the benchmark operating condition. Take the changes of the boundary parameters taken in the variable operating condition thermodynamic calculation of the waste heat boiler evaporator in Step 2 relative to the benchmark operating condition boundary parameters and perform logarithmic transformation as the feature of the multiple linear regression algorithm. Take the changes of the overall heat transfer coefficient in the variable operating condition thermodynamic calculation results of the waste heat boiler evaporator in Step 2 relative to the benchmark operating condition heat transfer coefficient and perform logarithmic transformation as the prediction result of the multiple linear regression.

[0011] Step 4: Use the features and prediction results of the multiple linear regression algorithm determined in Step 3 as the dataset of the multiple linear regression algorithm, divide the dataset into training set and test set, carry out training of the multiple linear regression algorithm, and build a simplified calculation model for the variable operating condition thermodynamic calculation of the waste heat boiler evaporator based on the training results.

[0012] A further improvement of this invention is that, in steps 1 and 2, the traditional variable-condition thermodynamic calculation model for waste heat boiler evaporators is as follows:

[0013] The water-side heat transfer coefficient of the evaporator is obtained by solving the water-side parameters of the waste heat boiler evaporator under varying operating conditions and the geometric parameters of the evaporator tube bundle diameter and number.

[0014]

[0015] Where: h w The water-side heat transfer coefficient in a variable-condition evaporator is given in W / m³. 2 .℃; The specific heat on the water side of the evaporator is expressed in J / kg·℃. The viscosity of the water side in the evaporator is kg / ms; The thermal conductivity of the evaporator water side is W / m·℃; d i The inner diameter of the evaporator tube bundle is in mm.

[0016] A further improvement of this invention is that, in steps 1 and 2, the formula for calculating the heat transfer coefficient of the flue gas side under varying operating conditions of the waste heat boiler evaporator is as follows:

[0017]

[0018] Where: hg The heat transfer coefficient on the flue gas side of the evaporator under variable operating conditions, in W / m³. 2 .℃; C1, C3, and C5 are coefficients; is the outer diameter of the evaporator tube bundle, mm; h is the fin height, mm; The average temperature of the inlet and outlet flue gas is expressed in °C. Temperature at the fin, in °C; The value is the flue gas flow rate, in kg / s; Specific heat of flue gas, J / kg·℃; Viscosity of the evaporator flue gas, kg / ms; is the thermal conductivity of the evaporator flue gas side, W / m·℃.

[0019] A further improvement of this invention is that, in steps 1 and 2, the formula for calculating the overall heat transfer coefficient of the waste heat boiler evaporator under varying operating conditions is as follows:

[0020]

[0021] In the formula: The overall heat transfer coefficient in a variable-condition evaporator is given in W / m³. 2 .℃; A t The outer surface area of ​​the fin per unit fin length, in meters. 2 / m;A i The inner surface area of ​​the fin per unit fin length, in meters. 2 / m;ff i The fouling factor inside the finned tube is m. 2 ℃ / W; A w The average inner and outer surface areas of the fin per unit fin length, in meters. 2 / m;K m The thermal conductivity of the fin wall is W / m. 2 .℃;ff o The fouling factor on the outside of the finned tube, m 2 ℃ / W; η is the efficiency of the fin.

[0022] A further improvement of this invention is that, in steps 1 and 2, the logarithmic mean temperature difference of the evaporator is calculated based on the overall heat transfer coefficient of the waste heat boiler evaporator under varying operating conditions:

[0023]

[0024] Where: LMTD is the logarithmic mean temperature difference of the evaporator; e w2 The enthalpy of the water at the evaporator inlet is given in kJ / kg; e w1 A is the enthalpy of water at the evaporator outlet, kJ / kg. evp The heat exchange area of ​​the evaporator is in meters (m). 2 .

[0025] A further improvement of this invention is that, in step 2, the logarithmic mean temperature difference of the evaporator can also be obtained through thermodynamic calculations using the inlet and outlet flue gas side and water side parameters of the waste heat boiler evaporator:

[0026]

[0027] In the formula: T g1 T represents the inlet flue gas temperature of the evaporator, in °C. g2 T represents the evaporator outlet flue gas temperature, in °C. w1 The temperature of the water at the evaporator inlet is ℃.

