Carbon capture performance comprehensive detection method and system based on integrated system

Abstract

The invention relates to the technical field of comprehensive detection, and discloses a carbon capture performance comprehensive detection method and system based on an integrated system. The method comprises the following steps: firstly, establishing a residual and waste heat compensation system model, and taking residual and waste heat compensation as a first carbon capture performance parameter; a carbon capture prediction value is obtained through prediction based on an SVM prediction model, a carbon absorption amount prediction value is obtained according to a grey prediction model, the ratio of the carbon capture prediction value to the carbon absorption amount prediction value is marked as the carbon capture rate, and the carbon capture rate serves as a second carbon capture performance parameter; training a compensation fuzzy neural network according to the first carbon capture performance parameter and the second carbon capture performance parameter, optimizing a compensation fuzzy neural network model parameter by using a piebald optimization algorithm, and outputting a final carbon capture performance parameter; and finally, synthesizing the final carbon capture performance parameter and the residual and waste heat utilization efficiency to detect the carbon capture cycle performance.

Description

Technical Field

[0001] The present invention relates to the technical field of comprehensive detection, and specifically to a comprehensive detection method and system for carbon capture performance based on an integrated system. Background Art

[0002] The carbon capture technology converts thermal power units into carbon capture units, which is an effective means to achieve low-carbon power at present. However, the traditional carbon capture technology introduces wind power generation, increasing the uncertainty and load of carbon capture, which is not conducive to economy and environmental protection. Moreover, the traditional carbon capture system has low industrial waste heat recovery efficiency and high maintenance cost, with relatively low comprehensive performance and unable to achieve the efficient utilization of waste heat compensation in the carbon capture system. Summary of the Invention

[0003] In view of the problems in the related art, the present invention provides a comprehensive detection method and system for carbon capture performance based on an integrated system to overcome the above-mentioned technical problems existing in the prior related art.

[0004] To solve the above technical problems, the present invention is realized through the following technical solutions:

[0005] The present invention provides a comprehensive detection method for carbon capture performance based on an integrated system, including the following steps:

[0006] S1. Calculate the parameters of the waste heat compensation system, use the thermal economic evaluation index and Construct a waste heat compensation system model, use the waste heat energy level as the evaluation index of the waste heat compensation system model, and regard the waste heat compensation as the first carbon capture performance parameter;

[0007] S2. Use a temperature and pressure measuring device to obtain the carbon capture temperature and pressure, train an SVM prediction model, predict carbon capture based on the SVM prediction model to obtain a carbon capture prediction value, and adjust the current waste heat compensation according to the carbon capture prediction value; then obtain a carbon absorption amount prediction value according to the grey prediction model, record the ratio of the carbon capture prediction value to the carbon absorption amount prediction value as the carbon capture rate, and regard the carbon capture rate as the second carbon capture performance parameter;

[0008] S3. Construct a carbon capture performance parameter matrix from the first carbon capture performance parameter and the second carbon capture performance parameter, train a compensation fuzzy neural network, perform performance optimization control on the carbon capture cycle system based on the compensation fuzzy neural network model, use the spotted kingfisher optimization algorithm to optimize the parameters of the compensation fuzzy neural network model, and output the final carbon capture performance parameter; comprehensively detect the performance of the carbon capture cycle system based on the final carbon capture performance parameter and the waste heat utilization efficiency.

[0009] The invention first calculates the thermal economic evaluation index in the waste heat compensation system model and uses Measure the residual waste heat energy level of the boiler and turbine in a coal-fired power plant, obtain the residual waste heat compensation efficiency, and take the residual waste heat compensation as the first carbon capture performance parameter; secondly, train an SVM prediction model using the carbon capture temperature and pressure, and predict the carbon capture based on the SVM prediction model to obtain the carbon capture prediction value; then construct a time series according to the grey theory, and obtain the carbon absorption prediction value by calculating the area and integral within a time period on the time series, calculate the carbon capture rate, and take the carbon capture rate as the second carbon capture performance parameter. Train an SVM prediction model using pressure and temperature as the main parameters. The SVM has a high prediction accuracy. Use the grey model to process the time series, weaken the uncertainty, and improve the prediction accuracy; then construct a carbon capture performance parameter matrix from the first carbon capture performance parameter and the second carbon capture performance parameter, train a compensation fuzzy neural network, and perform performance optimization control on the carbon capture cycle system based on the compensation fuzzy neural network model. This neural network is suitable for system control under conditions where accurate measurement is difficult, has a fast training speed, and has a good error tolerance rate; and use the spotted kingfisher optimization algorithm to optimize the parameters of the compensation fuzzy neural network model. This algorithm can effectively avoid local optima and has strong exploration performance. Finally, comprehensively detect the performance of the carbon capture cycle system.

[0010] Preferably, the S1 includes the following steps:

[0011] S11. The parameters of the residual waste heat compensation system include the thermal economy evaluation index and The thermal economy evaluation index includes the coal consumption rate of the boiler and turbine in the coal-fired power plant, the cycle thermal efficiency of the boiler and turbine in the coal-fired power plant, and the power generation efficiency of the boiler and turbine in the coal-fired power plant; set the heat consumption of the boiler and turbine in the coal-fired power plant as a1, the calorific value of standard coal of the boiler and turbine in the coal-fired power plant as a2, the efficiency of the boiler and turbine in the coal-fired power plant as b1, and the pipeline efficiency of the boiler and turbine in the coal-fired power plant as b2. Then the calculation formula for the coal consumption rate A′ of the boiler and turbine in the coal-fired power plant is as follows,

[0012]

[0013] Set the main steam enthalpy value of the boiler and turbine in the coal-fired power plant as a3, the feed water enthalpy value of the main steam of the boiler and turbine in the coal-fired power plant as a4, the primary reheated steam fraction of the boiler and turbine in the coal-fired power plant as c1, the secondary reheated steam fraction of the boiler and turbine in the coal-fired power plant as c2, the primary reheated specific enthalpy rise of the boiler and turbine in the coal-fired power plant as b3, the secondary reheated specific enthalpy rise of the boiler and turbine in the coal-fired power plant as b4, the exhaust fraction of the boiler and turbine in the coal-fired power plant as a5, the enthalpy value of the turbine in the boiler and turbine in the coal-fired power plant as a6, the extraction steam fraction of the heater in the boiler and turbine in the coal-fired power plant as a′, the extraction steam enthalpy value of the heater in the boiler and turbine in the coal-fired power plant as a″, and the number of extraction stages as a. Then the calculation formulas for the cycle thermal efficiency and power generation efficiency of the boiler and turbine in the coal-fired power plant are as follows,

[0014]

[0015] Among them, A″ represents the circulating thermal efficiency of the boiler-turbine unit in a coal-fired power plant, and A″′ represents the power generation efficiency of the boiler-turbine unit in a coal-fired power plant. represents the extraction steam fraction of the heater in the boiler-turbine unit of a coal-fired power plant with the extraction steam stage number of i1. represents the extraction steam enthalpy value of the heater in the boiler-turbine unit of a coal-fired power plant with the extraction steam stage number of i1, where i1 = 1, 2, 3,..., a.

