Method and system for controlling reaction temperature of process for synthesizing methanol through hydrogenation of carbon dioxide

Abstract

The embodiment of the invention provides a method and system for controlling the reaction temperature of a process for synthesizing methanol through carbon dioxide hydrogenation, and the method comprises the steps: setting a data threshold value of a synthesis tower, and carrying out synthesis tower intake air screening based on the data threshold value; performing model training by taking intake air data of intake air of the synthesis tower and reaction temperature of the synthesis tower as input variables of a neural network model, adjusting the number of model neurons, calculating a ratio result of sample similarity of the neurons and a weighted sample output value, and outputting predicted intake air flow of the synthesis tower based on the ratio result; obtaining synthetic tower attribute data, and calculating theoretical air inlet flow based on the synthetic tower attribute data; the flow difference value of the predicted air inlet flow and the theoretical air inlet flow is compared, the difference value change rate of the flow difference value is collected, a preset fuzzy control rule is obtained on the basis of the difference value change rate, the opening degree of an electromagnetic valve is adjusted on the basis of the fuzzy control rule, the air inlet flow is controlled, control over reaction heat is achieved, and therefore the purpose of controlling the reaction temperature is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of methanol processes, and particularly to a method and system for controlling the reaction temperature of a process for synthesizing methanol by hydrogenating carbon dioxide. Background Art

[0002] As a low-carbon and environmentally friendly energy alternative, methanol shows great application potential in fields such as transportation, energy storage, and chemical raw materials. However, traditional methanol production methods often rely on fossil fuels, which not only exacerbates the carbon emission problem but also goes against the current requirements of sustainable development. Therefore, exploring a new process for synthesizing methanol by combining carbon dioxide with green power hydrogen production not only can effectively utilize the greenhouse gas carbon dioxide but also can achieve the green transformation of methanol production, which is of great significance for promoting the optimization of the energy structure and environmental protection.

[0003] However, since most current projects for synthesizing methanol by hydrogenating carbon dioxide are at the pilot or small-scale test level, there are still problems with the synthesis process, such as non-linearity, large delay, and large time variability, and the control effect is not ideal. Especially for the process of synthesizing methanol by hydrogenating carbon dioxide, due to the obvious exothermic effect of the reaction, the reaction temperature is more uncontrollable, and it is very easy to exceed the preset heating temperature, resulting in serious consequences such as catalyst sintering and reaction termination. When the reaction temperature is low, the catalytic efficiency is affected and the reaction efficiency becomes poor. Therefore, finding a method for effectively controlling the reaction temperature of the synthesis tower is an urgent problem to be solved at present. Summary of the Invention

[0004] In view of the problems existing in the prior art, an embodiment of the present invention provides a method and system for controlling the reaction temperature of a process for synthesizing methanol by hydrogenating carbon dioxide, which realizes controlling the intake air flow in real time according to the reaction temperature through neural network and fuzzy control, so as to control the reaction heat generation and thus control the reaction synthesis temperature.

[0005] An embodiment of the present invention provides a method for controlling the reaction temperature of a process for synthesizing methanol by hydrogenating carbon dioxide, the method comprising: Setting a data threshold for the synthesis tower and screening the intake air of the synthesis tower based on the data threshold; Taking the intake air data of the synthesis tower intake air and the synthesis tower reaction temperature as input variables of a neural network model for model training, adjusting the number of neurons corresponding to the intake air data in the model training, and calculating the ratio result of the sample similarity of the neurons and the weighted sample output value, and outputting the predicted intake air flow of the synthesis tower based on the ratio result; Obtaining the synthesis tower attribute data and calculating the theoretical intake air flow based on the synthesis tower attribute data; Compare the flow difference between the predicted intake air flow and the theoretical intake air flow, collect the change rate of the flow difference, obtain a preset fuzzy control rule based on the change rate, and adjust the solenoid valve opening based on the fuzzy control rule.

[0006] In one embodiment, the method further includes: Standardize the intake air data to obtain input variables, and set the number of neurons in the input layer based on the variable dimension of the input variables; Transmit the input variables to the pattern layer, set the number of neurons in the pattern layer based on the number of samples of the input variables, and calculate the dynamic relationship between the synthesis temperature of the reaction tower and other data in the input variables; Based on the dynamic relationship, calculate the sample similarity and weighted sample output value of the neurons through the summation layer; Output the ratio result of the sample similarity and weighted sample output value of the neurons through the output layer, and output the intake air flow corresponding to the synthesis temperature of the reaction tower in the summary of the input data based on the calculation formula of the ratio result, which is the predicted intake air flow.