[0028] A further improvement of this invention is that, in step 3, the calculation formula is as follows:

[0029] The calculation results under the design operating condition of the waste heat boiler are used as the benchmark operating condition, that is:

[0030]

[0031] In the formula: The overall heat transfer coefficient of the evaporator under the reference operating conditions is expressed in W / m². 2 .℃;W g,b The flue gas flow rate is given under the baseline operating conditions, in kg / s. The specific heat of the flue gas under the reference operating conditions is J / kg·℃; The viscosity of the flue gas under the reference operating conditions is kg / ms; km. g,b The thermal conductivity of the flue gas side under the reference operating conditions is W / m. 2 .℃;W w,b The water flow rate under the baseline operating conditions is expressed in kg / s. The water-side specific heat is given under the baseline operating conditions, in J / kg·℃. Water-side viscosity under reference operating conditions, kg / ms; km w,b The thermal conductivity of the water side under the reference operating conditions is W / m. 2 .℃;

[0032] Thermodynamic calculations for the waste heat boiler evaporator under varying operating conditions are performed within typical ranges of flue gas temperature, flow rate, composition, and water-side temperature and pressure. The changes in each boundary parameter of the thermodynamic calculation model under varying operating conditions relative to the baseline operating condition boundary parameters, and the change in the overall heat transfer coefficient in the thermodynamic calculation results under varying operating conditions relative to the baseline operating condition heat transfer coefficient are also calculated.

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042] In the formula: W g,o The flue gas flow rate under varying operating conditions is expressed in kg / s. The specific heat of the flue gas under varying operating conditions is expressed in J / kg·℃. For flue gas viscosity under varying operating conditions, kg / ms; km g,o The thermal conductivity of the flue gas side under varying operating conditions, W / m 2 .℃;W w,o The water flow rate under varying operating conditions is expressed in kg / s. The specific heat of the water side under varying operating conditions is expressed in J / kg·℃. For water-side viscosity under varying operating conditions, kg / ms; km w,o Thermal conductivity of the water side under varying operating conditions, W / m 2 .℃;r wg r represents the change in flue gas flow rate under varying operating conditions relative to the baseline operating condition. cpg r represents the change in flue gas specific heat under varying operating conditions relative to the reference operating condition. cpg r represents the change in flue gas specific heat under varying operating conditions relative to the reference operating condition. vg r represents the change in flue gas viscosity under varying operating conditions relative to the reference flue gas viscosity; kmg r represents the change in the thermal conductivity of the flue gas side under varying operating conditions relative to the thermal conductivity of the flue gas side under the reference operating conditions. ww r represents the change in water-side flow rate under variable operating conditions relative to the water-side flow rate under reference operating conditions. cpw r represents the change in water-side specific heat under varying operating conditions relative to the reference operating condition. vw r represents the change in water-side viscosity under varying operating conditions relative to the water-side viscosity under reference operating conditions. kmw r represents the change in water-side thermal conductivity under varying operating conditions relative to the reference operating condition. u This represents the change in the overall heat transfer coefficient of the evaporator under varying operating conditions relative to the overall heat transfer coefficient of the evaporator under reference operating conditions.

[0043] Logarithmic transformations were performed on the changes in each boundary parameter of the variable operating condition thermodynamic calculation model of the waste heat boiler evaporator relative to the baseline operating condition boundary parameter, and on the change in the overall heat transfer coefficient in the variable operating condition thermodynamic calculation results of the waste heat boiler evaporator relative to the baseline operating condition heat transfer coefficient. These transformations were then used as the features and prediction results of the multiple linear regression algorithm, respectively.

[0044]

[0045]

[0046] In the formula: X represents the characteristics of the multiple linear regression algorithm; Y represents the prediction result of the multiple linear regression algorithm.

[0047] A further improvement of this invention is that, in step 4, the calculation formula is as follows:

[0048] Based on the calculation results of the variable operating condition thermodynamic calculation model of the waste heat boiler evaporator, a multiple linear regression algorithm is constructed using the features and prediction results.

[0049] + + + +

[0050] In the formula: a0~a7 are all training results of the multiple linear regression algorithm;

[0051] The multiple linear regression algorithm is trained based on the training set, and a simplified thermodynamic calculation model for the waste heat boiler evaporator under varying operating conditions is constructed based on the training results, namely:

[0052] .