[0016] S12. The waste heat quantity b of the boiler-turbine unit in a coal-fired power plant is obtained from the coal consumption rate, the circulating thermal efficiency, and the power generation efficiency of the boiler-turbine unit in a coal-fired power plant. When the waste heat of the boiler-turbine unit in a coal-fired power plant enters the capture system, assuming the temperature before entering the capture system is b′, the temperature after coming out of the capture system is b″, and the absorption efficiency is α, then the actually available waste heat quantity The calculation formula is as follows.

[0017]

[0018] The actually available waste heat quantity enters the collection device. Assuming the enthalpy before the actually available waste heat quantity enters the collection device is B′, the enthalpy after the actually available waste heat quantity enters the collection device is B″, the entropy before the actually available waste heat quantity enters the collection device is C′, the entropy after the actually available waste heat quantity enters the collection device is C″, and the temperature change is recorded as ΔB, then The calculation formula for C is as follows.

[0019] C = (B″ - B′) - ΔB(C″ - C′);

[0020] Using the to measure the waste heat energy level of the boiler-turbine unit in a coal-fired power plant, then the calculation formula for the waste heat energy level β of the boiler-turbine unit in a coal-fired power plant is as follows.

[0021]

[0022] Assuming the energy consumption threshold is ω, when the waste heat energy level of the boiler-turbine unit in a coal-fired power plant is greater than ω, the waste heat compensation efficiency is high; otherwise, the waste heat compensation efficiency is low. The waste heat compensation is used as the first carbon capture performance parameter.

[0023] This invention calculates the waste heat compensation system parameters, uses the thermal economic evaluation index and constructs a waste heat compensation system model, calculates the enthalpy of the waste heat and obtains the waste heat energy level, uses the waste heat energy level as the evaluation index of the waste heat compensation system model, and uses the waste heat compensation as the first carbon capture performance parameter.

[0024] Preferably, the S2 includes the following steps:

[0025] S21. Obtain the carbon capture temperature and carbon capture pressure at \(i\) moments during the carbon capture process to get a carbon capture temperature and pressure set, denoted as \(D=\{(d_1,e_1),(d_2,e_2),(d_3,e_3),\cdots,(d i ,e i )\}, where \(d i \) and \(e i \) respectively represent the carbon capture temperature and carbon capture pressure at the \(i\)-th moment; obtain the carbon capture temperature and carbon capture pressure data during the carbon capture process in previous years to get a carbon capture temperature and pressure sample set, and divide the carbon capture temperature and pressure sample set into a carbon capture temperature and pressure training set and a carbon capture temperature and pressure test set; set the radial basis function as the kernel function of the SVM (Support Vector Machine), the kernel function width is \(\chi\), and the penalty parameter is \(\delta\). Input the carbon capture temperature and pressure training set into the SVM for training, and set the maximum number of iterations as The current number of iterations is When the current number of iterations is equal to the maximum number of iterations, stop the iteration to obtain a trained SVM;

[0026] Input the carbon capture temperature and pressure test set into the trained SVM, set the error threshold as \(\xi\). When the error between the output result of the trained SVM and the actual value is less than \(\xi\), obtain the SVM prediction model; otherwise, adjust the kernel function width and penalty parameter until the error between the output result of the trained SVM and the actual value is less than \(\xi\);

[0027] S22. Input the carbon capture temperature and pressure set into the SVM prediction model to output a carbon capture prediction value. Set the first threshold as \(\omega_1\). When the absolute value of the difference between the carbon capture prediction value and the current carbon capture value is greater than \(\omega_1\), adjust the current surplus waste heat compensation so that the absolute value of the difference between the carbon capture prediction value and the current carbon capture value is less than \(\omega_1\); otherwise, do not adjust the current surplus waste heat compensation; obtain the current carbon absorption amount and calculate the carbon absorption amount prediction value. The specific process is as follows:

[0028] S221. Denote the carbon absorption amount from the carbon absorption amount in previous years to the current carbon absorption amount as a carbon absorption amount set, denoted as \(D'=\{d_1',d_2',d_3',\cdots,d j '\}, where \(d j '\) represents the \(j\)-th carbon absorption amount. Generate a time series for the time corresponding to the carbon absorption amount in the carbon absorption amount set where represents that the time series is of the time; use the grey theory to construct a whitenization differential equation for the time series. The whitenization differential equation includes a development coefficient, a grey action quantity, and a background value; plot the carbon absorption amounts in the carbon absorption amount set on a rectangular coordinate system according to the time series to obtain a carbon absorption amount image. Connect the times in the time series in the carbon absorption amount image and time respectively corresponding to the carbon absorption amounts in the carbon absorption amount sets to obtain a time series curve, and integrating the time series curve at time and time to obtain an integral value, i.e., the grey action; calculating the area of the time series curve on the rectangular coordinate system at time and time to obtain an area value, i.e., the background value;

[0029] S222. Substitute the background value and the grey action into the whiting differential equation to obtain the development coefficient. Then, the calculation formula for the predicted value of the carbon absorption amount at time is as follows

[0030]

[0031] where represents the predicted value of the carbon absorption amount at time , g1 represents the development coefficient, g2 represents the grey action, and e represents a constant;

[0032] S23. Repeat S221 and S222 to obtain the predicted value of the carbon absorption amount, calculate the ratio of the predicted value of the carbon capture and the predicted value of the carbon absorption amount, denoted as the carbon capture rate, and use the carbon capture rate as the second carbon capture performance parameter.

