An optimization method for the best operating conditions of an oxygen-enriched combustion boiler

Abstract

The present invention relates to an optimization method for the optimal operating conditions of an oxy-fuel combustion boiler, which includes the following steps: Based on the structure and dimensions of the oxy-fuel combustion boiler in Ansys Fluent, select a fluid dynamics model, determine the operating parameters required for numerical simulation calculations, and determine the results output through numerical simulation calculations; Use the genetic algorithm in MATLAB to determine the number of variables and the limit range according to the determined operating parameters required for input, generate an initial population, and set the operating parameters in the Ansys Fluent numerical simulation calculations; Construct a numerical simulation calculation model for boiler combustion; Construct a Pareto front multi-objective optimization model with four parameter optimization objectives including the effective heat utilization of the boiler, the flame center position, the highest flame temperature, and the boiler thermal efficiency, set the termination conditions, perform multi-objective optimization calculations, and obtain the optimal operating parameters.

Description

Technical Field

[0001] The invention belongs to the field of regulating oxygen-enriched combustion boilers, and particularly relates to an optimization method for the optimal operating conditions of oxygen-enriched combustion boilers. Background Art

[0002] With the improvement of environmental protection standards, power generation enterprises or heating enterprises are expected to reduce the emissions of pollutants (NO x ) in coal-fired boilers, and on this basis, to improve the combustion efficiency of pulverized coal as much as possible, ensure the stable and safe operation of the boiler, and ensure the effective utilization of heat in the boiler.

[0003] Oxygen-enriched combustion technology is an efficient energy-saving combustion technology and an emission reduction technology that is relatively easy to implement. This technology mixes pure oxygen and recirculated flue gas mainly composed of CO 2 in a certain proportion (the mixed gas is called auxiliary combustion gas or combustion gas), and then sends it into the furnace to mix and burn with the fuel. Since almost all nitrogen in the traditional combustion method is eliminated, the CO 2 concentration in the flue gas generated by oxygen-enriched combustion is relatively high. For the flue gas containing high-concentration CO 2 generated by combustion, a part of it is mixed with pure oxygen in a recirculation manner and then enters the furnace for drying, transporting the fuel, and controlling the combustion temperature in the furnace; the other part can be directly sent into the compression equipment for CO 2 storage after relatively simple impurity removal treatment, such as condensation, drying, etc. This technology is applicable to both the transformation of old boilers in power plants and the construction of new boilers, and has the advantages of low cost, high reliability, and good inheritance with the existing boiler combustion technology. It is also the most easily accepted CO 2 emission reduction technology route.

[0004] When adopting oxygen-enriched combustion technology, N 2 is replaced by CO 2 . Due to the specific volume and gas radiation characteristics of CO 2 , the local maximum temperature will be reduced, and at the same time, the flame center will move downward, and the combustion characteristics are worse than those of air combustion. Changing the mixing ratio of O 2 and CO 2 , and adjusting the proportion of the primary air volume, secondary air volume, and tertiary air volume can improve the comprehensive performance of the oxygen-enriched combustion boiler. Therefore, finding the optimal mixing ratio of O 2 and CO 2 , and adjusting the proportion of the primary air volume, secondary air volume, and tertiary air volume is of great significance for the application of oxygen-enriched combustion boilers. Summary of the Invention

[0005] The present invention provides an optimization method for the optimal operating conditions of an oxy-fuel combustion boiler. This method is realized based on the co-simulation of Ansys Fluent and MATLAB, and filters out the operating condition parameters (the optimal ratio of O 2 and CO 2 : k, 1-k; the optimal ratio of the primary air volume, secondary air volume and tertiary air volume: a, b, c, a + b + c = 1) whose comprehensive performance is closer to that of air combustion when using oxy-fuel combustion. To achieve the above object, the technical solution of the present invention is as follows:

[0006] An optimization method for the optimal operating conditions of an oxy-fuel combustion boiler, which is realized based on the co-simulation of Ansys Fluent and MATLAB, includes the following steps:

[0007] Step 1: Based on the structure and size of the oxy-fuel combustion boiler in Ansys Fluent, select a fluid dynamics model, determine the operating condition parameters required for numerical simulation calculations, and determine the results output through numerical simulation calculations. The results output through numerical simulation calculations determined include four parameters: the effectively utilized heat of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency;

[0008] Step 2: Using the genetic algorithm in MATLAB, for the operating condition parameters required to be input determined in Step 1, determine the number of variables and the limit range, generate an initial population, and set the operating condition parameters in the Ansys Fluent numerical simulation calculations.

[0009] Step 3: Build a numerical simulation calculation model of boiler combustion in Ansys Fluent, and start numerical simulation calculations using the built numerical simulation calculation model and output the calculation results.

[0010] Step 4: According to the calculation results output in Step 3, build a Pareto front multi-objective optimization model with the four parameters of the effectively utilized heat of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency as the optimization target parameters in the genetic algorithm in MATLAB, set the termination conditions, and perform multi-objective optimization calculations;

[0011] Step 5: If the termination conditions set in Step 4 are not met, the population undergoes mutation and cross-selection. After filtering out unfavorable mutations based on the existing results, return to Step 3 for calculation until the termination conditions are met, end the mutation, output the optimal results, and obtain the optimal operating condition parameters.

[0012] Preferably, in Step 1, the operating condition parameters required for numerical simulation calculations determined include: O 2 and CO 2The ratio is k:(1 - k); the ratio of the primary air flow rate, secondary air flow rate, and tertiary air flow rate is a:b:(c = 1 - a - b).

[0013] In step 4, the specific steps for constructing the Pareto front multi-objective optimization model are as follows:

[0014] (1) The four parameters of the effective heat utilization of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency are used as optimization objectives and expressed as follows:

[0015] f j , j = 1, 2, 3, 4 (1)

[0016] Among them, j represents the serial number of the optimization objective parameter, and f 1 , f 2 , f 3 , f 4 respectively represent the effective heat utilization of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency;

[0017] The result set obtained from numerical simulation calculations is expressed as follows:

[0018] f ij , i = 1, 2, 3, …, n; j = 1, 2, 3, 4 (2)

[0019] Among them, i represents different population individuals, with a total of n, and f i1 , f i2 , f i3 , f i4 respectively represent the values of the effective heat utilization of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency of different individuals in the population;

[0020] (2) The effective heat utilization Q 0 of the boiler, the position of the flame center h 0 , the highest flame temperature T 0 , and the boiler thermal efficiency η 0 obtained by air combustion are set as target values and expressed as follows:

[0021]

[0022] (3) The target values are dimensionless processed and expressed as The result data set is dimensionless processed and expressed as F ij ;

[0023] (4) Calculate the degree of closeness between the calculation result under the working condition parameters and the target value. The closer to the target value, the closer to the optimal result. The calculation method of the degree of closeness is as follows:

[0024]

[0025] Set the termination calculation value d * , and set the termination condition as the proximity degree d between the calculation result and the target value ij less than or equal to the termination calculation value d * ;

[0026] When the termination condition is met, obtain the minimum distance between the calculation result and the target value;

[0027] Select the population individual corresponding to the minimum distance S * , as well as the corresponding values of k, a, b, and c, and obtain the optimal operating parameters.

[0028] The formula for dimensionless processing of the target value in step (3) is as follows:

[0029]

[0030] The formula for dimensionless processing of the result data set in step (3) is as follows:

[0031]

[0032] The termination calculation value is:

[0033] Advantages of the present invention:

[0034] 1. This method is realized based on the co-simulation of Ansys Fluent and MATLAB, and can quickly find the operating conditions where the combustion characteristics of the boiler under oxy-fuel combustion are close to those of the boiler under air combustion, that is, the ratio k of O 2 , then the ratio of CO 2 is 1 - k; the primary air volume ratio a; the secondary air volume ratio b, and the tertiary air volume ratio is c.

