Optimal circulation heat storage performance calcium-based spherical particle preparation combination prediction method based on optimization algorithm
Through the method based on the optimization algorithm, combined with multi-component experiments and rotary granulation method, a combination prediction model for the optimal cyclic heat storage performance preparation of calcium-based pellets is constructed, which solves the problem that the preparation combination of calcium-based pellets cannot be accurately known in the prior art, and achieves rapid and accurate prediction and cost reduction.
Patent Information
- Application Number
- CN202510203282.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-24
- Publication Date
- 2025-05-27
AI Technical Summary
In the prior art, the preparation combination of calcium-based pellets cannot accurately obtain the optimal cyclic heat storage performance, resulting in high costs and long time.
The method based on the optimization algorithm was used to prepare calcium-based pellets through multi-component experiments and rotary granulation method. Multiple cyclic heat storage performance experiments were performed using a horizontal fixed bed experimental bench to construct a combination prediction model for the optimal cyclic heat storage performance preparation of calcium-based pellets.
The optimal preparation combination of calcium-based pellets is achieved quickly and accurately predicted, which reduces production costs and energy consumption, and solves the problem that preparation combinations cannot be accurately known in the prior art.
Smart Images
Figure CN120048405A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of energy technologies, and particularly to a combined prediction method for preparing calcium-based pellets with optimal cyclic heat storage performance based on an optimization algorithm. Background Art
[0002] Currently, the working temperature of the third-generation solar thermal power plant technology based on supercritical CO 2 power cycle exceeds 700 °C, and it is increasingly urgent to develop energy storage technologies for high-temperature operation. Thermochemical energy storage (TCES) is an important medium- and high-temperature energy storage technology. Among thermochemical energy storage materials, the calcium looping (CaL) technology, namely the heat storage based on the calcination / carbonation cycle of CaCO 3 (reversible reaction: ΔH = 178 kJ / mol), has great application potential in the third-generation CSP system due to its high energy storage density (about 3.2 GJ / m3), high reaction temperature (750 - 950 °C), and low cost.
[0003] In most of the existing inventions, powdered calcium-based materials are selected, but powdered materials are not suitable for use in actual cyclic heat storage systems, resulting in high costs, long time, and inability to accurately obtain the preparation combination conditions of calcium-based pellets under the optimal cyclic heat storage performance. Summary of the Invention
[0004] Object of the Invention: The object of the present invention is to provide a combined prediction method for preparing calcium-based pellets with optimal cyclic heat storage performance based on an optimization algorithm, so as to solve the problems of excessively high costs, long time, and inability to accurately obtain the preparation combination of calcium-based pellets under the optimal cyclic heat storage performance in the existing technical methods.
[0005] Technical Solution: The combined prediction method for preparing calcium-based pellets with optimal cyclic heat storage performance based on an optimization algorithm according to the present invention is based on multi-component experiments and includes the following steps:
[0006] (1) Select different types of biomass pore formers and different contents of inert carriers, and use the rotary granulation method to prepare experimental calcium-based pellets.
[0007] (2) Use a horizontal fixed-bed experimental bench to conduct multiple cyclic heat storage performance experiments and form a database.
[0008] (3) Based on the database and the optimization algorithm, construct a prediction model for the preparation combination of calcium-based pellets with the optimal cyclic heat storage performance.
[0009] Further, step (1) is specifically as follows: During the preparation of calcium-based pellets, the types of biomass pore-forming agents are rice husk, wood chips, bagasse, and peanut shell; then they are all subjected to crushing treatment and passed through a 200-mesh sieve to obtain 0.075-mm particles. Among them, the blending ratios of different types of biomass pore-forming agents are 2%, 5%, 10%, 15%, and 20%; the inert carrier is TiO 2 , and the mass fractions are 0%, 2%, 5%, 10%, and 20% respectively; during the granulation process, it is pre-calcined in a tubular furnace at 850 °C, with an air flow rate of 1.5 L / min for 15 min; the finally formed calcium-based pellets have three particle sizes of 0.6 - 1 mm, 1 - 1.5 mm, and 1.5 - 2 mm.