[0033] The present invention trains an SVM prediction model through the carbon capture temperature and carbon capture pressure sets, predicts the carbon capture based on the SVM prediction model to obtain the predicted value of the carbon capture; then constructs a time series according to the grey theory, calculates the predicted value of the carbon absorption amount, records the ratio of the predicted value of the carbon capture and the predicted value of the carbon absorption amount as the carbon capture rate, and uses the carbon capture rate as the second carbon capture performance parameter; the SVM prediction has high accuracy, and using the grey model can weaken the uncertainty and improve the prediction accuracy.

[0034] Preferably, the S3 includes the following steps

[0035] S31. Take the first i' carbon capture performance parameters of the first carbon capture performance parameter and the second carbon capture performance parameter to form a carbon capture performance parameter set, denoted as E = {(h1, k1), (h2, k2), (h3, k3),..., (h i′ , k i′ )}, where h i′ represents the i'-th first carbon capture performance parameter, and k i′ represents the i'-th second carbon capture performance parameter; set the feature dimension of the carbon capture performance parameter set as j', and generate a carbon capture performance parameter matrix F as follows

[0036]

[0037] Among them, h i′j′ represents the first carbon capture performance parameter of the i'-th with a characteristic dimension of j', and h i′j′ represents the second carbon capture performance parameter of the i'-th with a characteristic dimension of j'.

[0038] Use the relational degree clustering method to cluster the carbon capture performance parameter matrix. Set i'' reference values to form a reference vector, and take i'' carbon capture performance parameters from the carbon capture performance parameter set as comparison vectors. Calculate the relational degree between the reference vector and the comparison vector. The calculation formula is as follows.

[0039]

[0040] Among them, G represents the relational degree between the reference vector and the comparison vector, ε represents the width of the membership function, ||E'-E''|| represents the Euclidean distance between the reference vector and the comparison vector, and exp represents the exponential function with the constant e as the base.

[0041] Set the relational degree threshold as ψ. When the relational degree between the reference vector and the comparison vector is less than ψ, set the relational degree between the reference vector and the comparison vector to 0; otherwise, retain the relational degree between the reference vector and the comparison vector.

[0042] S32. Then calculate the clustering degree between the reference vector and the comparison vector. When the clustering degree between the reference vector and the comparison vector is the same as the relational degree between the reference vector and the comparison vector, take the value of i'' as the number of clusters; otherwise, change the value of i'' and repeat S31 until the clustering degree between the reference vector and the comparison vector is the same as the relational degree between the reference vector and the comparison vector.

[0043] Select carbon capture performance parameters from the comparison vectors as the initial cluster centers to obtain an initial cluster center set. Calculate the distances from the carbon capture performance parameters in the comparison vectors to the initial cluster centers in the initial cluster center set, denoted as the initial cluster distance set. Find the minimum initial cluster distance corresponding to the carbon capture performance parameters in the comparison vectors in the initial cluster distance set, denoted as the initial sample data. Add the initial sample data to the initial cluster center set to obtain a new cluster center set; then find the new cluster centers from the new cluster center set and repeat S32 to obtain the final cluster center set. Set the squared error threshold as ζ. When the squared error of the final cluster centers in the final cluster center set is less than ζ, do not repeat S32; otherwise, repeat S32 until the squared error of the final cluster centers in the final cluster center set is less than ζ; denote the final cluster center set as the clustering center set of the comparison vectors.

[0044] S33. Construct an initial fuzzy model with the clustering center set of the comparison vectors. The initial fuzzy model includes fuzzy rules and membership functions; divide the carbon capture performance parameter matrix into a j'-dimensional carbon capture performance parameter training set and a j'-dimensional carbon capture performance parameter test set, and input the j'-dimensional carbon capture performance parameter training set into the compensated fuzzy neural network. The j'-dimensional carbon capture performance parameter training set passes through the fuzzification layer, and the membership function fuzzifies the j'-dimensional carbon capture performance parameter training set to obtain a fuzzified set. The fuzzified set passes through the fuzzy inference layer to match the fuzzy rules and convert the fuzzified set into a fuzzified subset. The fuzzified subset passes through the compensation operation layer and performs compensation function operations to output results in the defuzzification layer;

[0045] Set the membership function of the compensated fuzzy neural network as the Gaussian function, the center of the membership function is φ, and the width of the membership function is The compensation degree of the compensation function of the compensated fuzzy neural network is γ, the global error threshold is σ. The j'-dimensional carbon capture performance parameter training set is input into the compensated fuzzy neural network. When the error of the output result of the compensated fuzzy neural network is less than σ, the trained compensated fuzzy neural network is obtained; otherwise, adjust the center, width of the membership function, and the compensation degree of the compensation function until the error of the output result of the compensated fuzzy neural network is less than σ. Use the spotted kingfisher optimization algorithm to optimize and adjust the center, width of the membership function, and the compensation degree of the compensation function. The specific steps are as follows:

[0046] S331. Set the fitness function as the error between the output result of the compensated fuzzy neural network and the actual value. The fitness function represents the ability of the spotted kingfisher individual to solve problems; the initial position of the p-th spotted kingfisher in the spotted kingfisher population is The upper bound of the search range of the p-th spotted kingfisher in the spotted kingfisher population is l', the lower bound of the search range of the p-th spotted kingfisher in the spotted kingfisher population is l'', k1 represents a random number and k1 ∈ [0, 1]. The calculation formula for the initial position of the p-th spotted kingfisher in the spotted kingfisher population is as follows,

[0047]

[0048] When the spotted kingfisher is in the perching and hovering stage, the fitness function value is calculated at this time. Set the current iteration number as q, and the position of the spotted kingfisher at the q-th iteration is The position of the spotted kingfisher at the (q + 1)-th iteration is The position of the p-th spotted kingfisher in the spotted kingfisher population at the q-th iteration is k2 represents a random number and k2 ∈ [0, 1], η represents the state parameter. Then the calculation formula for the position of the spotted kingfisher at the (q + 1)-th iteration is as follows,

[0049]

[0050] Set the maximum number of iterations to Q, the jumping factor to κ, k3 represents a random number and k3 ∈ [0, 1], λ represents the roosting coefficient. The calculation formula for the state parameter η of the pied kingfisher during roosting is as follows.

[0051]

[0052] The calculation formula for the state parameter η of the pied kingfisher during hovering is as follows.