[0007] In one embodiment, the method further includes: Among them, is the input vector of the i-th training sample, is the smoothing factor; The calculation formula of the sample similarity of the neurons includes: The calculation formula of the weighted sample output value of the neurons includes: In one embodiment, the method further includes: = = Among them, X i and Y i are the values of the training samples X0 and Y0, n is the training sample size, X0 ∈ R 3×n matrix, Y0 ∈ R 1×n matrix, (X) is the predicted output of the intake air flow of the network under the condition of variable X.

[0008] In one embodiment, the method further includes: Obtain the synthesis tower attribute data, where the attribute data includes the initial temperature of the reaction gas, pressure, the volume of the reaction gas in the synthesis tower, and the catalyst efficiency. Set the ideal temperature, and based on the thermodynamic principle, calculate the theoretical inlet gas flow rate. The calculation formula includes: V = = Wherein, T1 is the initial temperature of the reaction gas, P1 is the pressure, V is the volume of the reaction gas in the synthesis tower, ɑ is the catalyst efficiency, and T2 is the set ideal temperature.

[0009] In one embodiment, the method further includes: Calculate the flow difference between the predicted inlet gas flow rate and the theoretical inlet gas flow rate, and compare the change rate of the flow difference within a preset period. Set corresponding fuzzy variables for the flow difference and the difference change rate, and determine the universe of discourse corresponding to the fuzzy variables in combination with the historical data of the synthesis tower; Set the fuzzy subsets corresponding to the universe of discourse, and set the fuzzy subset of the valve opening. Based on fuzzy inference, adjust the valve opening, and set the adjustment period and quantization factor; Input the quantization domain corresponding to the universe of discourse, and output the valve opening level corresponding to the fuzzy subset in the corresponding valve opening based on the quantization domain.

[0010] An embodiment of the present invention provides a reaction temperature control system for a carbon dioxide hydrogenation to methanol process. The system includes: A screening module, configured to set the synthesis tower data threshold and perform synthesis tower inlet gas screening based on the data threshold; A training module, configured to use the inlet gas data of the synthesis tower inlet gas and the synthesis tower reaction temperature as input variables of a neural network model for model training, adjust the number of neurons corresponding to the inlet gas data in the model training, and calculate the ratio result of the sample similarity of the neurons to the weighted sample output value, and output the predicted inlet gas flow rate of the synthesis tower based on the ratio result; A theory module, configured to obtain the synthesis tower attribute data and calculate the theoretical inlet gas flow rate based on the synthesis tower attribute data; An adjustment module, configured to compare the flow difference between the predicted inlet gas flow rate and the theoretical inlet gas flow rate, collect the change rate of the flow difference, obtain a preset fuzzy control rule based on the change rate, and adjust the solenoid valve opening based on the fuzzy control rule.

[0011] In one embodiment, the system further includes: An input module, configured to standardize the inlet gas data to obtain input variables, and set the number of neurons in the input layer based on the variable dimension of the input variables; A mode layer module for transmitting the input variables to the mode layer, setting the number of neurons in the mode layer based on the number of samples of the input variables, and calculating the dynamic relationship between the synthesis temperature of the reaction tower and other data in the input variables; A summation module for calculating the sample similarity and weighted sample output value of neurons through a summation layer based on the dynamic relationship; An output module for outputting the ratio result of the sample similarity and weighted sample output value of neurons through an output layer, and outputting the intake air flow corresponding to the synthesis temperature of the reaction tower in the input data summary based on the calculation formula of the ratio result, which is the predicted intake air flow.

[0012] An embodiment of the present invention provides an electronic device, including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method described in one or more embodiments.

[0013] An embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned method for controlling the reaction temperature of the carbon dioxide hydrogenation to synthesize methanol process are realized.

[0014] In view of the above, in one or more embodiments of this specification, a synthesis tower data threshold is set, and synthesis tower intake air is screened based on the data threshold; the intake air data of the synthesis tower intake air is used as the input variable of the neural network model for model training, the number of neurons corresponding to the intake air data in the model training is adjusted, and the ratio result of the sample similarity and weighted sample output value of the neurons is calculated, and the predicted intake air flow of the synthesis tower is output based on the ratio result; the synthesis tower attribute data is obtained, and the theoretical intake air flow is calculated based on the synthesis tower attribute data; the flow difference between the predicted intake air flow and the theoretical intake air flow is compared, the difference change rate of the flow difference is collected, the preset fuzzy control rule is obtained based on the difference change rate, and the solenoid valve opening is adjusted based on the fuzzy control rule. In this way, by controlling the intake air volume and intake air ratio, the reaction temperature can be better controlled, and there is a certain "foresight" for the temperature change during the reaction process, overcoming the disadvantages of large nonlinear region errors and slow reaction that are prone to occur in general synthesis processes, and realizing intelligent control of the reaction temperature of the synthesis tower. Description of the Drawings

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a flowchart of a method for controlling the reaction temperature of a carbon dioxide hydrogenation to methanol process provided by an embodiment of this specification.