[0010] Further, in step (1), during the experiment, CaCO3, TiO2, and the biomass pore-forming agent are first placed in an oven and dried at 90 - 95 °C for 14 - 16 h; after physical mixing, a premixed powder is obtained. The powder is placed in a laboratory granulation drum and rotated at a specified speed of 30 - 35 revolutions per minute. A polyvinylpyrrolidone PVP solution with a concentration of 0.1 - 0.015 mg / ml is sprayed at intervals of 10 - 15 min to form seed particles; after the formation process of the seed particles is stable, an appropriate amount of dry powder is added to adjust the particle size; when the pellet particle size reaches the required size, tumbling is carried out at 90 - 95 revolutions per minute for 45 - 60 min.
[0011] Further, step (2) is specifically as follows: The cyclic heat storage experiment is carried out on a fixed-bed experimental bench. The sample is placed in the constant-temperature section at 850 °C for the carbonation reaction. After the reaction tube is sealed, CO 2 is introduced at a gas flow rate of 1.5 L / min, and after pressurization, the calcium-based pellet sample is fed in; during the process of taking out the sample, CO 2 is continuously introduced, and the sample is placed in a drying dish to cool; after the sample is weighed, the reaction gas is switched, and calcination is carried out at 850 °C in a pure N 2 atmosphere with a gas flow rate of 1 L / min. After the calcination is completed, the sample is quickly taken out, weighed, and then the sample is put back again to repeat the above steps, which is the pressurized carbonation cyclic heat storage experiment of calcium-based pellets.
[0012] Further, for the multi-round cyclic pressurized carbonation heat storage performance experiment of calcium-based pellets, the pressure range is atmospheric pressure, 0.2 - 0.8 MPa, and the pressure is increased in steps of 0.2 MPa, with a total of 5 groups of pressures; during the weighing process, it is easy to react with CO2 in the air. For the same pellet sample, 3 groups of repeated experiments are carried out, and the average value is taken; the inner diameter of the central tube of the fixed-bed reactor for the experimental level is 74 mm, and the heating length is 1000 mm, among which the central constant-temperature section is 25 mm.
[0013] Further, step (3) is specifically as follows: Based on the database and the collaborative filtering algorithm based on the regression model, a combined prediction model for the preparation of the best cyclic heat storage performance of calcium-based pellets is constructed. The particle size of the pelletizing particles, the type and content of the biomass pore-forming agent, and the mass fraction of the inert component are input as 4 groups of data, and the cyclic heat storage performance under different pressures is used as the data output. The heat storage performance is measured by the effective conversion rate of the heat release cycle of the calcium-based material and the heat storage density, and the sum is obtained by multiplying their respective cycle numbers. The highest total value is the optimal preparation combination of calcium-based pellets.
[0014] Further, let the optimal cyclic heat storage performance be P, the mass fraction of TiO 2 be T, the type of biomass pore-forming agent be B, the corresponding content of the biomass pore-forming agent be C, and the pellet diameter be D. Find the mathematical relationship between P and the variables T, B, C, D. Among them, the formula of SVR is as follows:
[0015]
[0016] where x = [T, B, C, D], the input variables; x i = [T i , B i , C i , D i , the variable values of the support vectors; K(x i , x), the kernel function, used to calculate the similarity between x i and x; α i and weights, which determine the contribution of each support vector to the prediction; b, the bias term, is a constant;
[0017] It is stipulated that the kernel function is the Gaussian kernel function, as follows:
[0018] K(x i , x) = exp(-γ||x i - x|| 2 )
[0019] By setting the error function to be minimized, as follows:
[0020]
[0021] where w is the difference between the true value and the predicted value in the database, C is the regularization parameter (the smaller C, the simpler the model; the larger C, the lower the model error), the slack variable, used to measure the error beyond the tolerance range;
[0022] Finally, the prediction result of the preparation combination under the best cyclic heat storage performance of calcium-based pellets is obtained.