[0053]

[0054] Among them, μ represents the hovering coefficient, f p represents the fitness function value of the p-th pied kingfisher in the pied kingfisher population. represents the -th pied kingfisher in the pied kingfisher population;

[0055] S332. Continuously iterate the fitness function value. During the diving stage of the pied kingfisher, the pied kingfisher continuously updates its position to find the best fitness function value. Set the hunting abilities of the pied kingfisher to h1 and h2, the control parameter to ν, the current best position of the pied kingfisher to x′, the hunting position of the pied kingfisher to x″, k4 represents a random number and k4 ∈ [0, 1]. Then the updated position of the pied kingfisher at the (q + 1)-th iteration is calculated as follows.

[0056]

[0057] After the pied kingfisher updates its position, during the random generation stage, randomly select two individual pied kingfishers in the pied kingfisher population, denoted as and Set the position at the q-th iteration to be The position at the q-th iteration is The predation efficiency of the pied kingfisher is k5 represents a random number and k5 ∈ [0, 1]. Continue to find the best fitness function value and replace the current fitness function value with the best fitness function value. Then the updated position of the pied kingfisher at the (q + 1)-th iteration is calculated as follows.

[0058]

[0059] When the current number of iterations is equal to the maximum number of iterations, stop the iteration to obtain the final position of the pied kingfisher. The coordinate values of the three-dimensional coordinates corresponding to the final position of the pied kingfisher are respectively the center of the membership function, the width, and the compensation degree of the compensation function.

[0060] S34. Input the j'-dimensional carbon capture performance parameter test set into the trained compensatory fuzzy neural network to output the final carbon capture performance parameter. Set the work done by the turbine in the carbon capture cycle system as m, the power consumption of the compressor as n, and the carbon capture input quantity as o, and obtain the waste heat utilization efficiency from the second law of thermodynamics The calculation formula is as follows

[0061]

[0062] Set the second threshold as ω2 and the third threshold as ω3. When the final performance parameter is greater than ω2 and the waste heat utilization efficiency is greater than ω3, the performance detection result of the carbon capture cycle system is qualified, and there is no need to adjust the carbon capture cycle system; otherwise, adjust the carbon capture cycle system and increase the waste heat compensation until the final performance parameter is greater than ω2 and the waste heat utilization efficiency is greater than ω3 to achieve the performance detection of the carbon capture cycle system.

[0063] The present invention combines the first carbon capture performance parameter and the second carbon capture performance parameter to train a compensatory fuzzy neural network, performs performance optimization control on the carbon capture cycle system based on the compensatory fuzzy neural network model; and uses the pied kingfisher optimization algorithm to optimize the parameters of the compensatory fuzzy neural network model, and finally detects the carbon capture cycle system. This neural network is applicable to system control under difficult-to-accurately-measure conditions, has a fast training speed, has a good fault tolerance rate and can effectively avoid local optima, and this algorithm can avoid local optima and has strong development performance.

[0064] The present invention also discloses a system for a comprehensive carbon capture performance detection method based on an integrated system, specifically including: a waste heat compensation module, a carbon capture rate prediction module, a compensatory fuzzy neural network module, and a carbon capture performance detection module;

[0065] The waste heat compensation module is used to use the thermal economy evaluation index and construct a waste heat compensation system model to calculate the waste heat energy level;

[0066] The carbon capture rate prediction module predicts carbon capture based on the SVM prediction model to obtain the carbon capture prediction value, and then obtains the carbon capture rate according to the grey model;

[0067] The compensatory fuzzy neural network module is used to train the compensatory fuzzy neural network and use the pied kingfisher optimization algorithm to optimize the parameters of the compensatory fuzzy neural network model;

[0068] The carbon capture performance detection module is used to perform performance detection on the carbon capture cycle system by combining the final carbon capture performance parameter and the waste heat utilization efficiency.

[0069] The present invention has the following beneficial effects:

[0070] 1. The invention first calculates the thermal economic evaluation index in the waste heat compensation system model and uses to measure the waste heat energy level of the boiler and turbine in a coal-fired power plant, obtains the waste heat compensation efficiency, and takes the waste heat compensation as the first carbon capture performance parameter.

[0071] 2. The invention uses the carbon capture temperature and pressure to train an SVM prediction model, predicts carbon capture based on the SVM prediction model to obtain a carbon capture prediction value; then constructs a time series according to the grey theory to obtain a carbon absorption prediction value, calculates the carbon capture rate, and takes the carbon capture rate as the second carbon capture performance parameter. The SVM used has a high prediction accuracy, and using the grey model is beneficial to improving the prediction accuracy.

[0072] 3. The invention trains a compensation fuzzy neural network with the first carbon capture performance parameter and the second carbon capture performance parameter, performs performance optimization control on the carbon capture cycle system based on the compensation fuzzy neural network model, and uses the spotted kingfisher optimization algorithm to optimize the parameters of the compensation fuzzy neural network model, and finally performs performance detection on the carbon capture cycle system. This neural network has a fast training speed and a good fault tolerance rate. This algorithm can effectively avoid local optima and has strong development performance.

[0073] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0075] Figure 1 FIG. is a schematic flow chart of the comprehensive detection of carbon capture performance by a carbon capture performance comprehensive detection system based on an integrated system provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0076] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0077] In the description of the present invention, it should be understood that the terms "opening", "upper", "lower", "top", "middle", "inner", etc. indicating the orientation or position relationship are only for facilitating the description of the invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the invention.

[0078] This embodiment discloses a comprehensive detection method for carbon capture performance based on an integrated system, which specifically includes the following contents:

[0079] S1. Calculate the parameters of the surplus waste heat compensation system, use the thermal economy evaluation index and Construct a surplus waste heat compensation system model, use the surplus waste heat energy level as the evaluation index of the surplus waste heat compensation system model, and regard the surplus waste heat compensation as the first carbon capture performance parameter.