[0035] 2. This method constructs a Pareto front LINMAP multi-objective optimization model in the optimization process, comprehensively considers various important parameters in boiler combustion, and the found optimal operating conditions are relatively close to the boiler characteristics under air combustion in all aspects. Description of the Drawings

[0036] Figure 1 Flowchart of an optimization method for the optimal operating conditions of an oxy-fuel combustion boiler

[0037] Figure 2 System diagram of an oxy-fuel combustion boiler

[0038] Appendix Figure 2Among them, 1. Cold ash hopper, 2. Primary air inlet, 3. Secondary air inlet, 4. Tertiary air inlet, 5. Water wall, 6. Separating platen superheater, 7. Rear platen superheater, 8. Final stage superheater, 9. Final stage reheater, 10. Vertical low-temperature superheater, 11. Vertical low-temperature superheater, 12. Horizontal low-temperature superheater, 13. Economizer. Specific implementation mode

[0039] The present invention is an improved strategy proposed for the problems existing in the prior art. To better illustrate the present invention, the following is combined with the attached Figure 1 and 2 to further illustrate the present invention.

[0040] The present invention is implemented based on Ansys Fluent and MATLAB. Therefore, it is crucial to realize the co-simulation of Ansys Fluent and MATLAB. The present invention needs to start MATLAB and Ansys Fluent simultaneously, and realizes indirect parameter transfer by accessing the shared data folder through the I / O operation of the file.

[0041] Based on the realization of the above co-simulation of Ansys Fluent and MATLAB, the present invention specifically includes the following steps:

[0042] Step 1: Build an oxy-fuel combustion boiler model in Ansys Fluent based on the actual structure and size of the boiler. The specific model is as Figure 2 shown. Specifically, it consists of a cold ash hopper, a primary air inlet 2, a secondary air inlet 3, a tertiary air inlet 4, a water wall 5, a separating platen superheater 6, a rear platen superheater 7, a final stage superheater 8, a final stage reheater 9, a vertical low-temperature superheater 10, a vertical low-temperature superheater 11, a horizontal low-temperature superheater 12, and an economizer 13. The following models are adopted in the numerical simulation calculation process: the N-S equation in fluid dynamics, the turbulent Realizable k-ε model, the Euler-Lagrange method stochastic orbit model, the two-step competing reaction rate model, the P1 radiation heat transfer model, and the turbulent chemical reaction is the Eddy Dissipation Model (EDM). The working condition parameters required for the numerical simulation calculation include: the ratio of O 2 and CO 2 : k, 1-k; the ratio of the primary air volume, secondary air volume, and tertiary air volume: a, b, c, c = 1 - a - b. The numerical simulation calculation results include four parameters: the effectively utilized heat of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency.

[0043] Step 2: The genetic algorithm in the MATLAB program generates an initial population according to the number of variables and the limit range: randomly generate N initial data, each data is an individual, and N individuals form a population. The chromosome of each individual consists of three independent characteristic values, which are as follows: (1) O2 The ratio k of CO 2 is 1 - k; (2) the primary air flow rate ratio a; (3) the secondary air flow rate ratio b, then the tertiary air flow rate ratio is c = 1 - a - b. And input them into Ansys Fluent and set them as the working condition parameters required for numerical simulation calculation.

[0044] Step 3: The working condition parameters required for the Ansys Fluent numerical simulation calculation in Step 1 and the settings in Step 2 are input and set after being encoded by MATLAB, and then the numerical simulation calculation starts from the model established in Ansys Fluent in Step 1 and the calculation results are output.

