A zinc pyridine carboxylate complex composite catalyst

Abstract

The present application relates to the technical field of composite catalysts, and specifically discloses a zinc pyridine carboxylate complex composite catalyst, wherein the zinc pyridine carboxylate complex composite catalyst is prepared by the following preparation method, the preparation method comprising feeding a zinc pyridine carboxylate complex, a double metal cyanide catalyst and a mixed solvent into a high pressure reactor at one time; using an ultrasonic device to ultrasonically dissolve a mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent; and preparing the zinc pyridine carboxylate complex composite catalyst by reacting in an inert gas atmosphere. In particular, in order to improve the efficiency of ultrasonic dissolution, the parameters of the ultrasonic device are adaptively controlled based on the dissolution rate of the reactants during the ultrasonic dissolution process to improve the dissolution efficiency and dissolution effect.

Description

Technical Field

[0001] The present application relates to the technical field of composite catalysts, and more specifically, to a zinc pyridinecarboxylate complex composite catalyst. Background Art

[0002] Carbon dioxide (CO2), as an abundant carbon resource, has received extensive attention in recent years, especially in its application as a raw material for the industrial production of aliphatic polycarbonates. Polycarbonate materials have the characteristics of high transparency, durability and biodegradability, so they have broad application prospects in the healthcare, food packaging and automotive industries. In particular, polypropylene carbonate (PPC) has a "carbon fixation" function, which helps to alleviate the greenhouse effect and plastic pollution problems, and exhibits good ductility, oxygen barrier properties and biocompatibility.

[0003] However, in the preparation process of polypropylene carbonate (PPC), the existing industrial process mainly adopts heterogeneous catalysis, such as ternary rare earth catalysts, zinc carboxylate catalysts and bimetallic catalysts, but the existing catalysts have problems such as poor catalytic activity and poor selectivity, resulting in poor quality of PPC products, low conversion rate, and long reaction time, which limits its industrial production and application. Two major problems are particularly prominent in the existing PPC production technology: one is that it is difficult to achieve high conversion rate and high carbonate content at the same time. In the preparation process of polypropylene carbonate (PPC), low temperature helps to increase the carbonate chain content in the product, but the conversion rate is low; high temperature helps to increase the conversion rate, but it is inevitably at the expense of reducing the carbonate chain content. Therefore, the high conversion rate and high carbonate content of polypropylene carbonate have not been able to be taken into account.

[0004] Second, the existing catalyst activity is low, resulting in a long production process time, usually requiring 20 hours or even longer, increasing energy consumption and reducing production efficiency. In addition, due to the high viscosity of PPC, high-temperature reactions are prone to temperature runaway, and the generated PPC may wrap the carbon dioxide required for the reaction, hindering further reactions and thus reducing the conversion rate.

[0005] Therefore, an optimized composite catalyst is desired, which can solve the problem that the prior art polypropylene carbonate cannot achieve both high conversion rate and high carbonate content. Summary of the invention

[0006] The present application provides a zinc pyridinecarboxylate complex composite catalyst, which can achieve a high propylene oxide conversion rate on the basis of ensuring a high carbonate content, the conversion rate is above 95%, and significantly shortens the polymerization reaction time, so that the product has a high carbonate content, a high molecular weight, a high conversion rate and a low by-product content, solves the problem that the prior art polypropylene carbonate cannot have both a high conversion rate and a high carbonate content, and solves the problem that the production process time is too long, and is very suitable for industrial production.

[0007] In a first aspect, a zinc pyridine carboxylate complex composite catalyst is provided, wherein the zinc pyridine carboxylate complex composite catalyst comprises a zinc pyridine carboxylate complex and a double metal cyanide catalyst, wherein the molar ratio of the zinc pyridine carboxylate complex to the double metal cyanide catalyst is 1:1-60, and the structural formula of the zinc pyridine carboxylate complex is as shown in Formula 1:

[0008]

[0009] Formula 1,

[0010] Wherein, R1 is an electron withdrawing group selected from -CX3 (X=F), -CN, -F, -Cl, -Br, -I; R2 is an electron donating group selected from -O-, -NHR3, -NH2, -OH, -OR4, -R5, -Ph, wherein R3, R4, R5 are any alkyl groups.

[0011] In a possible implementation, the average volume particle size of the zinc pyridine carboxylate complex composite catalyst is 10 nm to 16 nm, and the specific surface area of ​​the zinc pyridine carboxylate complex composite catalyst is 16 m 2 / g~170 m 2 / g.