[0017] Figure 2 It is a schematic structural diagram of a neural network training provided by an embodiment in this specification.

[0018] Figure 3 It is a schematic diagram of a fuzzy processing of the predicted intake air volume and the theoretical intake air volume provided by an embodiment in this specification.

[0019] Figure 4 It is another diagram of a method for controlling the reaction temperature of a carbon dioxide hydrogenation to methanol process provided by an embodiment in this specification.

[0020] Figure 5 It is a schematic structural diagram of a reaction temperature control system for a carbon dioxide hydrogenation to methanol process provided by an embodiment of this specification.

[0021] Figure 6 It is a schematic structural diagram of an electronic device provided by an embodiment of this specification. Detailed implementation manners

[0022] Now, the subject matter described herein will be discussed with reference to example embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein, and is not a limitation on the scope of protection, applicability, or examples set forth in the claims. The functions and arrangements of the elements discussed can be changed without departing from the scope of protection of the content of this specification. Each example can omit, substitute, or add various processes or components as needed. For example, the described method can be executed in a different order from the described order, and each step can be added, omitted, or combined. Additionally, the features described relative to some examples can also be combined in other examples.

[0023] As used herein, the term "comprising" and variations thereof are open-ended terms meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other definitions may be included below, either explicitly or implicitly. Unless explicitly stated in the context, the definition of a term is consistent throughout the specification.

[0024] As Figure 1 shown, an embodiment of the present invention provides a method for controlling the reaction temperature of a carbon dioxide hydrogenation to methanol process, including: Step S102, set the data threshold of the synthesis tower, and screen the inlet gas of the synthesis tower based on the data threshold.

[0025] Specifically, the inlet gas data in the synthesis tower is preliminarily screened. Generally, the heat supply temperature in the synthesis tower is basically unchanged, and the temperature in the tower is mainly affected by the heat released during the reaction, while the reaction heat is mainly affected by the amount of syngas, that is, the inlet gas data. The inlet gas data may include the temperature in the synthesis tower, the current hydrogen inlet flow rate, the current carbon dioxide inlet flow rate, and the hydrogen-carbon ratio in the synthesis tower, etc. The acquisition method of the inlet gas data can be through the PLC main control system related to the synthesis tower, and the real-time data of each valve and temperature sensor in the synthesis tower is obtained through the main control system. Determine the current inlet gas data through the obtained real-time data and perform the next screening. The screening process may include: determining the preset threshold of each data. For example, taking temperature as an example, the set threshold may be 200 °C (catalyst deactivation temperature) to 250 °C (starting point of sintering risk). Further, data such as the hydrogen inlet volume, carbon dioxide inlet volume, hydrogen-carbon ratio, temperature rise, etc. can all be screened according to the corresponding data thresholds to keep the inlet gas in the synthesis tower within the preset normal range.

[0026] Step S104, use the inlet gas data of the synthesis tower inlet and the reaction temperature of the synthesis tower as input variables of the neural network model for model training, adjust the number of neurons corresponding to the inlet gas data in the model training, and calculate the ratio result of the sample similarity of the neurons and the weighted sample output value, and output the predicted inlet gas flow rate of the synthesis tower based on the ratio result.

[0027] Specifically, perform model training on the inlet gas data with a generalized regression neural network model to determine the prediction result of the inlet gas flow rate. The model structure of the network structure of the generalized regression neural network model (GRNN) inlet gas flow rate prediction model is as Figure 2 shown, in Figure 2Among them, the neural network includes 4 layers, namely the input layer, the pattern layer, the summation layer, and the output layer. Among them, compared with other neural network models, GRNN does not require backpropagation and can complete prediction through a single forward calculation, making it suitable for real-time control scenarios.

[0028] 1. The characteristic parameters of the input layer can include the intake air data in the above steps. Standardize the intake air data to avoid weight imbalance caused by dimensional differences in various data. Let the dimension of variable X be the number of neurons in the input layer. For example, when the intake air data includes the tower temperature, hydrogen intake, and carbon dioxide intake, the dimension is 3 neurons. The input layer can directly transmit the input variables to the pattern layer.

[0029] 2. The pattern layer receives the input variables. The number of neurons in the pattern layer is the same as the number of training samples n. For example, if the number of samples of the intake air data is 1000, then the corresponding n is 1000. A large sample size can cover more working conditions and improve the robustness of the model. Each neuron corresponds to a different training sample as follows: And calculate the similarity between the input and the training samples through the Gaussian kernel function, and capture the dynamic relationship between temperature and other intake air data through nonlinear modeling: Among them, is the input vector of the i-th training sample, is the smoothing factor, which is used to control the sensitivity of the model to local data.

[0030] 3. The summation layer is used to integrate the output of the pattern layer and generate the normalized weights, including two types of calculations represented by two types of neurons.