[0053] The present invention has at least the following beneficial technical effects:

[0054] This invention provides a method for modeling variable operating conditions of waste heat boiler evaporators based on linear regression. It employs a traditional variable operating condition model for waste heat boiler evaporators to perform variable operating condition calculations, and trains a multiple linear regression algorithm in machine learning based on the calculation results. Finally, it establishes a simplified model for calculating variable operating conditions of waste heat boiler evaporators based on the training results of the multiple linear regression algorithm. This simplified model avoids the complex iterative calculations of traditional methods for calculating variable operating conditions of waste heat boiler evaporators while maintaining high accuracy. It can be used for variable operating condition thermodynamic performance analysis and online performance monitoring of waste heat boilers. Attached Figure Description

[0055] Figure 1 This is a flowchart of a variable operating condition modeling method for waste heat boiler evaporators based on multiple linear regression.

[0056] Figure 2 This is a schematic diagram of the coefficients a0~a7 obtained by training the multiple linear regression algorithm.

[0057] Figure 3 A comparison chart of the overall heat transfer coefficient calculated by the model and the actual value of the overall heat transfer coefficient, used to simplify the model calculation.

[0058] Figure 4 A diagram showing the deviation between the calculated overall heat transfer coefficient and the actual overall heat transfer coefficient, used to simplify the model calculation. Detailed Implementation

[0059] The present invention will now be described in detail with reference to the accompanying drawings:

[0060] According to the traditional variable-condition thermodynamic calculation model of waste heat boiler evaporators, the overall heat transfer coefficient of the evaporator is directly affected by the flue gas flow rate, specific heat, dynamic viscosity, and thermal conductivity, as well as the water flow rate, specific heat, dynamic viscosity, and thermal conductivity. Therefore, it is necessary to determine the boundary parameters (such as...) that affect the variable-condition thermodynamic performance of waste heat boiler evaporators. Figure 1 As shown), that is

[0061]

[0062] Where: U is the overall heat transfer coefficient of the evaporator, W / m 2 .℃;W g The flow rate is expressed as flue gas flow rate, kg / s. Specific heat of flue gas, J / kg·℃; Flue gas viscosity, kg / ms; km g The thermal conductivity on the flue gas side is W / m. 2 .℃;W w The flow rate is expressed as kg / s. Specific heat on the water side, J / kg·℃; Water-side viscosity, kg / ms; km w The thermal conductivity on the water side is W / m. 2 .℃.

[0063] The calculation results under the design operating condition of the waste heat boiler are used as the benchmark operating condition, that is:

[0064]

[0065] In the formula: U evp,b The overall heat transfer coefficient of the evaporator under the reference operating conditions is expressed in W / m². 2 .℃;W g,b The flue gas flow rate is given under the baseline operating conditions, in kg / s. The specific heat of the flue gas under the reference operating conditions is J / kg·℃; The viscosity of the flue gas under the reference operating conditions is kg / ms; km.g,b The thermal conductivity of the flue gas side under the reference operating conditions is W / m. 2 .℃;W w,b The water flow rate under the baseline operating conditions is expressed in kg / s. The water-side specific heat is given under the baseline operating conditions, in J / kg·℃. Water-side viscosity under reference operating conditions, kg / ms; km w,b The thermal conductivity of the water side under the reference operating conditions is W / m. 2 .℃.

[0066] like Figure 1 As shown, under the typical variation range of flue gas temperature, flow rate, composition, and water-side temperature and pressure under variable operating conditions, the thermodynamic calculation of the waste heat boiler evaporator under variable operating conditions is performed. The changes in each boundary parameter of the variable operating condition thermodynamic calculation model relative to the baseline operating condition boundary parameters, and the change in the overall heat transfer coefficient in the variable operating condition thermodynamic calculation results relative to the baseline operating condition heat transfer coefficient are calculated.