[0012] In a possible implementation, the zinc pyridinecarboxylate complex composite catalyst has a Zn-Zn bond with a bond length of 0.01Å to 3.0Å and a Zn-Co bond with a bond length of 0.1Å to 5.0Å.

[0013] In a possible implementation, the zinc pyridinecarboxylate complex composite catalyst has the following elemental composition, by mass percentage: Zn 10% to 30%, Co 5% to 20%, C 30% to 40%, N 4% to 20%, O 10% to 20%, and other substituent elements 1% to 20%.

[0014] In a possible implementation, the double metal cyanide catalyst has a structural formula of: Zn3[Co(CN)6]2:x ZnCl2•y tBuOH·z H2O, wherein the x value ranges from 0.5 to 2.0, the y value ranges from 0.9 to 3.0, and the z value ranges from 1.0 to 3.5.

[0015] In a possible implementation, the method for preparing the zinc pyridinecarboxylate complex composite catalyst comprises the steps of:

[0016] Adding a zinc pyridine carboxylate complex, a double metal cyanide catalyst and a mixed solvent into a high-pressure reactor at one time, wherein the mixed solvent is a mixed solvent comprising a polar solvent and a non-polar solvent, wherein the polar solvent is selected from one or more of water, acetonitrile, methanol, ethanol, acetone, isopropanol, acetic acid, and ether, and the non-polar solvent is selected from one or more of toluene, benzene, carbon tetrachloride, hexane, and dichloromethane;

[0017] Using an ultrasonic device to ultrasonically dissolve the mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent;

[0018] The zinc pyridinecarboxylate complex composite catalyst is prepared by reaction in an inert gas atmosphere.

[0019] In a possible implementation, the reaction temperature of the high-pressure reaction is 20° C.-150° C., and the reaction time is 12 h-38 h; and the pressure of the inert gas atmosphere is 0.2-1 MPa.

[0020] In a possible implementation, ultrasonic dissolution of a mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent is performed using an ultrasonic device, comprising:

[0021] Obtaining a time queue of ultrasonic power values ​​of the ultrasonic device;

[0022] A time queue for obtaining the real-time temperature of the solution collected by the temperature sensor;

[0023] Inputting the time queue of the ultrasonic power value and the time queue of the real-time temperature of the solution into a temporal encoder based on a 1D-CNN model to obtain a sequence of local temporal correlation feature vectors of the ultrasonic power and a sequence of local temporal correlation feature vectors of the real-time temperature of the solution;

[0024] Inputting the sequence of the local time series associated feature vectors of the ultrasonic power and the sequence of the local time series associated feature vectors of the real-time temperature of the solution into the node time series aggregation network guided by feature significance respectively to obtain the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector;

[0025] Inputting the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector into a feature local-global interactive response module based on feature decoupling to obtain an ultrasonic power-solution temperature local-global interactive response matrix;

[0026] The ultrasonic power-solution temperature local-global interaction response matrix is ​​input into a classifier-based ultrasonic power adaptive optimization module to obtain an optimization instruction, wherein the optimization instruction is used to indicate whether the ultrasonic power value at the next time point should increase, decrease, or remain unchanged.

[0027] In a possible implementation, the sequence of the local temporal correlation feature vectors of the ultrasonic power and the sequence of the local temporal correlation feature vectors of the solution real-time temperature are respectively input into a feature significance-guided node temporal aggregation network to obtain an ultrasonic power temporal significant aggregation representation vector and a solution real-time temperature temporal significant aggregation representation vector, including:

[0028] Calculating a characteristic significance description factor of each ultrasonic power local time series correlation feature vector in the sequence of the ultrasonic power local time series correlation feature vectors, wherein the characteristic significance description factor is related to the mean and variance of each ultrasonic power local time series correlation feature vector;

[0029] Constructing a characteristic significance attenuation factor of each ultrasonic power local time series association feature vector based on the distance span between each ultrasonic power local time series association feature vector in the sequence of ultrasonic power local time series association feature vectors and the current ultrasonic power local time series association feature vector;

[0030] Modulating the characteristic significance description factors of the respective ultrasonic power local time series associated characteristic vectors based on the characteristic significance attenuation factors of the respective ultrasonic power local time series associated characteristic vectors to obtain a sequence of characteristic significance attenuation description factors;

[0031] Inputting the sequence of feature significance attenuation description factors into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors;

[0032] Taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum of the sequence of the ultrasound power local time series associated feature vectors to obtain the ultrasound power time series significant aggregation representation vector;

[0033] Among them, based on the characteristic significance attenuation factors of the respective ultrasonic power local time series associated feature vectors, the characteristic significance description factors of the respective ultrasonic power local time series associated feature vectors are modulated to obtain a sequence of characteristic significance attenuation description factors, including: calculating the positional product between the characteristic significance attenuation factors of the respective ultrasonic power local time series associated feature vectors and the characteristic significance description factors of the respective ultrasonic power local time series associated feature vectors to obtain a sequence of characteristic significance attenuation description factors.