[0031] The transfer function of the first type of neuron is: Process the outputs of all neurons in the pattern layer through arithmetic summation, and accumulate the similarities of all samples.

[0032] The transfer function of the second type of neuron is: Process the outputs of all neurons in the pattern layer through weighted summation, and weighted accumulate the output values of all samples.

[0033] Among them, samples with higher similarity share more in the prediction result, and the denominator S D can be regarded as a normalization factor to ensure the physical rationality of the output.

[0034] Since the output variable of this model is the intake air flow rate, which is a one-dimensional vector, so Y iis the connection weight between the i-th neuron in the pattern layer and the neurons in the summation layer. Y i is the i-th element of the training sample Y.

[0035] 4. Corresponding results of the neurons in the output layer: = In summary, the data for model training in this embodiment includes: Let the input variable X = (Xa, Xn,..., Xp) T , the output variable Y, the known variable X, and the input samples are used for training. Under the condition that the input variable is X0, the output variable Y = Y0. Then, the predicted output of the model under the conditions of variables X0 and Y0 is: = In the formula: X i , Y i are the values of the training samples X0 and Y0, n is the training sample size of 1200, X0 ∈ R 3×n matrix, Y0 ∈ R 1×n matrix; σ is the smoothing factor of the GRNN; (X) is the predicted output of the intake air flow of the network under the condition of variable X.

[0036] In the formula, X = (X a , X n , ……, X p )T indicates that the synthesis temperature of the reaction tower of the system at this time is X a , the hydrogen intake flow rate is X n , the carbon dioxide intake flow rate is X p and other related parameters, represents the intake air flow predicted by the model when the system input is X.

[0037] Step S106: Obtain the synthesis tower attribute data and calculate the theoretical intake air flow based on the synthesis tower attribute data.

[0038] Specifically, obtain the attribute data inside the synthesis tower, including the volume of each part of the synthesis tower, the ratio of hydrogen to carbon dioxide, the intake air pressure and temperature under ideal conditions, etc. Based on the thermodynamic principle, establish a theoretical flow model inside the synthesis tower and conduct cross-validation with the GRNN prediction value to ensure the physical rationality of the control instruction. Among them, the calculation steps of the theoretical flow can include: Based on the synthesis tower being a constant-volume system where the gas does not do external work (W = 0), according to the second law of thermodynamics ΔU = W + Q, the energy Q required for the gas to adjust the temperature can be determined as Q = ΔU = + (Calculation formula 1).

[0039] Wherein, T1 is the initial temperature, T2 is the adjusted temperature (set ideal temperature), V1 is the gas volume before adjustment, V2 is the gas volume after adjustment, and n is the amount of substance of the reaction gas.

[0040] When the synthesis tower is inlet with gas, the inlet ratio of hydrogen to carbon dioxide can usually be set to 1:3, and the reaction ratio is also 1:3. The density of the synthesis gas can be calculated. .

[0041] Based on the formula n = = , it can be converted to: The amount of substance of the synthesis gas n 合 = V (Calculation formula 2) Wherein, for and , it can be determined by querying relevant gas data. And based on the fixed volume in the synthesis tower, it can be determined that = 0 (Calculation formula 3).

[0042] Substituting the above Calculation formulas 2 and 3 into Calculation formula 1, we can get Q = ΔU = (Calculation formula 4), wherein, is the molar heat capacity at constant volume.

[0043] For the reaction of hydrogenation of carbon dioxide to methanol in the synthesis tower in theory, CO2 + 3H2 = CH3OH + H2O, the reaction enthalpy ΔH in the chemical formula can be determined by relevant data of the methanol reaction, and the catalyst efficiency is ɑ. The reaction heat Q generated can be obtained as Q = ɑnΔH.

[0044] That is, the amount of substance of the reaction n 反 = (Calculation formula 5) Combining with the gas state equation PV = nRT (Calculation formula 6) Assuming that the initial temperature and pressure of the gas are T1 and P1 respectively, the volume of the reaction gas in the synthesis tower is V, and the set ideal temperature is T2. Combining Calculation formulas 4, 5, and 6, the data of the corresponding increase or decrease in the inlet gas flow rate for adjusting the temperature can be calculated: V = = Step S108, comparing the flow difference between the predicted inlet gas flow rate and the theoretical inlet gas flow rate, collecting the change rate of the flow difference, obtaining a preset fuzzy control rule based on the change rate, and adjusting the solenoid valve opening based on the fuzzy control rule.

[0045] Specifically, after comparing and determining the predicted flow rate and theoretical flow rate of GNRR, the flow rate difference is obtained, and the difference change rate of the flow rate difference is collected. Nonlinear dynamic adjustment of the intake air flow rate is achieved through fuzzy logic, improving the system robustness while ensuring the control accuracy. The architecture of the processing system can be as Figure 3 shown, including: 1. Processing of input variables, including: Deviation signal (intake air quantity deviation e) e = v - k, where v is the predicted flow rate of GRNN and k is the actual intake air quantity.