[0067]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] In the formula: The overall heat transfer coefficient of the evaporator under varying operating conditions is expressed in W / m². 2 .℃;W g,o The flue gas flow rate under varying operating conditions is expressed in kg / s. The specific heat of the flue gas under varying operating conditions is expressed in J / kg·℃. For flue gas viscosity under varying operating conditions, kg / ms; km g,o The thermal conductivity of the flue gas side under varying operating conditions, W / m 2 .℃;W w,o The water flow rate under varying operating conditions is expressed in kg / s. The specific heat of the water side under varying operating conditions is expressed in J / kg·℃. For water-side viscosity under varying operating conditions, kg / ms; km w,o Thermal conductivity of the water side under varying operating conditions, W / m 2 .℃;r wg r represents the change in flue gas flow rate under varying operating conditions relative to the baseline operating condition. cpg r represents the change in flue gas specific heat under varying operating conditions relative to the reference operating condition. cpg r represents the change in flue gas specific heat under varying operating conditions relative to the reference operating condition. vg r represents the change in flue gas viscosity under varying operating conditions relative to the reference flue gas viscosity; kmg r represents the change in the thermal conductivity of the flue gas side under varying operating conditions relative to the thermal conductivity of the flue gas side under the reference operating conditions. ww r represents the change in water-side flow rate under variable operating conditions relative to the water-side flow rate under reference operating conditions. cpw r represents the change in water-side specific heat under varying operating conditions relative to the reference operating condition. vw r represents the change in water-side viscosity under varying operating conditions relative to the water-side viscosity under reference operating conditions. kmw r represents the change in water-side thermal conductivity under varying operating conditions relative to the reference operating condition. u This represents the change in the overall heat transfer coefficient of the evaporator under varying operating conditions relative to the overall heat transfer coefficient of the evaporator under reference operating conditions.

[0077] like Figure 1 As shown, logarithmic transformations are performed on the changes in each boundary parameter of the variable operating condition thermodynamic calculation model of the waste heat boiler evaporator relative to the baseline operating condition boundary parameter, and on the change in the overall heat transfer coefficient in the variable operating condition thermodynamic calculation results of the waste heat boiler evaporator relative to the baseline operating condition heat transfer coefficient. These transformations are then used as the features and prediction results of the multiple linear regression algorithm, respectively.

[0078]

[0079]

[0080] In the formula: X represents the characteristics of the multiple linear regression algorithm; Y represents the prediction result of the multiple linear regression algorithm.

[0081] Based on the calculation results of the variable operating condition thermodynamic calculation model of the waste heat boiler evaporator, a multiple linear regression algorithm is constructed using the features and prediction results.

[0082] + + + +

[0083] In the formula: a0~a7 are all training results of the multiple linear regression algorithm.

[0084] like Figure 1 As shown, the features and prediction results of the multiple linear regression algorithm are used as the dataset for the multiple linear regression algorithm. The dataset is randomly divided into a training set (accounting for two-thirds of the dataset) and a test set (accounting for one-third of the dataset) according to the ratio of one-third to one-third of the dataset.

[0085] like Figure 1 As shown, the multiple linear regression algorithm is trained based on the training set, and a simplified thermodynamic calculation model for the waste heat boiler evaporator under varying operating conditions is constructed based on the training results, namely:

[0086]

[0087] Example:

[0088] Taking the design operating condition of a three-pressure waste heat boiler in a combined cycle unit as the baseline operating condition, and performing variable-condition thermodynamic calculations of the waste heat boiler evaporator within the typical variation range of its boundary parameters, the changes in each boundary parameter of the variable-condition thermodynamic calculation model of the waste heat boiler evaporator relative to the baseline operating condition boundary parameters, and the changes in the overall heat transfer coefficient in the variable-condition thermodynamic calculation results of the waste heat boiler evaporator relative to the baseline operating condition heat transfer coefficient were calculated. Some calculation results are shown in Table 1. The calculation results in Table 1 were divided into training and testing sets for training and testing of multiple linear regression. The training results are shown below. Figure 2 As shown in the figure (the parameters in the figure are the coefficients a0~a7 of the multiple linear regression algorithm).

[0089] A simplified thermodynamic calculation model for a waste heat boiler evaporator under varying operating conditions is constructed based on the training results. The overall heat transfer coefficient calculated by this model under both the training and test sets is compared with the actual values. Figure 3 As shown, the deviation is as follows Figure 4 As shown. Figure 4 The results show that, under both the training and testing sets, the maximum deviation between the total heat transfer coefficient calculated by the simplified model and the true value of the total heat transfer coefficient is only 0.08%, indicating that the simplified model has good accuracy.

[0090] Table 1. Variations of boundary parameters and overall heat transfer coefficient relative to the baseline condition under varying operating conditions.