[0023] A combined prediction system for preparing calcium-based pellets with the best cyclic heat storage performance based on an optimization algorithm according to the present invention includes:
[0024] A preparation module: used to select different types of biomass pore-forming agents and different contents of inert carriers, and use the rotary granulation method to prepare experimental calcium-based pellets;
[0025] An experiment module: used to conduct multiple cyclic heat storage performance experiments using a horizontal fixed-bed experimental bench and form a database;
[0026] An algorithm module: used to construct a combined prediction model for preparing calcium-based pellets with the best cyclic heat storage performance based on the database and the optimization algorithm.
[0027] An electronic device according to the present invention includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is loaded into the processor, it realizes a combined prediction method for preparing calcium-based pellets with the best cyclic heat storage performance based on an optimization algorithm according to any one of the above.
[0028] A storage medium according to the present invention stores a computer program, and when the computer program is executed by a processor, it realizes a combined prediction method for preparing calcium-based pellets with the best cyclic heat storage performance based on an optimization algorithm according to any one of the above.
[0029] Beneficial effects: Compared with the prior art, the present invention has the following remarkable advantages: The present invention selects different types of biomass pore-forming agents and different contents of inert carriers, uses the rotary granulation method to prepare experimental calcium-based pellets, conducts multiple cyclic heat storage performance experiments using a horizontal fixed-bed experimental bench, and the combined prediction model for preparing calcium-based pellets with the best cyclic heat storage performance constructed based on experimental data and the optimization algorithm is fast in calculation and accurate in results, providing a beneficial effect for solving the problem of accurately preparing calcium-based pellets with the best cyclic heat storage performance, saving production costs and reducing energy consumption. Description of the Drawings
[0030] Figure 1 is the process of preparing calcium-based pellets of the present invention;
[0031] Figure 2 is a schematic diagram of a pressurized high-temperature fixed-bed reactor of the present invention;
[0032] Figure 3 is a schematic diagram of the size of the fixed-bed reactor of the present invention;
[0033] Figure 4 is a logic diagram of a combined prediction model for preparing calcium-based pellets with the best cyclic heat storage performance based on an optimization algorithm of the present invention. Detailed Embodiments
[0034] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.
[0035] As Figure 1 shown in -4, an embodiment of the present invention provides a combined prediction method for preparing calcium-based pellets with the best cyclic heat storage performance based on an optimization algorithm, including the following steps:
[0036] (1) Use the rotary granulation method to prepare experimental calcium-based pellets, and the preparation process is as Figure 1 shown. First, CaCO 3 , TiO 2 , and biomass pore formers (bagasse, wood chips, rice husks, peanut shells) are first placed in an oven and dried at 90 - 95 °C for 14 - 16 h; after physical mixing, a premixed powder is obtained. The powder is placed in a laboratory granulation drum and rotated at a specified speed (30 - 35 revolutions per minute), and a polyvinylpyrrolidone (PVP) solution with a concentration of 0.1 - 0.015 mg / ml is sprayed at intervals of 10 - 15 min to form seed particles; after the formation process of the seed particles is stable, an appropriate amount of dry powder is added to adjust the particle size; when the pellet particle size reaches the required size, a tumbling process at 90 - 95 revolutions per minute for 45 - 60 min is carried out to further improve its mechanical strength and form pellets. During the preparation of calcium-based pellets, the main types of biomass pore formers are rice husk wood chips, bagasse, and peanut shells (all obtained by crushing and passing through a 200-mesh sieve to obtain 0.075-mm particles), and the mixing ratios of different biomass pore formers are 2%, 5%, 10%, 15%, and 20%; the main inert carrier is TiO 2 , and the mass fractions are 0%, 2%, 5%, 10%, and 20% respectively; during the granulation process, it is pre-calcined in a tubular furnace at 850 °C with an air flow rate of 1.5 L / min for 15 min; the finally formed calcium-based pellets have three particle sizes: 1) 0.6 - 1 mm, 2) 1 - 1.5 mm, and 3) 1.5 - 2 mm.