[0046] ec = (v - k)(t + 1) / (v - k)t (t = N, collecting the values of e in different time periods, sampling every 5 minutes) is the change rate of the difference.

[0047] Among them, let the fuzzy variables of the input variables e and ec be E and EC, and the universes of discourse of E and EC can be determined based on historical data statistics.

[0048] 2. Setting up the fuzzy rule base Select E and EC as the input to form fuzzy subsets with 7 variable languages. The fuzzy subset of E is {FG, JG, GZ, SZ, SR, JR, FR}, and the fuzzy subset of EC is {FK, FZ, FM, ZR, ZM, ZZ, ZK}. In addition, the corresponding output P can also determine the fuzzy subset as {JC, JD, ZR, JD, JC, JZ, JC} according to the output result.

[0049] For the above fuzzy subsets, for example, when the deviation is extremely large (FG) and the change rate drops extremely fast (FK), the valve opening needs to be increased significantly (JC). Among them, for the setting rule table of the fuzzy subsets, it can include: EC\E FG (negative large) JG (negative relatively large) GZ (negative medium) SZ (negative small) SR (moderate) JR (positive relatively large) FR (positive large) FK (negative fast) JC (decrease) JC JC JD ZR JD JD FZ (negative medium fast) JC JC JD JD JD JZ JZ FM (negative slow) JD JD JD ZR JZ JZ JC ZR (zero) ZR JD JD ZR JD JD ZR ZM (positive slow) JD JZ JZ JD JD JD JC ZZ (positive medium fast) JZ JZ JD JD JD JC JC ZK (positive fast) JZ JD JD JC JC JC JC Rule 1: If the error E is negative large (FG) and the error change rate EC is negative fast (FK), then the output P is decrease (JC). The error is negative and decreasing rapidly, and the control amount needs to be reduced to avoid overshoot.

[0050] Rule 2: If the error E is moderate (SR) and the error change rate EC is zero (ZR), then the output P is zero (ZR). The system is close to the stable state and there is no need to adjust the control amount.

[0051] Rule 3: If the error E is positive large (FR) and the error change rate EC is positive fast (ZK), then the output P is decrease (JC). The error is positive and increasing rapidly, and the control amount needs to be strongly reduced (possibly adjusted in the reverse direction).

[0052] 3. Defuzzification of the output variable Based on the above fuzzy subset P = {ZR, JD, JZ, JC}, different valve openings can be represented respectively. For example, they can represent closed, slightly open, moderately open, and fully open. It is also possible to combine with the quantization factor and map the fuzzy subset to the corresponding physical quantity data within the domain of the valve opening.

[0053] The output variable of the fuzzy control is the solenoid valve opening. By changing the solenoid valve opening, the intake air volume of the system can be controlled. It can be set that one cycle is 10 minutes, and the basic domain of the control variable output p is [0, 50]. Four fuzzy languages are selected for the fuzzy variable P, and the valve opening is divided into four levels. The fuzzy subset of P is {ZR, JD, JZ, JC}. The scale factor kp = 20 is set, and the quantization domain is {-m, -m + 1, …, 0, 1, m, m + 1}. The quantization domain of each parameter can be set as {-3, -2, -1, 0, 1, 2, 3}, EC = {-3, -2, -1, 0, 1, 2, 3}, P = {0, 1, 2, 3}.

[0054] In the fuzzy control system, defuzzification is the key step to convert the linguistic variables (such as "slightly open", "moderately open") obtained from fuzzy reasoning into precise physical quantities (such as the specific numerical value of the valve opening). The following is a step-by-step explanation of the logic and process of defuzzifying the output variable: 1. Mapping of the fuzzy output variable to the physical quantity Fuzzy subset P: {ZR, JD, JZ, JC} Assume the corresponding valve opening levels: ZR: Closed (0% opening) JD: Slightly open (low opening, e.g., 25%) JZ: Moderately open (medium opening, e.g., 50%) JC: Fully open (high opening, e.g., 75% or 100%) In practical applications, the physical quantities corresponding to the levels need to be adjusted according to the control accuracy. For example, if it is divided into "closed, slightly open, moderately open, fully open", there are 4 levels, which is the same as the number of fuzzy subsets. 2. Domain and quantization parameters Basic domain: The actual range of the physical quantity, which is the valve opening [0, 50] here (unit: % or scale value).

[0055] Quantization domain: The value range of the fuzzy variable in the discrete space, usually a symmetric integer interval.

[0056] Quantization domain of EC and E: {-3, -2, -1, 0, 1, 2, 3} (a total of 7 levels, symmetric distribution, suitable for positive and negative inputs).