[0091]

Claims

1. A method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression, characterized in that, Includes the following steps: Step 1: Use the traditional waste heat boiler evaporator variable operating condition thermodynamic calculation model to identify the boundary parameters affecting the variable operating condition thermodynamic performance of the waste heat boiler evaporator. Step 2: Using the traditional waste heat boiler evaporator variable operating condition thermodynamic calculation model, and within the range of boundary parameter variation determined in Step 1, perform variable operating condition thermodynamic calculations of the waste heat boiler evaporator to obtain a series of calculation results; Step 3: Take the thermodynamic calculation results of the waste heat boiler evaporator under the design operating condition as the benchmark operating condition. Take the changes of the boundary parameters taken in the variable operating condition thermodynamic calculation of the waste heat boiler evaporator in Step 2 relative to the benchmark operating condition boundary parameters and perform logarithmic transformation as the feature of the multiple linear regression algorithm. Take the changes of the overall heat transfer coefficient in the variable operating condition thermodynamic calculation results of the waste heat boiler evaporator in Step 2 relative to the overall heat transfer coefficient of the benchmark operating condition and perform logarithmic transformation as the prediction result of the multiple linear regression. Step 4: Use the features and prediction results of the multiple linear regression algorithm determined in Step 3 as the dataset for the multiple linear regression algorithm. Divide the dataset into a training set and a test set, and train the multiple linear regression algorithm. Based on the training results, construct a simplified calculation model for the variable operating condition thermodynamic calculation of the waste heat boiler evaporator; the calculation formula is as follows. Based on the calculation results of the variable operating condition thermodynamic calculation model of the waste heat boiler evaporator, a multiple linear regression algorithm is constructed using the features and prediction results. + + + + In the formula: a0~a7 are all training results of the multiple linear regression algorithm; r wg This refers to the change in flue gas flow rate under varying operating conditions relative to the flue gas flow rate under the baseline operating conditions. r cpg r represents the change in flue gas specific heat under varying operating conditions relative to the reference operating condition. vg r represents the change in flue gas viscosity under varying operating conditions relative to the reference flue gas viscosity; kmg This represents the change in the thermal conductivity of the flue gas side under varying operating conditions relative to the thermal conductivity of the flue gas side under the reference operating conditions. r ww This refers to the change in water-side flow rate under variable operating conditions relative to the water-side flow rate under the baseline operating conditions. r cpw r represents the change in water-side specific heat under varying operating conditions relative to the reference operating condition. vw r represents the change in water-side viscosity under varying operating conditions relative to the water-side viscosity under reference operating conditions. kmw This represents the change in the water-side thermal conductivity under variable operating conditions relative to the water-side thermal conductivity under the reference operating conditions. r u This represents the change in the overall heat transfer coefficient of the evaporator under varying operating conditions relative to the overall heat transfer coefficient of the evaporator under reference operating conditions. The multiple linear regression algorithm is trained based on the training set, and a simplified thermodynamic calculation model for the waste heat boiler evaporator under varying operating conditions is constructed based on the training results, namely: 。 2. The method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression according to claim 1, characterized in that, In steps 1 and 2, the traditional thermodynamic calculation model for variable operating conditions of a waste heat boiler evaporator is as follows: The water-side heat transfer coefficient of the evaporator is obtained by solving the water-side parameters of the waste heat boiler evaporator under varying operating conditions and the geometric parameters of the evaporator tube bundle diameter and number. Where: h w The water-side heat transfer coefficient in a variable-condition evaporator is given in W / m³. 2 .℃; The specific heat on the water side of the evaporator is expressed in J / kg·℃. The viscosity of the water side in the evaporator is kg / ms; The thermal conductivity of the evaporator water side is W / m·℃; d i The inner diameter of the evaporator tube bundle is in mm.

3. The method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression according to claim 2, characterized in that, In steps 1 and 2, the formula for calculating the heat transfer coefficient of the flue gas side of the waste heat boiler evaporator under varying operating conditions is as follows: Where: h g The heat transfer coefficient on the flue gas side of the evaporator under variable operating conditions, in W / m³. 2 .℃; C1, C3, and C5 are coefficients; is the outer diameter of the evaporator tube bundle, mm; h is the fin height, mm; The average temperature of the inlet and outlet flue gas is expressed in °C. Temperature at the fin, in °C; The value is the flue gas flow rate, in kg / s; Specific heat of flue gas, J / kg·℃; Viscosity of the evaporator flue gas, kg / ms; is the thermal conductivity of the evaporator flue gas side, W / m·℃.