[0037] (2) The cyclic heat storage experiment is carried out on a fixed-bed experimental bench, as Figure 2 shown. The sample is placed in the constant temperature section (850 °C) for the carbonation reaction. After the reaction tube is sealed, CO 2 is introduced at a gas flow rate of 1.5 L / min, and after pressurization, it is sent into the calcium-based pellet sample. During the process of taking out the sample, CO 2 is continuously introduced to avoid high-temperature calcination of the sample during the taking-out process, and then it is placed in a drying dish to cool. After the sample is weighed, the reaction gas is switched to pure N 2Calcination was carried out at 850 °C under an atmosphere, the gas flow rate was 1 L / min. After the calcination was completed, the sample was quickly taken out. After weighing, the sample was put in again to repeat the above steps, which was the pressurized carbonation cycle heat storage experiment of calcium-based pellets. For the multi-cycle pressurized carbonation heat storage performance experiment of calcium-based pellets, the pressure range was from atmospheric pressure to 0.2 - 0.8 MPa and the pressure was increased in steps of 0.2 MPa, with a total of 5 groups of pressures; since the reaction was carried out at high temperature, it was easy to react with CO in the air during the weighing process. 2 Therefore, for the same pellet sample, 3 sets of repeated experiments were carried out for the experiment, and the average value was taken to reduce the experimental error; the size of the horizontal fixed-bed platform described in this experiment was that the inner diameter of the central tube was 74 mm and the heating length was 1000 mm, where the central constant temperature section was 25 mm, as Figure 3 shown.
[0038] The effective conversion rate and heat storage density of the calcium-based material were calculated by weighing the mass change of the sample before and after the carbonation reaction of the calcium-based material to measure the heat storage performance of the material. The effective conversion rate represents the ratio of the mass of CaO that actually reacts during each carbonation process to the total mass of the sample before the carbonation reaction, as shown in Equation (1).
[0039]
[0040] In the formula, N is the number of heat storage cycles; X N is the effective conversion rate of the calcium-based material in the Nth heat storage cycle; m car,N and m cal,N are the masses of the calcium-based material after the Nth carbonation and the Nth calcination, respectively, in g; m 0 is the total mass of the sample before the carbonation reaction, in g; M CaO and M CO2 are the molar masses of CaO and CO 2 , respectively, in g / mol.
[0041] The heat storage density represents the maximum heat that can be released per unit mass of the calcium-based material during each carbonation process, as shown in Equation (2).
[0042]
[0043] In the formula, Q g,N is the mass heat storage density of the calcium-based material in the Nth heat storage cycle, in kJ / kg; △H 0 is the reaction heat of the carbonation reaction under standard conditions, calculated as 178 kJ / mol.
[0044] (3) Based on the database and the Support Vector Regression (SVR) algorithm, a combined prediction model for the best cyclic heat storage performance of calcium-based pellets is constructed. First, establish the relationship between each variable and the obtained result, that is, pellet particle size, biomass pore-forming agent type, biomass pore-forming agent content, inert component mass fraction, and the obtained result is the judgment value under the Nth heat storage cycle, set as N×X N ×Q g,N . Use support vector regression as the optimization algorithm to find the relationship between the weights of each variable and the final cyclic heat storage performance. The goal of SVR is to find a function F(x) to approximate the given samples, and finally determine the change of the output function through the mutual influence relationship coefficients between different variables.
[0045] Assume that the optimal cyclic heat storage performance is P, the mass fraction of TiO 2 is T, the type of biomass pore-forming agent is B, the corresponding biomass pore-forming agent content is C, and the pellet particle size is D. Our goal is to find the mathematical relationship between P and these variables T, B, C, D.
[0046] The basic formula of SVR is as follows:
[0047]
[0048] Among them, x = [T, B, C, D], the input variables; x i = [T i , B i , C i , D i , the variable values of the support vectors; K(x i , x), the kernel function, used to calculate the similarity between x i and x; α i and weights, which determine the contribution of each support vector to the prediction; b, the bias term, is a constant.
[0049] Specify the kernel function as the Gaussian kernel function, as follows:
[0050] K(x i , x) = exp(-γ||x i - x|| 2 )
[0051] By setting the error function to be minimized, as follows:
[0052]
[0053] Among them, w is the difference between the true value and the predicted value in the database, C is the regularization parameter (the smaller C, the simpler the model; the larger C, the lower the model error), slack variables, used to measure the error beyond the tolerance range.
[0054] Finally, the predicted results of the preparation combination under the best cyclic heat storage performance of calcium-based pellets are obtained.