[0080] The said S1 includes the following steps:

[0081] S11. The parameters of the surplus waste heat compensation system include the thermal economy evaluation index and The said thermal economy evaluation index includes the coal consumption rate of the boiler and turbine in a coal-fired power station, the cycle thermal efficiency of the boiler and turbine in a coal-fired power station, and the power generation efficiency of the boiler and turbine in a coal-fired power station; set the heat consumption of the boiler and turbine in a coal-fired power station as a1, the calorific value of standard coal of the boiler and turbine in a coal-fired power station as a2, the efficiency of the boiler and turbine in a coal-fired power station as b1, and the pipeline efficiency of the boiler and turbine in a coal-fired power station as b2. Then the calculation formula of the coal consumption rate A′ of the boiler and turbine in a coal-fired power station is as follows,

[0082]

[0083] Set the main steam enthalpy value of the boiler and turbine in a coal-fired power station as a3, the feed water enthalpy value of the main steam of the boiler and turbine in a coal-fired power station as a4, the primary reheat steam fraction of the boiler and turbine in a coal-fired power station as c1, the secondary reheat steam fraction of the boiler and turbine in a coal-fired power station as c2, the primary reheat specific enthalpy rise of the boiler and turbine in a coal-fired power station as b3, the secondary reheat specific enthalpy rise of the boiler and turbine in a coal-fired power station as b4, the exhaust steam fraction of the boiler and turbine in a coal-fired power station as a5, the enthalpy value of the intermediate turbine in a coal-fired power station as a6, the extraction steam fraction of the heater in a coal-fired power station as a′, the extraction steam enthalpy value of the heater in a coal-fired power station as a″, and the number of extraction stages as a. Then the calculation formulas of the cycle thermal efficiency and the power generation efficiency of the boiler and turbine in a coal-fired power station are as follows,

[0084]

[0085] Among them, A″ represents the cycle thermal efficiency of the boiler and turbine in a coal-fired power station, and A″′ represents the power generation efficiency of the boiler and turbine in a coal-fired power station. represents the extraction steam fraction of the heater in a coal-fired power station with the number of extraction stages as i1. represents the extraction steam enthalpy value of the heater in a coal-fired power station with the number of extraction stages as i1, where i1 = 1, 2, 3,..., a;

[0086] S12. Obtain the surplus waste heat b of the coal-fired power plant boiler by the coal consumption rate, the cycle thermal efficiency, and the power generation efficiency of the coal-fired power plant boiler. When the surplus waste heat of the coal-fired power plant boiler enters the capture system, set the temperature before entering the capture system as b′, the temperature after coming out of the capture system as b″, and the absorption efficiency as α. Then the actually available surplus waste heat The calculation formula is as follows

[0087]

[0088] The actually available surplus waste heat enters the collection device. Set the enthalpy before the actually available surplus waste heat enters the collection device as B′, the enthalpy after the actually available surplus waste heat enters the collection device as B″, the entropy before the actually available surplus waste heat enters the collection device as C′, the entropy after the actually available surplus waste heat enters the collection device as C″, and the temperature change is recorded as ΔB. Then The calculation formula of C is as follows

[0089] C = (B″ - B′) - ΔB(C″ - C′);

[0090] Use the above to measure the surplus waste heat energy level of the coal-fired power plant boiler. Then the calculation formula of the surplus waste heat energy level β of the coal-fired power plant boiler is as follows

[0091]

[0092] Set the energy consumption threshold as ω. When the surplus waste heat energy level of the coal-fired power plant boiler is greater than ω, the surplus waste heat compensation efficiency is high; otherwise, the surplus waste heat compensation efficiency is low. Take the surplus waste heat compensation as the first carbon capture performance parameter;

[0093] S2. Use the carbon capture temperature and pressure to train an SVM prediction model, predict carbon capture based on the SVM prediction model to obtain a carbon capture prediction value, and adjust the current surplus waste heat compensation according to the carbon capture prediction value; then obtain a carbon absorption amount prediction value according to the grey prediction model, and record the ratio of the carbon capture prediction value to the carbon absorption amount prediction value as the carbon capture rate. Take the carbon capture rate as the second carbon capture performance parameter;

[0094] S2 includes the following steps

[0095] S21. Obtain the carbon capture temperature and carbon capture pressure at i moments during the carbon capture process to obtain a carbon capture temperature and pressure set, denoted as D = {(d1, e1), (d2, e2), (d3, e3),..., (d i , e i )}, where d i and e irespectively represent the carbon capture temperature and carbon capture pressure at the \(i\)-th moment; obtain the carbon capture temperature and carbon capture pressure data during the carbon capture process in previous years to get a carbon capture temperature-pressure sample set, and divide the carbon capture temperature-pressure sample set into a carbon capture temperature-pressure training set and a carbon capture temperature-pressure test set; set the radial basis function as the kernel function of the SVM, the width of the kernel function is \(\chi\), and the penalty parameter is \(\delta\), input the carbon capture temperature-pressure training set into the SVM for training, and set the maximum number of iterations as The current number of iterations is When the current number of iterations is equal to the maximum number of iterations, stop the iteration to obtain a trained SVM;

[0096] Input the carbon capture temperature-pressure test set into the trained SVM, set the error threshold as \(\xi\), when the error between the output result of the trained SVM and the actual value is less than \(\xi\), obtain the SVM prediction model; otherwise, adjust the width of the kernel function and the penalty parameter until the error between the output result of the trained SVM and the actual value is less than \(\xi\);

[0097] S22. Input the carbon capture temperature-pressure set into the SVM prediction model, output the carbon capture prediction value, set the first threshold as \(\omega_1\), when the absolute value of the difference between the carbon capture prediction value and the current carbon capture value is greater than \(\omega_1\), then regulate the current surplus waste heat compensation to make the absolute value of the difference between the carbon capture prediction value and the current carbon capture value less than \(\omega_1\), otherwise do not regulate the current surplus waste heat compensation; obtain the current carbon absorption amount, and calculate the carbon absorption amount prediction value. The specific process is as follows:

[0098] S221. Denote the carbon absorption amounts from previous years to the current carbon absorption amount as a carbon absorption amount set, denoted as \(D'=\{d_1',d_2',d_3',\cdots,d j '\}\), where \(d j ' represents the \(j\)-th carbon absorption amount, and the time corresponding to the carbon absorption amount in the carbon absorption amount set generates a time series where represents that the time series is of time; use grey theory to construct a whitenization differential equation for the time series, and the whitenization differential equation includes a development coefficient, a grey action quantity, and a background value; plot the carbon absorption amounts in the carbon absorption amount set on a rectangular coordinate system according to the time series to obtain a carbon absorption amount image, and connect the carbon absorption amounts in the carbon absorption amount set corresponding to the time and the time respectively in the carbon absorption amount image to obtain a time series curve, integrate the time series curve within the time and the time to obtain an integral value, which is the grey action quantity; calculate the time series curve at the time and the time Obtain the area on the internal rectangular coordinate system, and the obtained area value is the background value;