4. The method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression according to claim 3, characterized in that, In steps 1 and 2, the formula for calculating the overall heat transfer coefficient of the waste heat boiler evaporator under varying operating conditions is as follows: In the formula: The overall heat transfer coefficient in a variable-condition evaporator is given in W / m³. 2 .℃; A t The outer surface area of ​​the fin per unit fin length, in meters. 2 / m;A i The inner surface area of ​​the fin per unit fin length, in meters. 2 / m;ff i The fouling factor inside the finned tube is m. 2 ℃ / W; A w The average inner and outer surface areas of the fin per unit fin length, in meters. 2 / m;K m The thermal conductivity of the fin wall is W / m. 2 .℃;ff o The fouling factor on the outside of the finned tube, m 2 ℃ / W; η is the efficiency of the fin.

5. The method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression according to claim 4, characterized in that, In steps 1 and 2, the logarithmic mean temperature difference of the evaporator is calculated based on the overall heat transfer coefficient of the waste heat boiler evaporator under varying operating conditions: Where: LMTD is the logarithmic mean temperature difference of the evaporator; e w2 The enthalpy of the water at the evaporator inlet is given in kJ / kg; e w1 A is the enthalpy of water at the evaporator outlet, kJ / kg. evp The heat exchange area of ​​the evaporator is in meters (m). 2 .

6. The method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression according to claim 5, characterized in that, In steps 1 and 2, the logarithmic mean temperature difference of the evaporator can also be obtained through thermodynamic calculations using the inlet and outlet flue gas side and water side parameters of the waste heat boiler evaporator: In the formula: T g1 T represents the inlet flue gas temperature of the evaporator, in °C. g2 T represents the evaporator outlet flue gas temperature, in °C. w1 The temperature of the water at the evaporator inlet is ℃.

7. The method for modeling variable operating conditions of a waste heat boiler evaporator based on linear regression according to claim 6, characterized in that, In step 3, the calculation formula is as follows: The calculation results under the design operating condition of the waste heat boiler are used as the benchmark operating condition, that is: In the formula: The overall heat transfer coefficient of the evaporator under the reference operating conditions is expressed in W / m². 2 .℃;W g,b The flue gas flow rate is given under the baseline operating conditions, in kg / s. The specific heat of the flue gas under the reference operating conditions is J / kg·℃; The viscosity of the flue gas under the reference operating conditions is kg / ms; km. g,b The thermal conductivity of the flue gas side under the reference operating conditions is W / m. 2 .℃;W w,b The water flow rate under the baseline operating conditions is expressed in kg / s. The water-side specific heat is given under the baseline operating conditions, in J / kg·℃. Water-side viscosity under reference operating conditions, kg / ms; km w,b The thermal conductivity of the water side under the reference operating conditions is W / m. 2 .℃; Thermodynamic calculations for the waste heat boiler evaporator under varying operating conditions are performed within typical ranges of flue gas temperature, flow rate, composition, and water-side temperature and pressure. The changes in each boundary parameter of the thermodynamic calculation model under varying operating conditions relative to the baseline operating condition boundary parameters, and the change in the overall heat transfer coefficient in the thermodynamic calculation results under varying operating conditions relative to the baseline operating condition heat transfer coefficient are also calculated. In the formula: W g,o The flue gas flow rate under varying operating conditions is expressed in kg / s. The specific heat of the flue gas under varying operating conditions is expressed in J / kg·℃. For flue gas viscosity under varying operating conditions, kg / ms; km g,o The thermal conductivity of the flue gas side under varying operating conditions, W / m 2 .℃;W w,o The water flow rate under varying operating conditions is expressed in kg / s. The specific heat of the water side under varying operating conditions is expressed in J / kg·℃. For water-side viscosity under varying operating conditions, kg / ms; km w,o Thermal conductivity of the water side under varying operating conditions, W / m 2 .℃; Logarithmic transformations were performed on the changes in each boundary parameter of the variable operating condition thermodynamic calculation model of the waste heat boiler evaporator relative to the baseline operating condition boundary parameter, and on the change in the overall heat transfer coefficient in the variable operating condition thermodynamic calculation results of the waste heat boiler evaporator relative to the baseline operating condition heat transfer coefficient. These transformations were then used as the features and prediction results of the multiple linear regression algorithm, respectively. In the formula: X represents the characteristics of the multiple linear regression algorithm; Y represents the prediction result of the multiple linear regression algorithm.

Images