[0034] In a possible implementation, a feature significance description factor of each ultrasonic power local time series correlation feature vector in the sequence of the ultrasonic power local time series correlation feature vector is calculated, wherein the feature significance description factor is related to the mean and variance of each ultrasonic power local time series correlation feature vector, including:

[0035] Respectively calculating the mean and variance of the ultrasonic power local time series correlation feature vector to obtain the ultrasonic power local time series correlation feature mean and the ultrasonic power local time series correlation feature variance;

[0036] After calculating the fourth power of the position difference between the ultrasonic power local time series association feature vector and the ultrasonic power local time series association feature mean, the expected value of the fourth power modulated semantic deviation vector is calculated to obtain the ultrasonic power local time series association feature expected factor;

[0037] The position-based division between the expected factor of the ultrasonic power local time series correlation feature and the square of the ultrasonic power local time series correlation feature variance is calculated to obtain a feature significance description factor of the ultrasonic power local time series correlation feature vector.

[0038] In a possible implementation, constructing a characteristic significance attenuation factor of each ultrasonic power local time series association feature vector based on the distance span between each ultrasonic power local time series association feature vector in the sequence of ultrasonic power local time series association feature vectors and the current ultrasonic power local time series association feature vector includes:

[0039] Calculating the distance span between other ultrasonic power local time series association feature vectors in the sequence of the ultrasonic power local time series association feature vectors and the current ultrasonic power local time series association feature vector to obtain a feature value distance span;

[0040] Extracting the maximum value of the current ultrasonic power local time series correlation feature vector to obtain the current ultrasonic power local time series correlation feature maximum value;

[0041] Extracting the maximum value of the other ultrasonic power local time series correlation feature vectors to obtain the maximum value of other ultrasonic power local time series correlation features;

[0042] The difference between the current ultrasonic power local time series correlation feature maximum value and the other ultrasonic power local time series correlation feature maximum values ​​is calculated and then divided by the feature value distance span to obtain the feature significance attenuation factor.

[0043] In a possible implementation, the sequence of feature significance attenuation description factors is input into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors, including:

[0044] Taking the negative number of each feature significance attenuation description factor in the sequence of feature significance attenuation description factors as an exponential power, calculating an exponential function value with a natural constant e as a base to obtain a sequence of feature significance attenuation class support description factors;

[0045] Calculating the reciprocal of the sum of each feature significance attenuation class support description factor in the sequence of feature significance attenuation class support description factors and a constant one to obtain a sequence of feature significance attenuation weight factors;

[0046] Based on a predetermined threshold, the sequence of feature significance attenuation weight factors is masked to obtain a sequence of masked feature significance attenuation weight factors, wherein, in response to each feature significance attenuation weight factor in the sequence of feature significance attenuation weight factors being greater than the predetermined threshold, the feature significance attenuation weight factor greater than the predetermined threshold is taken as the original value, otherwise it is set to zero.

[0047] In a possible implementation, taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum of the sequence of the ultrasound power local time series correlation feature vectors to obtain the ultrasound power time series significant aggregation representation vector, including:

[0048] Taking the sequence of the mask feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum of each other ultrasonic power local time series association feature vector in the sequence of ultrasonic power local time series association feature vectors to obtain a historical ultrasonic power local time series association feature vector;

[0049] The position-based sum of the historical ultrasonic power local time series correlation feature vector and the current ultrasonic power local time series correlation feature vector is calculated to obtain the ultrasonic power time series significant aggregation representation vector.