[0055] (1) The calcium-based pellets for experiments are prepared by the rotary granulation method, and the preparation process is as Figure 1 shown. First, CaCO3, TiO2, and biomass pore-forming agents (bagasse, wood chips, rice husks, peanut shells) are first placed in an oven and dried at 90 - 95 °C for 14 - 16 h; after physical mixing, the premixed powder is obtained. The powder is placed in a laboratory granulation drum and rotated at a specified speed (30 - 35 revolutions per minute). A 0.1 - 0.015 mg / ml polyvinylpyrrolidone (PVP) solution is sprayed at intervals of 10 - 15 min to form seed particles; after the formation process of the seed particles is stable, an appropriate amount of dry powder is added to adjust the particle size; when the pellet particle size reaches the required size, a tumbling process at 90 - 95 revolutions per minute for 45 - 60 min is carried out to further improve its mechanical strength and form pellets. During the preparation of calcium-based pellets, the main types of biomass pore-forming agents are 1) rice husks, 2) wood chips, 3) bagasse, 4) peanut shells (all obtained by crushing and passing through a 200-mesh sieve to obtain 0.075-mm particles), and the mixing ratios of different biomass pore-forming agents are 2%, 5%, 10%, 15%, and 20%; the main inert carrier is TiO 2 , and the mass fractions are 0%, 2%, 5%, 10%, and 20% respectively; during the granulation process, it is pre-calcined in a tubular furnace at 850 °C with an air flow rate of 1.5 L / min for 15 min; finally, three types of calcium-based pellet particle sizes are formed, 1) 0.6 - 1 mm, 2) 1 - 1.5 mm, 3) 1.5 - 2 mm.
[0056] (2) The cyclic heat storage experiment is carried out on a fixed-bed experimental bench, as Figure 2As shown in the figure, the sample was placed in the constant temperature section (850 °C) for carbonation reaction. After the reaction tube was sealed, CO2 was introduced at a gas flow rate of 1.5 L / min, and after pressurization, the calcium-based pellet sample was fed in. During the process of taking out the sample, CO2 was continuously introduced to avoid high-temperature calcination of the sample during the taking-out process, and then it was placed in a drying dish to cool. After the sample was weighed, the reaction gas was switched, and calcination (850 °C) was carried out under a pure N2 atmosphere with a gas flow rate of 1 L / min. After the calcination was completed, the sample was quickly taken out, weighed, and then the sample was put in again to repeat the above steps, which was the pressurized carbonation cycle heat storage experiment of calcium-based pellets. For the multi-cycle pressurized carbonation heat storage performance experiment of calcium-based pellets, the pressure range was from atmospheric pressure to 0.2 - 0.8 MPa, and the pressure was increased in steps of 0.2 MPa, with a total of 5 groups of pressures; since the reaction was carried out at high temperature and it was easy to react with CO2 in the air during the weighing process, therefore, 3 sets of repeated experiments were carried out for the same pellet sample, and the average value was taken to reduce the experimental error; the size of the horizontal fixed-bed platform described in this experiment was that the diameter of the central tube was 74 mm and the heating length was 1000 mm, among which the central constant temperature section was 25 mm, as Figure 3 shown.
[0057] The effective conversion rate and heat storage density of the calcium-based material were calculated by weighing the mass change of the sample before and after the carbonation reaction of the calcium-based material to measure the heat storage performance of the material. The effective conversion rate represents the ratio of the mass of CaO actually reacting during each carbonation process to the total mass of the sample before the carbonation reaction, as shown in Equation (1).
[0058]
[0059] In the formula, N is the number of heat storage cycles; X N is the effective conversion rate of the calcium-based material in the Nth heat storage cycle; m car,N and m cal,N are the masses of the calcium-based material after the Nth carbonation and the Nth calcination, respectively, g; m 0 is the total mass of the sample before the carbonation reaction, g; M CaO and M CO2 are the molar masses of CaO and CO 2 respectively, g / mol.
[0060] The heat storage density represents the maximum heat that can be released per unit mass of the calcium-based material during each carbonation process, as shown in Equation (2).