[0045] Step 4: The calculation results in Step 3 are input into the Pareto - front multi - objective optimization model based on the genetic algorithm in MATLAB with the four parameters of the effective heat utilization of the boiler, the flame center position, the maximum flame temperature, and the boiler thermal efficiency as the optimization target parameters, which is called the LINMAP multi - objective optimization model, and multi - objective optimization calculation is carried out. The Pareto - front LINMAP multi - objective optimization model is as follows:

[0046] Taking the four parameters of the effective heat utilization of the boiler, the flame center position, the maximum flame temperature, and the boiler thermal efficiency as the optimization objectives is expressed as follows:

[0047] f j , j = 1, 2, 3, 4 (1)

[0048] j represents the optimization target parameters, with a total of 4. f 1 , f 2 , f 3 , f 4 respectively represent the effective heat utilization of the boiler, the flame center position, the maximum flame temperature, and the boiler thermal efficiency.

[0049] Expressing the result set obtained from the numerical simulation calculation as follows:

[0050] f ij , i = 1, 2, 3, …, n; j = 1, 2, 3, 4 (2)

[0051] i represents different population individuals, with a total of n. f i1 , f i2 , f i3 , f i4 respectively represent the effective heat utilization of the boiler, the flame center position, the maximum flame temperature, and the boiler thermal efficiency of different individuals in the population.

[0052] Taking the effective heat utilization of the boiler (Q 0 ) obtained by air combustion, the flame center position (h 0)、The maximum flame temperature (T 0 ) and the boiler thermal efficiency (η 0 ) are set as target values, which are respectively expressed as follows:

[0053]

[0054] If the dimensions of the four parameters of the effective heat utilization of the boiler, the flame center position, the maximum flame temperature and the boiler thermal efficiency are different, in order to ignore the influence of different dimensions, it is necessary to non-dimensionalize the target values and the result data sets respectively during the multi-objective optimization calculation:

[0055]

[0056]

[0057] Calculate the degree of closeness between the calculation results under the operating conditions and the target values. The closer to the target value, the closer to the optimal result. The calculation method of the degree of closeness is as follows:

[0058]

[0059] Set the termination condition as the degree of closeness between the calculation results and the target values being 95%. Specifically, it is expressed as follows:

[0060]

[0061] When the value of (6) is less than or equal to formula (7), the termination condition is satisfied, and the minimum distance between the result and the target value is obtained, which is expressed as follows:

[0062] S * =Min(d ij ), i = 1, 2, 3, …, n; j = 1, 2, 3, 4 (8)

[0063] The obtained minimum distance S * Select the population individuals corresponding to this distance, as well as the corresponding values of k, a, b, and c, to obtain the optimal operating conditions parameters.

[0064] Step 5: If the termination condition of formula (7) in Step 4 is not satisfied, the initial population undergoes mutation and cross-selection. After screening out the unfavorable mutations according to the existing results, return to Step 3 for calculation until the termination condition of formula (7) is satisfied, end the mutation, and output the optimal result according to formula (8) (comprehensively compare the values of k, a, b, and c when closest to the target value).

[0065] The above is only an illustration of the specific embodiments of the present invention and should not be used to limit the scope of the protection of the rights of the present invention. All equivalent changes and modifications made according to the claims and the content of the specification of the present invention application are within the scope of protection of the present invention.

Claims

1. An optimization method for the optimal operating conditions of an oxy-fuel combustion boiler, realized based on the co-simulation of Ansys Fluent and MATLAB, includes the following steps: Step 1: Based on the structure and dimensions of the oxy-fuel combustion boiler in Ansys Fluent, select a fluid dynamics model, determine the operating parameters required for numerical simulation calculations, and determine the results output through numerical simulation calculations. The results output through numerical simulation calculations include: Four parameters: the effectively utilized heat of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency; Step 2: Use the genetic algorithm in MATLAB, the operating parameters required to be input determined in Step 1, determine the number of variables and the limit range, generate an initial population, and set the operating parameters in the Ansys Fluent numerical simulation calculation; Step 3: Build a numerical simulation calculation model for boiler combustion in Ansys Fluent, and start numerical simulation calculations using the built numerical simulation calculation model and output the calculation results; Step 4: According to the calculation results output in Step 3, build a Pareto front multi-objective optimization model with the four parameters of the effectively utilized heat of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency as the optimization target parameters in the genetic algorithm in MATLAB, set the termination conditions, and perform multi-objective optimization calculations; Step 5: If the termination conditions set in Step 4 are not met, the population mutates, and crossover selection is performed. After screening out unfavorable mutations based on the existing results, return to Step 3 for calculation until the termination conditions are met, end the mutation, output the optimal results, and obtain the optimal operating parameters.