[0050] In a possible implementation, the ultrasound power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are input into a feature local-global interactive response module based on feature decoupling to obtain an ultrasound power-solution temperature local-global interactive response matrix, including:

[0051] Performing feature decoupling on the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector to obtain a sequence of ultrasonic power local time series semantic representation vectors and a sequence of solution real-time temperature local time series semantic representation vectors;

[0052] Calculating the local semantic association interaction vectors between each corresponding group of the sequence of the ultrasonic power local temporal semantic representation vectors and the sequence of the solution real-time temperature local temporal semantic representation vectors to obtain a sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors;

[0053] Calculating the fine-grained attention weight of each ultrasonic power-solution temperature local temporal semantic association interaction vector in the sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights;

[0054] Inputting the sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights into a random deactivation module to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights after deactivation processing;

[0055] Taking the sequence of fine-grained attention weights of the local temporal sequence of ultrasonic power-solution temperature after the inactivation process as the weight distribution, weighted optimization is performed on the sequence of the local temporal sequence semantic association interaction vectors of ultrasonic power-solution temperature to obtain an optimized sequence of the local temporal sequence semantic association interaction vectors of ultrasonic power-solution temperature;

[0056] Matrixing the optimized sequence of ultrasound power-solution temperature local time series semantic association interaction vectors to obtain an ultrasound power-solution temperature time series fine-grained interaction global semantic representation matrix;

[0057] Performing autocorrelation interaction perception analysis on the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix to obtain the ultrasonic power-solution temperature local-global interaction response matrix;

[0058] The step of calculating the fine-grained attention weights of each ultrasonic power-solution temperature local temporal semantic association interaction vector in the sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights includes:

[0059] The ultrasonic power-solution temperature local time series semantic association interaction vector and the weight matrix are matrix-multiplied and then added to the bias vector by position to obtain a linear transformation ultrasonic power-solution temperature local time series semantic association interaction vector;

[0060] The linear transformation ultrasonic power-solution temperature local temporal semantic association interaction vector is activated using a tanh function and then vector-multiplied with a weight coefficient vector to obtain an ultrasonic power-solution temperature local temporal fine-grained attention weight.

[0061] In a possible implementation, calculating the local semantic association interaction vector between each group of corresponding ultrasonic power local temporal semantic representation vectors and solution real-time temperature local temporal semantic representation vectors in the sequence of the ultrasonic power local temporal semantic representation vectors and the sequence of the solution real-time temperature local temporal semantic representation vectors to obtain a sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors, including:

[0062] Perform position difference, position dot multiplication and position addition processing on each group of corresponding ultrasonic power local time series semantic representation vectors and solution real-time temperature local time series semantic representation vectors to obtain ultrasonic power-solution temperature local time series semantic difference representation vectors, ultrasonic power-solution temperature local time series semantic dot product representation vectors and ultrasonic power-solution temperature local time series semantic addition representation vectors;

[0063] Cascading the ultrasonic power-solution temperature local time series semantic difference representation vector, the ultrasonic power-solution temperature local time series semantic dot product representation vector and the ultrasonic power-solution temperature local time series semantic addition representation vector to obtain an ultrasonic power-solution temperature local time series semantic association multidimensional representation fusion vector;

[0064] The ultrasonic power-solution temperature local temporal semantic association multidimensional representation fusion vector is subjected to one-dimensional convolution encoding and then passed through a maximum pooling layer to obtain an ultrasonic power-solution temperature local temporal semantic association interaction vector.

[0065] In a possible implementation, the ultrasound power-solution temperature time series fine-grained interaction global semantic representation matrix is ​​subjected to autocorrelation interaction perception analysis to obtain the ultrasound power-solution temperature local-global interaction response matrix, including:

[0066] Performing matrix multiplication of the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix and the transposed matrix of the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix, and then dividing by the scale of the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix by position to obtain the ultrasonic power-solution temperature time series fine-grained interaction feature global dependency matrix;

[0067] The Relu function is used to activate the global dependency matrix of the ultrasonic power-solution temperature time series fine-grained interaction features to obtain the ultrasonic power-solution temperature local-global interaction response matrix.