[0061]
[0062] In the formula, Q g,N is the mass heat storage density of the calcium-based material in the Nth heat storage cycle, kJ / kg; △H 0 is the reaction heat of the carbonation reaction under standard conditions, calculated as 178 kJ / mol.
[0063] (3) Construct a combined prediction model for the best cyclic heat storage performance of calcium-based pellets based on the database and the support vector regression (SVR) algorithm. First, we establish the relationship between each variable and the obtained result, that is, pellet particle size, biomass pore-forming agent type, biomass pore-forming agent content, and inert component mass fraction. The obtained result is the judgment value under the Nth heat storage cycle, which is set as N×X N ×Q g,N . Support vector regression is used as the optimization algorithm to find the relationship between the weights of each variable and the final cyclic heat storage performance. The goal of SVR is to find a function F(x) to approximate the given samples, and finally determine the change of the output function through the mutual influence relationship coefficients between different variables.
[0064] Assume that the optimal cyclic heat storage performance is P, the mass fraction of TiO2 is T, the biomass pore-forming agent type is B, the corresponding biomass pore-forming agent content is C, and the pellet particle size is D. Our goal is to find the mathematical relationship between P and these variables T, B, C, D.
[0065] The basic formula of SVR is as follows:
[0066]
[0067] where x = [T, B, C, D], the input variables; x i = [T i , B i , C i , D i , the variable values of the support vectors; K(x i , x), the kernel function, used to calculate the similarity between x i and x; α i and are the weights, which determine the contribution of each support vector to the prediction; b is the bias term, which is a constant.
[0068] It is stipulated that the kernel function is the Gaussian kernel function, as follows:
[0069] K(x i , x) = exp(-γ||x i - x|| 2 )
[0070] By setting the error function to be minimized, as follows:
[0071]
[0072] where w is the difference between the true value and the predicted value in the database, C is the regularization parameter (the smaller C is, the simpler the model; the larger C is, the lower the model error), Slack variables are used to measure the error beyond the tolerance range.
[0073] Finally, the predicted result of the preparation combination under the best cyclic heat storage performance of calcium-based pellets is obtained.
Claims
1. A combined prediction method for preparing calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm, characterized in that: Based on multi-component experiments, the following steps are included: (1) Different types of biomass pore-forming agents and different contents of inert carriers were selected to prepare experimental calcium-based pellets using a rotary granulation method; (2) Conduct multiple cycles of heat storage performance experiments using a horizontal fixed bed test bench and form a database; (3) A combined prediction model for the optimal cyclic heat storage performance of calcium-based pellets was constructed based on the database and optimization algorithm.
2. The combined prediction method for preparing calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm according to claim 1, characterized in that: Step (1) is as follows: during the preparation of calcium-based pellets, the types of biomass pore-forming agents are rice husks, wood chips, bagasse, and peanut shells; then they are all crushed and passed through a 200-mesh sieve to obtain 0.075 mm particles, wherein the blending ratios of different types of biomass pore-forming agents are 2%, 5%, 10%, 15%, and 20%; the inert carrier is TiO2, and the mass fractions are 0%, 2%, 5%, 10%, and 20%, respectively; during the granulation process, the pellets are pre-calcined for 15 minutes at 850° C. and with an air flow of 1.5 L / min in a tubular furnace; the final calcium-based pellets have three particle sizes of 0.6-1 mm, 1-1.5 mm, and 1.5-2 mm.
3. The combined prediction method for preparing calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm according to claim 2 is characterized in that: In step (1), during the experiment, CaCO3, TiO2, and biomass pore-forming agent are first placed in an oven and dried at 90-95°C for 14-16 hours; after physical mixing, a premixed powder is obtained, the powder is placed in a laboratory granulation drum and rotated at a specified speed of 30-35 rpm, and 0.1-0.015 mg / ml polyvinyl pyrrolidone PVP solution is sprayed every 10-15 minutes to form seed particles; after the seed particle formation process is stable, an appropriate amount of dry powder is added to adjust the particle size; when the particle size of the spherical particles reaches the desired size, tumbling is performed for 45-60 minutes at 90-95 rpm.