[0099] S222. Substitute the background value and the grey action amount into the whiting differential equation to obtain the development coefficient. Then, the calculation formula for the predicted value of carbon absorption amount at time is as follows,

[0100]

[0101] where, represents the predicted value of carbon absorption amount at time , g1 represents the development coefficient, g2 represents the grey action amount, and e represents a constant;

[0102] S23. Repeat S221 and S222 to obtain the predicted value of carbon absorption amount, calculate the ratio of the predicted value of carbon capture and the predicted value of carbon absorption amount, denoted as the carbon capture rate, and use the carbon capture rate as the second carbon capture performance parameter;

[0103] S3. Construct a carbon capture performance parameter matrix based on the first carbon capture performance parameter and the second carbon capture performance parameter, and train a compensatory fuzzy neural network. Based on the compensatory fuzzy neural network model, perform performance optimization control on the carbon capture cycle system, use the spotted kingfisher optimization algorithm to optimize the parameters of the compensatory fuzzy neural network model, and output the final carbon capture performance parameter; Detect the performance of the carbon capture cycle system based on the comprehensive final carbon capture performance parameter and the waste heat utilization efficiency;

[0104] The S3 includes the following steps:

[0105] S31. Take the first i' carbon capture performance parameters of the first carbon capture performance parameter and the second carbon capture performance parameter to form a carbon capture performance parameter set, denoted as E = {(h1, k1), (h2, k2), (h3, k3),..., (h i′ , k i′ )}, where h i′ represents the i'-th first carbon capture performance parameter, and k i′ represents the i'-th second carbon capture performance parameter; Set the characteristic dimension of the carbon capture performance parameter set as j', and generate the carbon capture performance parameter matrix F as follows,

[0106]

[0107] where, h i′j′ represents the i'-th first carbon capture performance parameter with the characteristic dimension of j', and h i′j′ represents the i'-th second carbon capture performance parameter with the characteristic dimension of j';

[0108] The carbon capture performance parameter matrix is clustered using the relational degree clustering method. Set \(i''\) reference values to form a reference vector. Select \(i''\) carbon capture performance parameters from the carbon capture performance parameter set as comparison vectors. Calculate the relational degree between the reference vector and the comparison vector. The calculation formula is as follows.

[0109]

[0110] Where \(G\) represents the relational degree between the reference vector and the comparison vector, \(\varepsilon\) represents the width of the membership function, \(\vert\vert E'-E''\vert\vert\) represents the Euclidean distance between the reference vector and the comparison vector, and exp represents the exponential function with the constant \(e\) as the base.

[0111] Set the relational degree threshold as \(\psi\). When the relational degree between the reference vector and the comparison vector is less than \(\psi\), set the relational degree between the reference vector and the comparison vector to 0; otherwise, retain the relational degree between the reference vector and the comparison vector.

[0112] S32. Then calculate the clustering degree of the reference vector and the comparison vector. When the clustering degree of the reference vector and the comparison vector is the same as the relational degree between the reference vector and the comparison vector, take the value of \(i''\) as the number of clusters; otherwise, change the value of \(i''\) and repeat S31 until the clustering degree of the reference vector and the comparison vector is the same as the relational degree between the reference vector and the comparison vector.

[0113] Select carbon capture performance parameters from the comparison vectors as the initial cluster centers to obtain an initial cluster center set. Calculate the distances from the carbon capture performance parameters in the comparison vectors to the initial cluster centers in the initial cluster center set, denoted as the initial cluster distance set. Find the minimum initial cluster distance corresponding to the carbon capture performance parameters in the comparison vectors in the initial cluster distance set, denoted as the initial sample data. Add the initial sample data to the initial cluster center set to obtain a new cluster center set; then find the new cluster centers from the new cluster center set and repeat S32 to obtain the final cluster center set. Set the squared error threshold as \(\zeta\). When the squared error of the final cluster centers in the final cluster center set is less than \(\zeta\), do not repeat S32; otherwise, repeat S32 until the squared error of the final cluster centers in the final cluster center set is less than \(\zeta\); denote the final cluster center set as the clustering center set of the comparison vectors.

[0114] S33. Construct an initial fuzzy model with the set of clustering centers of the comparison vectors. The initial fuzzy model includes fuzzy rules and membership functions. Divide the carbon capture performance parameter matrix into a j'-dimensional carbon capture performance parameter training set and a j'-dimensional carbon capture performance parameter test set. Input the j'-dimensional carbon capture performance parameter training set into the compensatory fuzzy neural network. The j'-dimensional carbon capture performance parameter training set passes through the fuzzification layer. The membership function fuzzifies the j'-dimensional carbon capture performance parameter training set to obtain a fuzzified set. The fuzzified set passes through the fuzzy inference layer to match the fuzzy rules and convert the fuzzified set into a fuzzified subset. The fuzzified subset passes through the compensatory operation layer and undergoes compensatory function operations to output a result in the defuzzification layer.

[0115] Set the membership function of the compensatory fuzzy neural network as a Gaussian function. The center of the membership function is φ, and the width of the membership function is The compensation degree of the compensation function of the compensatory fuzzy neural network is γ, and the global error threshold is σ. The j'-dimensional carbon capture performance parameter training set is input into the compensatory fuzzy neural network. When the error of the output result of the compensatory fuzzy neural network is less than σ, a trained compensatory fuzzy neural network is obtained. Otherwise, adjust the center, width of the membership function, and the compensation degree of the compensation function until the error of the output result of the compensatory fuzzy neural network is less than σ. Use the pied kingfisher optimization algorithm to optimize and adjust the center, width of the membership function, and the compensation degree of the compensation function. The specific steps are as follows:

[0116] S331. Set the fitness function as the error between the output result of the compensatory fuzzy neural network and the actual value. The fitness function represents the ability of an individual pied kingfisher to solve problems. The initial position of the p-th pied kingfisher in the pied kingfisher population is The upper bound of the search range of the p-th pied kingfisher in the pied kingfisher population is l', and the lower bound of the search range of the p-th pied kingfisher in the pied kingfisher population is l''. k1 represents a random number and k1 ∈ [0, 1]. The formula for calculating the initial position of the p-th pied kingfisher in the pied kingfisher population is as follows,

[0117]

[0118] During the perching and hovering stages of the pied kingfisher, the fitness function value is calculated at this time. Set the current iteration number as q. The position of the pied kingfisher at the q-th iteration is The position of the pied kingfisher at the (q + 1)-th iteration is The position of the p-th pied kingfisher in the kingfisher population at the q-th iteration is k2 represents a random number and k2 ∈ [0, 1]. η represents the state parameter. Then the formula for calculating the position of the pied kingfisher at the (q + 1)-th iteration is as follows,

[0119]

[0120] Set the maximum number of iterations to Q, the jump factor to κ, k3 represents a random number and k3 ∈ [0, 1], λ represents the roosting coefficient. The calculation formula for the state parameter η of the pied kingfisher during roosting is as follows.