[0068] The present application provides a zinc pyridine carboxylate complex composite catalyst, the composite catalyst comprising a zinc pyridine carboxylate complex and a double metal cyanide catalyst (DMC) in a specific ratio and structure, the zinc pyridine carboxylate complex having both an electron-donating group and an electron-withdrawing group, the zinc pyridine carboxylate complex and DMC exert a synergistic effect after being composited, and have the characteristics of high selectivity and high activity in the low-temperature (30-75°C) catalytic carbon dioxide copolymerization reaction, the high activity being reflected in the catalytic activity reaching 510.97 g in 4 hours polyment / g cat The catalytic activity increased steadily or remained stable with the extension of polymerization time, and the catalytic activity could reach 535.13 g / cm2 in 12 h. polyment / g cat The selectivity is reflected in the polycarbonate content being above 98.0% and having a small molecular weight distribution, which is less than 1.8. On the basis of ensuring a high carbonate content, a high propylene oxide conversion rate can be achieved, and the conversion rate is above 95%. The polymerization reaction time is significantly shortened, so that the product has high carbonate content, high molecular weight, high conversion rate and low by-product content at the same time, which solves the problem that the prior art polypropylene carbonate cannot take into account both high conversion rate and high carbonate content, and solves the problem that the production process time is too long, and is very suitable for industrial production. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0070] Figure 1 This is the XRD pattern of the prepared ethyl-bromo-zinc pyridine carboxylate.

[0071] Figure 2 The WXRD pattern of the bimetallic catalyst prepared in the examples of this application.

[0072] Figure 3 This is the WXRD pattern of the prepared ethyl-bromo-pyridinecarboxylic acid zinc and bimetallic composite catalyst.

[0073] Figure 4The schematic flow chart of the preparation method of the zinc pyridinecarboxylate complex composite catalyst of the embodiment of the present application is shown in FIG.

[0074] Figure 5 This is a schematic flow chart of using an ultrasonic device to ultrasonically dissolve a mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent according to an embodiment of the present application.

[0075] Figure 6 This is a data flow diagram of using an ultrasonic device to ultrasonically dissolve a mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent in an embodiment of the present application. DETAILED DESCRIPTION

[0076] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0077] Accordingly, the present application proposes a zinc pyridine carboxylate complex composite catalyst, the composite catalyst comprising a zinc pyridine carboxylate complex and a double metal cyanide catalyst (DMC) of a specific ratio and structure, the zinc pyridine carboxylate complex having both an electron-donating group and an electron-withdrawing group, and the zinc pyridine carboxylate complex and DMC play a synergistic role after being composited; in the preparation process, by introducing a sensor-based data acquisition system, combined with advanced machine learning algorithms, real-time monitoring of the dissolution process is achieved, and the output of the ultrasonic power is adaptively regulated based on the actual dissolution state to ensure the reliability and stability of the dissolution process. The catalyst preparation can adaptively control the power of the ultrasonic wave based on the dynamic changes of the dissolution state during the ultrasonic dissolution process, thereby realizing a more intelligent ultrasonic dissolution process regulation, and providing a basis for the preparation of a highly selective and highly active zinc pyridine carboxylate complex composite catalyst. Therefore, the composite catalyst of the present invention has the characteristics of high selectivity and high activity in the catalytic carbon dioxide copolymerization reaction at low temperature (30-75°C), and the high activity is reflected in the catalytic activity of 510.97 g in 4 h. polyment / g catThe selectivity is reflected in the polycarbonate content being above 98.0% and having a small molecular weight distribution, which is less than 1.8. On the basis of ensuring a high carbonate content, a high propylene oxide conversion rate can be achieved, and the conversion rate is above 95%. The polymerization reaction time is significantly shortened, so that the product has high carbonate content, high molecular weight, high conversion rate and low by-product content at the same time, which solves the problem that the prior art polypropylene carbonate cannot take into account both high conversion rate and high carbonate content, and solves the problem that the production process time is too long, and is very suitable for industrial production.

[0078] Specifically, the zinc pyridine carboxylate complex composite catalyst comprises a zinc pyridine carboxylate complex and a double metal cyanide catalyst, the molar ratio of the zinc pyridine carboxylate complex to the double metal cyanide catalyst is 1:1-60, and the structural formula of the zinc pyridine carboxylate complex is shown in Formula 1:

[0079]

[0080] Formula 1,

[0081] Wherein, R1 is an electron withdrawing group selected from -CX3 (X=F), -CN, -F, -Cl, -Br, -I; R2 is an electron donating group selected from -O-, -NHR3, -NH2, -OH, -OR4, -R5, -Ph, wherein R3, R4, R5 are any alkyl groups.

[0082] Optionally, in one embodiment of the present application, the average volume particle size of the zinc pyridine carboxylate complex composite catalyst is 10 nm to 16 nm, and the specific surface area of ​​the zinc pyridine carboxylate complex composite catalyst is 16 m 2 / g~170m 2 / g.

[0083] Optionally, in one embodiment of the present application, the zinc pyridinecarboxylate complex composite catalyst has a Zn-Zn bond with a bond length of 0.01Å to 3.0Å and a Zn-Co bond with a bond length of 0.1Å to 5.0Å.