[0057] Quantization domain of P: {0, 1, 2, 3}, a total of 4 levels, asymmetric distribution, applicable to non - negative outputs (e.g., the valve opening cannot be negative).

[0058] Scale factor (kp): Used to map the discrete values of the quantization domain to the physical quantities in the basic domain. The formula is: Actual value = Quantized value × kp Here, kp = 20, but attention should be paid to the matching between the quantization domain and the basic domain.

[0059] Among them, the analysis of the defuzzification process includes: 1. Quantization of the fuzzy inference result Suppose after reasoning through the fuzzy rule base, the fuzzy set of the output variable P obtains a quantized value z through the maximum membership degree method or the centroid method, and this value belongs to the quantization domain {0, 1, 2, 3} of P.

[0060] For example: If the reasoning result is "small opening (JD)", the corresponding quantized value z = 1; "medium opening (JZ)" corresponds to z = 2.

[0061] 2. Mapping from the quantized value to the physical quantity Convert the quantized value to the actual valve opening through the scale factor kp: Valve opening = z × kp Problem analysis: The quantization domain is {0, 1, 2, 3}. If kp = 20, the mapping result is {0, 20, 40, 60}, but the basic domain is [0, 50], and the maximum value 60 exceeds the range.

[0062] Correction suggestion: Adjust kp so that the maximum quantized value corresponds to the upper limit of the basic domain, that is: kp = Maximum value of the quantization domain / Upper limit of the basic domain = 50 / 3 ≈ 16.67. At this time, the mapping result is {0, 16.67, 33.33, 50}, which is within the range of [0, 50].

[0063] If rounding or truncation is allowed, when kp = 20, the quantized value 3 corresponds to 60, which can be forced to be limited to 50 (saturation processing).

[0064] 3. Periodic control and dynamic adjustment The system updates the output every 10 minutes as a cycle, which means calculating the valve opening every 10 minutes according to the current error (E) and the error change rate (EC), avoiding oscillations caused by frequent adjustments.

[0065] Example of control logic: If the current error is large and is growing positively (EC is positive and fast), the output quantized value is 3 (large opening), corresponding to an opening of 3 × kp; If the system is approaching stability (E = moderate, EC = zero), the output quantized value is 0 (closed), and the opening is maintained at 0.

[0066] A method for controlling the reaction temperature of a process for synthesizing methanol by hydrogenating carbon dioxide provided by an embodiment of the present invention sets data thresholds for a synthesis tower and screens the inlet gas of the synthesis tower based on the data thresholds; uses the inlet gas data of the synthesis tower as input variables of a neural network model for model training, adjusts the number of neurons corresponding to the inlet gas data in the model training, and calculates the ratio of the sample similarity of the neurons to the weighted sample output value, and outputs the predicted inlet gas flow of the synthesis tower based on the ratio result; obtains the synthesis tower attribute data, and calculates the theoretical inlet gas flow based on the synthesis tower attribute data; compares the flow difference between the predicted inlet gas flow and the theoretical inlet gas flow, collects the difference change rate of the flow difference, obtains a preset fuzzy control rule based on the difference change rate, and adjusts the solenoid valve opening based on the fuzzy control rule. In this way, by controlling the inlet gas volume and inlet gas ratio, the reaction temperature can be better controlled, and there is a certain "foresight" for the temperature change during the reaction process, overcoming the disadvantages of large nonlinear region errors and slow reaction that are prone to occur in general synthesis processes, and realizing the intelligent control of the reaction temperature of the synthesis tower.

[0067] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of a reaction temperature control system for a process for synthesizing methanol by hydrogenating carbon dioxide provided by an embodiment of the present application. As Figure 5 shown, the system includes: A screening module S502 for setting data thresholds for a synthesis tower and screening the inlet gas of the synthesis tower based on the data thresholds; A training module S504 for using the inlet gas data of the synthesis tower as input variables of a neural network model for model training, adjusting the number of neurons corresponding to the inlet gas data in the model training, and calculating the ratio of the sample similarity of the neurons to the weighted sample output value, and outputting the predicted inlet gas flow of the synthesis tower based on the ratio result; A theory module S506 for obtaining synthesis tower attribute data and calculating the theoretical inlet gas flow based on the synthesis tower attribute data; An adjustment module S508 for comparing the flow difference between the predicted inlet gas flow and the theoretical inlet gas flow, collecting the difference change rate of the flow difference, obtaining a preset fuzzy control rule based on the difference change rate, and adjusting the solenoid valve opening based on the fuzzy control rule.