[0121]

[0122] The calculation formula for the state parameter η of the pied kingfisher during hovering is as follows.

[0123]

[0124] Among them, μ represents the hovering coefficient, f p represents the fitness function value of the p-th pied kingfisher in the pied kingfisher population. represents the fitness function value of the

[0125] S332. Continuously iterate the fitness function value. During the diving stage of the pied kingfisher, the pied kingfisher continuously updates its position to find the best fitness function value. Set the hunting abilities of the pied kingfisher to h1 and h2, the control parameter to ν, the current best position of the pied kingfisher to x′, the hunting position of the pied kingfisher to x″, k4 represents a random number and k4 ∈ [0, 1]. Then the updated position of the pied kingfisher at the (q + 1)-th iteration The calculation formula is as follows.

[0126]

[0127] After the pied kingfisher updates its position, during the random generation stage, select two random pied kingfisher individuals in the pied kingfisher population, denoted as and Set the position at the q-th iteration to be The position at the q-th iteration to be The predation efficiency of the pied kingfisher is k5 represents a random number and k5 ∈ [0, 1]. Continue to find the best fitness function value and replace the current fitness function value with the best fitness function value. Then the updated position of the pied kingfisher at the (q + 1)-th iteration The calculation formula is as follows.

[0128]

[0129] When the current number of iterations is equal to the maximum number of iterations, stop the iteration to obtain the final position of the pied kingfisher. The coordinate values of the three-dimensional coordinates corresponding to the final position of the pied kingfisher are respectively the center of the membership function, the width, and the compensation degree of the compensation function.

[0130] S34. Input the j'-dimensional carbon capture performance parameter test set into the trained compensatory fuzzy neural network to output the final carbon capture performance parameter. Set the turbine work in the carbon capture cycle system as m, the compressor work consumption as n, and the carbon capture input quantity as o, and obtain the waste heat utilization efficiency according to the second law of thermodynamics The calculation formula is as follows

[0131]

[0132] Set the second threshold as ω2 and the third threshold as ω3. When the final performance parameter is greater than ω2 and the waste heat utilization efficiency is greater than ω3, the performance detection result of the carbon capture cycle system is qualified, and there is no need to adjust the carbon capture cycle system; otherwise, adjust the carbon capture cycle system, increase the waste heat compensation until the final performance parameter is greater than ω2 and the waste heat utilization efficiency is greater than ω3, so as to realize the performance detection of the carbon capture cycle system.

[0133] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0134] The preferred embodiments of the invention disclosed above are only used to help illustrate the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principle and practical application of the invention, so that those skilled in the art in the technical field can understand and utilize the invention well.

Claims

1. A comprehensive detection method for carbon capture performance based on an integrated system, characterized in that, The method includes the following steps: S1. Calculate the parameters of the waste heat compensation system, and use the thermal economic evaluation index and Construct a waste heat compensation system model, use the waste heat energy level as the evaluation index of the waste heat compensation system model, and regard the waste heat compensation as the first carbon capture performance parameter; S2. Use a temperature and pressure measurement device to obtain the carbon capture temperature and pressure, train an SVM prediction model, predict carbon capture based on the SVM prediction model to obtain a carbon capture prediction value, and adjust the current surplus waste heat compensation according to the carbon capture prediction value; then obtain a carbon absorption amount prediction value according to the grey prediction model, record the ratio of the carbon capture prediction value to the carbon absorption amount prediction value as the carbon capture rate, and use the carbon capture rate as the second carbon capture performance parameter; S3. Construct a carbon capture performance parameter matrix according to the first carbon capture performance parameter and the second carbon capture performance parameter, and train a compensation fuzzy neural network. Perform performance optimization control on the carbon capture circulation system based on the compensation fuzzy neural network model to obtain the final carbon capture performance parameter; Detect the carbon capture performance by integrating the final carbon capture performance parameter and the surplus waste heat utilization efficiency.

2. The comprehensive detection method for carbon capture performance based on an integrated system according to claim 1, wherein Specifically, S1 includes: S11. The parameters of the waste heat compensation system include the thermal economy evaluation index and Calculate the thermal economy evaluation index and Construct a waste heat compensation system model; S12. Use the to measure the residual waste heat level of the boiler and turbine in a coal-fired power plant, and use the compensation of the residual waste heat as the first carbon capture performance parameter.

3. The comprehensive detection method for carbon capture performance based on an integrated system according to claim 2, characterized in that Specifically, S2 includes: S21. Use a temperature and pressure measurement device to obtain the carbon capture temperature and carbon capture pressure during the carbon capture process, and obtain a carbon capture temperature-pressure set; obtain the carbon capture temperature and carbon capture pressure data during the carbon capture process in previous years, obtain a carbon capture temperature-pressure sample set, and divide the carbon capture temperature-pressure sample set into a carbon capture temperature-pressure training set and a carbon capture temperature-pressure test set; set the radial basis function as the kernel function of the SVM, with the kernel function width being χ and the penalty parameter being δ, input the carbon capture temperature-pressure training set into the SVM for training, and set the maximum number of iterations as The current number of iterations is When the current number of iterations is equal to the maximum number of iterations, stop the iteration and obtain a trained SVM; Input the carbon capture temperature and pressure test set into the trained SVM, set the error threshold as ξ. When the error between the output result of the trained SVM and the actual value is less than ξ, obtain the SVM prediction model; otherwise, adjust the kernel function width and penalty parameter until the error between the output result of the trained SVM and the actual value is less than ξ; S22. Input the carbon capture temperature and pressure set into the SVM prediction model to output a carbon capture prediction value; obtain the current carbon absorption amount and calculate the carbon absorption amount prediction value.