4. The combined prediction method for preparing calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm according to claim 1, characterized in that: Step (2) is as follows: the cyclic heat storage experiment is carried out in a fixed bed test bench, the sample is placed in a constant temperature section at 850°C for carbonation reaction, after the reaction tube is sealed, CO2 is introduced at a gas volume of 1.5L / min, and the calcium-based pellet sample is sent in after the pressure is increased; CO2 is continuously introduced during the process of taking out the sample, and the sample is placed in a drying dish for cooling; after the sample is weighed, the reaction gas is switched, and calcination is carried out at 850°C in a pure N2 atmosphere with a gas flow rate of 1L / min. After the calcination is completed, the sample is quickly taken out, and after weighing, the sample is put in again to repeat the above steps, which is the calcium-based pellet pressurized carbonation cyclic heat storage experiment.
5. The combined prediction method for preparing calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm according to claim 4 is characterized in that: Multiple cycles of calcium-based pellet pressurized carbonation heat storage performance experiments were conducted, with a pressure range of normal pressure, 0.2-0.8MPa and a step size of 0.2MPa for increasing the pressure, for a total of 5 groups of pressures; it is easy to react with CO2 in the air during the weighing process, and 3 groups of repeated experiments were conducted on the same pellet sample to take the average value; the central tube diameter of the experimental horizontal fixed bed reactor is 74mm, the heating length is 1000mm, of which the central constant temperature section is 25mm.
6. The combined prediction method for preparing calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm according to claim 1, characterized in that: Step (3) is as follows: Based on the database and the collaborative filtering algorithm based on the regression model, a prediction model for the optimal cyclic heat storage performance preparation combination of calcium-based pellets is constructed, and the particle size of the pelletizing particles, the type and content of the biomass pore-forming agent, and the mass fraction of the inert component are used as four groups of data input, and the cyclic heat storage performance under different pressures is used as data output. The heat storage performance is measured by the effective conversion rate of the heat output cycle of the calcium-based material and the heat storage density, which are multiplied by their respective number of cycles and added together. The highest total value is the optimal calcium-based pellet preparation combination.
7. The combined prediction method for preparing calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm according to claim 6 is characterized in that: Assume that the optimal cycle heat storage performance is P, the mass fraction of TiO2 is T, the type of biomass pore-forming agent is B, the corresponding biomass pore-forming agent content is C, and the particle size of the spherule is D, and find the mathematical relationship between P and the variables T, B, C, and D; among them, the formula of SVR is as follows: Among them, x=[T,B,C,D], input variable; x i =[T i ,B i ,C i ,D i ], the variable value of the support vector; K(x i ,x), kernel function, used to calculate x i Similarity with x; α i and The weight determines the contribution of each support vector to the prediction; b, the bias term, is a constant; The kernel function is specified as a Gaussian kernel function, as follows: K(x i ,x)=exp(-γ||x i -x|| 2 ) By setting the error function to be minimized, as follows: Where w is the difference between the true value and the predicted value of the database, C is the regularization parameter (the smaller C is, the simpler the model is, and the larger C is, the lower the model error is), ξ i , Slack variables are used to measure whether the error exceeds the tolerance range; Finally, the preparation combination prediction results of calcium-based pellets with optimal cyclic heat storage performance were obtained.
8. A combined prediction system for the preparation of calcium-based pellets with optimal cycle heat storage performance based on an optimization algorithm, characterized in that: It includes: preparation module: used to select different types of biomass pore-forming agents and different contents of inert carriers, and use the rotary granulation method to prepare experimental calcium-based pellets; Experimental module: used to conduct multiple cycles of heat storage performance experiments using a horizontal fixed bed test bench and form a database; Algorithm module: used to construct a combined prediction model for the preparation of optimal cyclic heat storage performance of calcium-based pellets based on database and optimization algorithm.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into the processor, it implements the combined prediction method for preparing calcium-based pellets with optimal cyclic heat storage performance based on an optimization algorithm according to any one of claims 1 to 7.
10. A storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements the combined prediction method for preparing calcium-based pellets with optimal cyclic heat storage performance based on an optimization algorithm according to any one of claims 1 to 7.