[0068] In another embodiment, a reaction temperature control system for a process for synthesizing methanol by hydrogenating carbon dioxide further includes: An input module for standardizing the inlet gas data to obtain input variables, and setting the number of neurons in the input layer based on the variable dimension of the input variables; A mode layer module for transmitting the input variables to the mode layer, setting the number of neurons in the mode layer based on the number of samples of the input variables, and calculating the dynamic relationship between the synthesis temperature of the reaction tower and other data in the input variables; A summation module for calculating the sample similarity and weighted sample output value of neurons through a summation layer based on the dynamic relationship; An output module for outputting the ratio result of the sample similarity and weighted sample output value of neurons through an output layer, and outputting the intake air flow corresponding to the synthesis temperature of the reaction tower in the input data summary based on the calculation formula of the ratio result, which is the predicted intake air flow.

[0069] Those skilled in the art can clearly understand that the technical solutions of the embodiments of the present application can be implemented by means of software and / or hardware. The "units" and "modules" in this specification refer to software and / or hardware that can independently complete or cooperate with other components to complete specific functions, where the hardware can be, for example, a Field-Programmable Gate Array (FPGA), an Integrated Circuit (IC), etc.

[0070] Each processing unit and / or module of the embodiments of the present application can be implemented by an analog circuit that implements the functions described in the embodiments of the present application, or can be implemented by software that executes the functions described in the embodiments of the present application.

[0071] See Figure 6 which shows a schematic structural diagram of an electronic device related to the embodiments of the present application. This electronic device can be used to implement Figure 1 the method in the embodiments shown. As Figure 6 shown, the electronic device 600 may include: at least one processor 601, at least one network interface 604, a user interface 603, a memory 605, and at least one communication bus 602.

[0072] Among them, the communication bus 602 is used to realize the connection and communication between these components.

[0073] Among them, the user interface 603 may include a display screen (Display), a camera (Camera). Optionally, the user interface 603 may further include a standard wired interface and a wireless interface.

[0074] Among them, the network interface 604 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0075] Among them, the processor 601 may include one or more processing cores. The processor 601 connects various parts within the entire electronic device 600 through various interfaces and circuits. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 605, and by invoking data stored in the memory 605, it performs various functions of the electronic device 600 and processes data. Optionally, the processor 601 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 601 may integrate a combination of one or more of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 601 and may be implemented separately by a single chip.

[0076] Among them, the memory 605 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 605 includes a non-transitory computer-readable storage medium. The memory 605 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 605 may include a program storage area and a data storage area. Among them, the program storage area can store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store the data involved in the above-mentioned various method embodiments. Optionally, the memory 605 may also be at least one storage device located far from the aforementioned processor 601. As Figure 6 shown, the memory 605, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and program instructions.

[0077] In Figure 6In the electronic device 600 shown, the user interface 603 is mainly used to provide an interface for the user to input and obtain the data input by the user. The processor 601 can be used to call the interactive application program generated based on images stored in the memory 605 and specifically perform the following operations: set the data threshold of the synthesis tower, and screen the intake air of the synthesis tower based on the data threshold; use the intake air data of the synthesis tower as the input variable of the neural network model for model training, adjust the number of neurons corresponding to the intake air data in the model training, and calculate the ratio result of the sample similarity of the neurons and the weighted sample output value, and output the predicted intake air flow of the synthesis tower based on the ratio result; obtain the synthesis tower attribute data, and calculate the theoretical intake air flow based on the synthesis tower attribute data; compare the flow difference between the predicted intake air flow and the theoretical intake air flow, collect the change rate of the flow difference, obtain the preset fuzzy control rule based on the change rate of the difference, and adjust the solenoid valve opening based on the fuzzy control rule.

[0078] This application also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the above method are implemented. Among them, the computer-readable storage medium can include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, and magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0079] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0080] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0081] In several embodiments provided in this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection between each other can be through some service interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0082] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0083] In addition, each functional unit in various embodiments of this application can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0084] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. And the aforementioned memory includes: USB flash drive, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk, or optical disc, etc., which can store program codes.

[0085] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable memory. The memory can include: flash drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disc, etc.

[0086] The foregoing describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for controlling the reaction temperature in the process of synthesizing methanol by hydrogenating carbon dioxide, the method comprising: Setting the data threshold of the synthesis tower and screening the inlet gas of the synthesis tower based on the data threshold; Taking the inlet gas data of the synthesis tower inlet and the reaction temperature of the synthesis tower as input variables of a neural network model for model training, adjusting the number of neurons corresponding to the inlet gas data in the model training, and calculating the ratio of the sample similarity of the neurons to the weighted sample output value. Based on the ratio result, the predicted inlet gas flow rate of the synthesis tower is output; Obtaining the synthesis tower attribute data and calculating the theoretical inlet gas flow rate based on the synthesis tower attribute data; Comparing the flow difference between the predicted inlet gas flow rate and the theoretical inlet gas flow rate, collecting the change rate of the flow difference, obtaining a preset fuzzy control rule based on the change rate, and adjusting the solenoid valve opening based on the fuzzy control rule.