[0084] Optionally, in one embodiment of the present application, the zinc pyridinecarboxylate complex composite catalyst has an elemental composition, by mass percentage, of: Zn 10% to 30%, Co 5% to 20%, C 30% to 40%, N 4% to 20%, O 10% to 20%, and other substituent elements totaling 1% to 20%.

[0085] Optionally, in one embodiment of the present application, the structural formula of the double metal cyanide catalyst is: Zn3[Co(CN)6]2:x ZnCl2•y tBuOH·z H2O, wherein the x value ranges from 0.5 to 2.0, the y value ranges from 0.9 to 3.0, and the z value ranges from 1.0 to 3.5.

[0086] Specifically, the present application uses the following detection method to analyze the elemental composition of the composite catalyst:

[0087] 1. Elemental analysis was used to analyze the elemental composition of the composite catalyst: the elemental analysis was measured using Vario Macro Cube-M336 equipment produced by Elementar of Germany.

[0088] 2. Catalyst WRXD spectrum and chemical bond length test method: Wide-angle X-ray diffraction (WXRD) spectrum and data are obtained by measuring with Rigaku's Ultima IV.

[0089] 3. Specific surface area test method: The Quantachrome automatic gas adsorption system (AUTOSORB-1-C) of the full adsorption instrument is used for detection, and the specific surface area is calculated using the BET technology N2 adsorption curve. Before the test, the catalyst is treated at 200℃ under vacuum for 1 hour to prevent dust or water adsorbed in the pores from affecting the test results.

[0090] 4. Average volume particle size test method: The average volume particle size is measured by Nicomp 380 ZLS equipment produced by PSS, USA. Before the measurement, the solid powder is dispersed with deionized water and the suspension is broken up with ultrasound.

[0091] Example 1 Preparation of zinc pyridine carboxylate complex:

[0092] (1) Add 16.28 g zinc oxide to 50 mL dilute hydrochloric acid solution and stir at 40 °C for 1 h to form a zinc hydrochloric acid solution;

[0093] (2) Add 1.95 g of 4-ethyl-2,6-pyridinedicarboxylic acid and 2.46 g of 4-bromo-2,6-pyridinedicarboxylic acid into 60 mL of deionized water and stir evenly at room temperature to form a pyridinecarboxylic acid mixed solution;

[0094] (3) adding the prepared zinc hydrochloric acid solution to the pyridine carboxylic acid mixed solution at a rate of 10 drops / min. After the addition, the reaction solution was heated to 80°C and refluxed for 8 h. During the reflux heating process, stirring was continued at a rate of 600 r / min to ensure the formation of the pyridine carboxylic acid zinc complex.

[0095] (4) After the reaction is completed, the mixture is cooled to room temperature, crystals are gradually precipitated, and the crystals are washed and dried to obtain a zinc pyridinecarboxylate complex intermediate containing crystal water;

[0096] (5) removing crystal water from the zinc pyridine carboxylate complex intermediate at a high temperature of 130° C. to obtain the ethyl-bromo-zinc pyridine carboxylate complex, which is retained for later use.

[0097] The zinc pyridine carboxylate complex prepared in Example 1 was subjected to elemental analysis by elemental analysis. The elemental analysis results (%) of the ethyl-bromo-zinc pyridine carboxylate complex were: C, 37.79; H, 2.97; Br, 15.71; N, 5.51; O, 15.17; Zn, 12.86. The WXRD spectrum of the zinc pyridine carboxylate complex prepared in Example 1 is shown in FIG. Figure 1 As shown, for ethyl-bromo-zinc pyridine carboxylate, its characteristic peaks appear at 2θ=11.7, 13.9, and 16.8°. After the interaction between the electron-withdrawing group and the electron-donating group, the characteristic peak intensity is higher and sharper, which further proves that the electron-withdrawing group and the electron-donating group can improve the crystallinity of the zinc pyridine carboxylate particles. Combined with the ratio of Br and C in the elemental analysis, it is proved that the electron-withdrawing group and the electron-donating group react well to obtain the ethyl-bromo-zinc pyridine carboxylate complex, and its structure is shown below:

[0098]

[0099] Preparation of double metal cyanide catalyst DMC:

[0100] The double metal cyanide catalyst DMC is prepared by a traditional method, which can be specifically prepared according to the following steps: 2.4 g Na3[Co(CN)6] is dissolved in 60 mL of aqueous solution to prepare a solution, and 9.6 g ZnCl2 is dissolved in 80 mL of water and 80 mL of tert-butanol mixed solution to prepare a solution. Under strong stirring, the Na3[Co(CN)6] solution is added dropwise to the ZnCl2 solution at a drop rate of 20 drops / minute. After the addition is completed, the mixed solution is allowed to react for 3 h while maintaining strong stirring. The resulting white suspension is washed with a mixed solution of 100 mL of water and 100 mL of tert-butanol under strong stirring, and the mixed solution is filtered. Then it is placed in a vacuum oven at 100°C to dry overnight. The resulting DMC (Zn3[Co(CN)6]2: 1.8ZnCl2•2.4 t BuOH·1.5 H2O) catalyst solid powder is light yellow. The WXRD spectrum of the prepared DMC catalyst is as follows Figure 2 As shown, for the bimetallic catalyst, its characteristic peak appears at 2 =17.8,19.36°, and the structure Zn3[Co(CN)6]2: 1.8ZnCl2•2.4 t The peak positions of BuOH·1.5 H2O characteristic peaks are consistent, belonging to zinc-cobalt bimetallic catalysts.

[0101] Preparation of composite catalyst:

[0102] 100 mL of toluene and 10 mL of deionized water were added to the stirring bar, followed by 1 mol of the ethyl-bromo-pyridinecarboxylic acid zinc complex in Example 1 and DMC Zn3[Co(CN)6]2: 1.8ZnCl2•2.4 t 5 mol of BuOH·1.5 H2O was added to a high-pressure reactor, and then the high-pressure reactor was pressurized to 1 MPa under argon, followed by ultrasonic dissolution for 20 min. After the dissolution was completed, the reaction was carried out at 150 °C for 38 h.

[0103] After the reaction was completed, the mixture was cooled to room temperature, and the suspension was washed with 100 mL of acetone. After washing three times, the mixture was filtered to obtain a white solid, which was then dried in a vacuum drying oven to obtain an ethyl-bromo-zinc pyridine carboxylate / DMC composite catalyst. The WXRD spectrum of the obtained ethyl-bromo-zinc pyridine carboxylate / DMC composite catalyst is shown in FIG. Figure 3 As shown, it can be seen intuitively that the ethyl-bromo-zinc pyridine carboxylate / DMC composite catalyst system has the crystal peak characteristics of ethyl-bromo-zinc pyridine carboxylate and DMC catalyst. For the ethyl-bromo-zinc pyridine carboxylate catalyst, its characteristic peaks appear at 2θ=11.7, 13.9, 16.8°; while for the DMC catalyst, its characteristic peaks appear at 2θ=17.8, 19.36°. In the WXRD curve of the ethyl-bromo-zinc pyridine carboxylate / DMC composite catalyst, the crystal peaks of ethyl-bromo-zinc pyridine carboxylate and DMC catalyst are obvious and the peak shape is very sharp, indicating that the ethyl-bromo-zinc pyridine carboxylate and DMC catalyst have been completely composited, and the composite process of the two has no effect on the crystal morphology of the catalyst particles.

[0104] In particular, the one-time feeding method into the autoclave simplifies the operation steps, reduces the complexity of intermediate processing and the potential risk of contamination. In addition, by using a mixed solvent system containing polar solvents and non-polar solvents, the morphology and structure of the catalyst can be better controlled. Polar solvents help dissolve the complex, while non-polar solvents help control the particle size and distribution, thereby improving the performance of the catalyst.

[0105] Moreover, in the process of dissolving the mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent, an ultrasonic dissolution method is adopted to accelerate the dissolution process and promote uniform mixing between the reactants, thereby helping to form more uniform catalyst particles. Specifically, the ultrasonic wave can destroy the aggregation between particles and increase the collision frequency between particles through the cavitation effect produced, thereby accelerating the dissolution rate. In addition, the ultrasonic wave can also promote the diffusion of the solute in the solvent, so that the reactants can be more fully contacted, thereby helping to carry out the dissolution reaction. However, how to accurately control the energy input during the ultrasonic treatment process is a challenge, because too strong or too weak ultrasonic waves may affect the preparation quality of the final zinc pyridine carboxylate complex composite catalyst.