2. The optimization method for the optimal operating conditions of the oxy-fuel combustion boiler according to claim 1, characterized in that In Step 1, the operating condition parameters to be input for numerical simulation calculation include: O 2 and CO 2 ratio, i.e., k:(1-k); the ratio of the primary air volume, secondary air volume and tertiary air volume, i.e., a:b:(c = 1-a-b).

3. The optimization method for the optimal operating conditions of the oxy-fuel combustion boiler according to claim 2, characterized in that In Step 4, the specific steps for building the Pareto front multi-objective optimization model are as follows: (1) Represent the four parameters of the effectively utilized heat of the boiler, the position of the flame center, the highest flame temperature, and the boiler thermal efficiency as optimization targets as follows: f j , j = 1, 2, 3, 4 (1) Among them, j represents the serial number of the optimization target parameter, and f 1 , f 2 , f 3 , f 4 respectively represent the effectively utilized heat of the boiler, the position of the flame center, the highest temperature of the flame, and the thermal efficiency of the boiler; The result set obtained from numerical simulation calculations is represented as follows: f ij , where i = 1, 2, 3, …, n; j = 1, 2, 3, 4 (2) Among them, i represents different population individuals, with a total of n, and f i1 , f i2 , f i3 , f i4 respectively represent the values of the effective heat utilization of the boiler, the position of the flame center, the highest temperature of the flame, and the thermal efficiency of the boiler for different individuals in the population; (2) The effective heat utilization Q of the boiler obtained by air combustion 0 , the flame center position h 0 , the maximum flame temperature T 0 and the boiler thermal efficiency η 0 are set as target values and are respectively expressed as follows: (3) Nondimensionalize the target value, expressed as Nondimensionalize the result data set, expressed as F ij ; (4) Calculate the degree of closeness between the calculation results under the operating parameters and the target values. The closer to the target value, the closer it represents to the optimal result. The calculation of the degree of closeness is as follows: Set the termination calculation value d * , and set the termination condition as the proximity degree d between the calculation result and the target value ij less than or equal to the termination calculation value d * ; when the termination condition is met, obtain the minimum distance between the calculation result and the target value; select the population individual corresponding to the minimum distance S * , as well as the corresponding k, a, b, and c values, to obtain the optimal working condition parameters 4. The optimization method for the optimal operating conditions of the oxy-fuel combustion boiler according to claim 3, characterized in that The formula for non-dimensionalizing the target value in Step (3) is as follows:

5. The optimization method for the optimal operating conditions of the oxy-fuel combustion boiler according to claim 4, characterized in that The formula for non-dimensionalizing the result data set in Step (3) is as follows:

6. The optimization method for the optimal operating conditions of the oxy-fuel combustion boiler according to claim 5, characterized in that Terminate calculated value 7. The optimization method for the optimal operating conditions of the oxy-fuel combustion boiler according to claim 1, characterized in that The selected hydrodynamic models include the Navier-Stokes equations, the turbulent Realizable k-ε model, the Euler-Lagrange method stochastic orbit model, the two-step competing reaction rate model, the P1 radiation heat transfer model, and the turbulent chemical reaction model.

8. The optimization method for the optimal operating conditions of an oxygen-enriched combustion boiler according to claim 7, characterized in that the turbulent chemical reaction model is the Eddy Dissipation Model (EDM).

Images