2. The method according to claim 1, characterized in that The model training includes: Normalizing the inlet gas data to obtain input variables, and setting the number of neurons in the input layer based on the variable dimension of the input variables; Transmitting the input variables to the pattern layer, setting the number of neurons in the pattern layer based on the number of samples of the input variables, and calculating the dynamic relationship between the synthesis temperature of the reaction tower and other data in the input variables; Based on the dynamic relationship, calculating the sample similarity of the neurons and the weighted sample output value through the summation layer; Outputting the ratio result of the sample similarity of the neurons to the weighted sample output value through the output layer, and based on the calculation formula of the ratio result, outputting the inlet gas flow rate corresponding to the synthesis temperature of the reaction tower summarized by the input data, which is the predicted inlet gas flow rate.

3. The method according to claim 2, characterized in that, The calculation formula for the dynamic relationship between the synthesis temperature of the reaction tower and other data includes: Among them, is the input vector of the i-th training sample, is the smoothing factor; The calculation formula for the sample similarity of the neurons includes: The calculation formula for the weighted sample output value of the neurons includes: 。 4. The method according to claim 3, wherein Based on the calculation formula of the ratio result, outputting the inlet gas flow rate corresponding to the synthesis temperature of the reaction tower summarized by the input data includes: = = Among them, X i , Y i are the values of the training samples X0 and Y0, n is the training sample size, X0 ∈ R 3×n matrix, Y0 ∈ R 1×n matrix, (X) is the predicted output of the intake air flow of the network under the condition of variable X.

5. The method according to claim 4, wherein Calculating the theoretical inlet gas flow rate based on the synthesis tower attribute data includes: Obtaining the synthesis tower attribute data, where the attribute data includes the initial temperature of the reaction gas, pressure, the volume of the reaction gas in the synthesis tower, and the catalyst efficiency. Setting the ideal temperature as, and combining with the thermodynamic principle, calculating the theoretical inlet gas flow rate. The calculation formula includes: V= = Where, T1 is the initial temperature of the reaction gas, P1 is the pressure, V is the volume of the reaction gas in the synthesis tower, ɑ is the catalyst efficiency, and T2 is the set ideal temperature.

6. The method according to claim 1, wherein The method further includes: Calculating the flow difference between the predicted inlet gas flow rate and the theoretical inlet gas flow rate, and comparing the change rate of the flow difference within a preset period. Setting corresponding fuzzy variables for the flow difference and the change rate of the difference, and determining the domain of discourse corresponding to the fuzzy variables in combination with the historical data of the synthesis tower; Setting the fuzzy subsets corresponding to the domain of discourse, and setting the fuzzy subset of the valve opening. Based on fuzzy reasoning, adjusting the valve opening, and setting the adjustment period and quantization factor; Inputting the quantization domain corresponding to the domain of discourse, and outputting the valve opening level corresponding to the fuzzy subset in the corresponding valve opening based on the quantization domain.

7. A comprehensive evaluation system for the disposal efficiency of security prevention and warning events, characterized in that, The system includes; A screening module for setting the data threshold of the synthesis tower and screening the inlet gas of the synthesis tower based on the data threshold; A training module, which is used to use the intake data of the synthesis tower inlet and the reaction temperature of the synthesis tower as input variables of a neural network model for model training, adjust the number of neurons corresponding to the intake data in the model training, calculate the ratio of the sample similarity of the neurons to the weighted sample output value, and output the predicted intake flow of the synthesis tower based on the ratio result; A theory module, which is used to obtain the synthesis tower attribute data and calculate the theoretical intake flow based on the synthesis tower attribute data; An adjustment module, which is used to compare the flow difference between the predicted intake flow and the theoretical intake flow, collect the change rate of the flow difference, obtain a preset fuzzy control rule based on the change rate, and adjust the solenoid valve opening based on the fuzzy control rule.

8. The system according to claim 7, wherein The system further includes: An input module, which is used to standardize the intake data to obtain input variables, and set the number of neurons in the input layer based on the variable dimension of the input variables; A pattern layer module, which is used to transmit the input variables to the pattern layer, set the number of neurons in the pattern layer based on the number of samples of the input variables, and calculate the dynamic relationship between the synthesis temperature of the reaction tower and other data in the input variables; A summation module, which is used to calculate the sample similarity of the neurons and the weighted sample output value through the summation layer based on the dynamic relationship; An output module, which is used to output the ratio result of the sample similarity of the neurons to the weighted sample output value through the output layer, and output the intake flow corresponding to the synthesis temperature of the reaction tower in the input data summary based on the calculation formula of the ratio result, which is the predicted intake flow.

9. An electronic device, including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1-6.

10. A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method according to any one of claims 1-6 is implemented.

Images