4. The comprehensive detection method for carbon capture performance based on an integrated system according to claim 3, characterized in that, Specifically, S22 includes: S221. Record the carbon absorption amount from previous years to the current carbon absorption amount as the carbon absorption amount set, construct a time series, use grey theory to construct a whitenization differential equation for the time series, and calculate the development coefficient, grey action quantity, and background value from the whitenization differential equation; S222. Calculate the carbon absorption amount prediction value from the development coefficient, grey action quantity, and background value.

5. A comprehensive detection method for carbon capture performance based on an integrated system according to claim 4, characterized in that Repeat S221 and S222 to obtain the carbon absorption amount prediction value, calculate the ratio of the carbon capture prediction value to the carbon absorption amount prediction value, record it as the carbon capture rate, and use the carbon capture rate as the second carbon capture performance parameter.

6. The comprehensive detection method for carbon capture performance based on an integrated system according to claim 5, wherein, Specifically, S3 includes: S31. Take the first carbon capture performance parameter and the second carbon capture performance parameter to form a carbon capture performance parameter set, generate a carbon capture performance parameter matrix, and use the relational degree clustering method to perform clustering processing on the carbon capture performance parameter matrix to obtain the relational degree between the reference vector and the comparison vector; S32. Then calculate the clustering degree of the reference vector and the comparison vector, obtain the final cluster center set through the clustering algorithm, and record the final cluster center set as the clustering center set of the comparison vector; S33. Construct an initial fuzzy model with the set of cluster centers of the comparison vectors. The initial fuzzy model includes fuzzy rules and membership functions. Divide the carbon capture performance parameter matrix into a j'-dimensional carbon capture performance parameter training set and a j'-dimensional carbon capture performance parameter test set. Set the membership function of the compensatory fuzzy neural network as a Gaussian function, the center of the membership function as φ, and the width of the membership function as The compensation degree of the compensation function of the compensatory fuzzy neural network is γ, the global error threshold is σ. The j'-dimensional carbon capture performance parameter training set is input into the compensatory fuzzy neural network. When the error of the output result of the compensatory fuzzy neural network is less than σ, the trained compensatory fuzzy neural network is obtained; otherwise, adjust the center, width of the membership function and the compensation degree of the compensation function until the error of the output result of the compensatory fuzzy neural network is less than σ. Use the pied kingfisher optimization algorithm to optimize and adjust the center, width of the membership function and the compensation degree of the compensation function; S34. Input the j'-dimensional carbon capture performance parameter test set into the trained compensation fuzzy neural network to output the final carbon capture performance parameter, and obtain the surplus waste heat utilization efficiency from the second law of thermodynamics; detect the carbon capture performance by integrating the final carbon capture performance parameter and the surplus waste heat utilization efficiency.

7. The integrated system-based comprehensive carbon capture performance detection method according to claim 6, wherein Specifically, S33 includes: S331. Set the fitness function as the error between the output result of the compensation fuzzy neural network and the actual value. The fitness function represents the ability of an individual of the pied kingfisher to solve problems. The initial position of the p-th pied kingfisher in the pied kingfisher population is The upper bound of the search range of the p-th pied kingfisher in the pied kingfisher population is l′, the lower bound of the search range of the p-th pied kingfisher in the pied kingfisher population is l″, k1 represents a random number and k1 ∈ [0, 1]. The calculation formula for the initial position of the p-th pied kingfisher in the pied kingfisher population is as follows. During the perching and hovering phases of the pied kingfisher, the fitness function value starts to be calculated. Let the current iteration number be q, and the position of the pied kingfisher at the q-th iteration be The position of the pied kingfisher at the (q + 1)-th iteration is In the pied kingfisher population, the position of the -th pied kingfisher at the q-th iteration is Let k2 denote a random number and k2 ∈ [0, 1], and η denote the state parameter. Then the calculation formula for the position of the pied kingfisher at the (q + 1)-th iteration is as follows. S332. Continuously iterate the fitness function value. During the diving stage of the pied kingfisher, the pied kingfisher continuously updates its position to find the optimal fitness function value. Set the hunting abilities of the pied kingfisher as h1 and h2, the control parameter as ν, the current best position of the pied kingfisher as x′, the hunting position of the pied kingfisher as x″, k4 represents a random number and k4 ∈ [0, 1]. Then, the updated position of the pied kingfisher in the (q + 1)-th iteration The calculation formula is as follows: After the pied kingfisher updates its position, during the growth stage, two random individuals from the pied kingfisher population are selected and denoted as and abs represents the absolute value. Set the position at the q-th iteration to be The position at the q-th iteration is The predation efficiency of the pied kingfisher is k5 represents a random number and k5 ∈ [0, 1]. Continue to find the optimal fitness function value and replace the current fitness function value with the optimal one. Then the updated position of the pied kingfisher at the (q + 1)-th iteration The calculation formula is as follows 8. A comprehensive detection method for carbon capture performance based on an integrated system according to claim 7, characterized in that When the current iteration number is equal to the maximum iteration number, stop the iteration to obtain the final position of the pied kingfisher. The coordinate values of the three-dimensional coordinates corresponding to the final position of the pied kingfisher are respectively the center of the membership function, the width, and the compensation degree of the compensation function.

9. A system for implementing a comprehensive detection method for carbon capture performance based on an integrated system as described in any one of claims 1-8, characterized in that, Specifically, it includes: a surplus waste heat compensation module, a carbon capture rate prediction module, a compensation fuzzy neural network module, and a carbon capture performance detection module; The waste heat compensation module is used to use the thermal economic evaluation index and construct a waste heat compensation system model to calculate the waste heat energy level; The carbon capture rate prediction module predicts carbon capture based on the SVM prediction model to obtain a carbon capture prediction value, and then obtains the carbon capture rate according to the grey model; The compensation fuzzy neural network module is used to train the compensation fuzzy neural network and optimize the parameters of the compensation fuzzy neural network model using the pied kingfisher optimization algorithm; The carbon capture performance detection module is used to detect the performance of the carbon capture cycle system by combining the final carbon capture performance parameters and the surplus waste heat utilization efficiency.

Images