[0106] Accordingly, considering that during the dissolution process, the drastic change in temperature during the dissolution process is not conducive to the reaction system. Too high a temperature may cause the solvent to evaporate or the catalyst to deactivate, while too low a temperature may slow down the dissolution process and affect the reaction efficiency. Therefore, it is very important to monitor the real-time temperature of the solution. Moreover, the dissolution process is usually accompanied by a thermal effect, and the dissolved substance may release heat (exothermic) or absorb heat (endothermic) when entering the solvent. By monitoring the temperature change of the solution in real time, the dynamic situation of the dissolution process can be understood and the speed of dissolution can be indirectly judged. If the dissolution is rapid, the temperature may change rapidly; conversely, if the dissolution is slow, the temperature change will be relatively gentle. Monitoring the temperature change helps to evaluate the progress of the dissolution process and adjust the ultrasonic power accordingly to ensure the optimal state of the dissolution process.

[0107] In particular, the present application also provides a process for preparing a zinc pyridinecarboxylate complex composite catalyst. Figure 4 Schematic flow chart of the preparation method of the zinc pyridine carboxylate complex composite catalyst of the embodiment of the present application. Figure 4 As shown, the preparation process of the zinc pyridine carboxylate complex composite catalyst comprises the following steps: S1, adding zinc pyridine carboxylate complex, double metal cyanide catalyst and mixed solvent into a high pressure reactor at one time, wherein the mixed solvent is a mixed solvent comprising polar solvent and non-polar solvent, wherein the polar solvent is selected from one or more of water, acetonitrile, methanol, ethanol, acetone, isopropanol, acetic acid and ether, and the non-polar solvent is selected from one or more of toluene, benzene, carbon tetrachloride, hexane and dichloromethane; S2, using ultrasonic equipment to ultrasonically dissolve the mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent; S3, preparing the zinc pyridine carboxylate complex composite catalyst by reaction in an inert gas atmosphere.

[0108] Based on this, in the technical scheme of the present application, by introducing a sensor-based data acquisition system, combined with advanced machine learning algorithms, real-time monitoring of the dissolution process is achieved, and the output of the ultrasonic power is adaptively regulated based on the actual dissolution state to ensure the reliable stability of the dissolution process. Specifically, the real-time temperature data of the solution is monitored and collected in real time by a temperature sensor, and the ultrasonic power value of the ultrasonic equipment is monitored and collected in real time, and then the data processing and analysis algorithm based on artificial intelligence and machine learning is introduced in the back end to perform time series analysis and fine-grained interactive response to the real-time temperature data and ultrasonic power value of the solution, so as to decode the ultrasonic power value at the next time point, so as to automatically optimize and control the ultrasonic power. In this way, the power of the ultrasonic wave can be adaptively controlled based on the dynamic change of the dissolution state during the ultrasonic dissolution process, so as to realize more intelligent ultrasonic dissolution process regulation, and provide a basis for the preparation of highly selective and highly active zinc pyridine carboxylate complex composite catalysts.

[0109] Figure 5 This is a schematic flow chart of using an ultrasonic device to ultrasonically dissolve a mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent according to an embodiment of the present application. Figure 6 This is a data flow diagram of using an ultrasonic device to ultrasonically dissolve a mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent in an embodiment of the present application. Figure 5 and Figure 6As shown, an ultrasonic device is used to ultrasonically dissolve a mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent, including: S21, obtaining a time queue of ultrasonic power values ​​of the ultrasonic device; S22, obtaining a time queue of the real-time temperature of the solution collected by a temperature sensor; S23, inputting the time queue of the ultrasonic power values ​​and the time queue of the real-time temperature of the solution into a timing encoder based on a 1D-CNN model to obtain a sequence of local time series correlation feature vectors of ultrasonic power and a sequence of local time series correlation feature vectors of the real-time temperature of the solution; S24, inputting the sequence of local time series correlation feature vectors of ultrasonic power and the sequence of local time series correlation feature vectors of the real-time temperature of the solution into a timing encoder based on a 1D-CNN model to obtain a sequence of local time series correlation feature vectors of ultrasonic power and a sequence of local time series correlation feature vectors of the real-time temperature of the solution. The sequence of feature vectors is respectively input into the feature significance guided node time series aggregation network to obtain the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector; S25, the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are input into the feature local-global interaction response module based on feature decoupling to obtain the ultrasonic power-solution temperature local-global interaction response matrix; S26, the ultrasonic power-solution temperature local-global interaction response matrix is ​​input into the classifier-based ultrasonic power adaptive optimization module to obtain the optimization instruction, and the optimization instruction is used to indicate that the ultrasonic power value at the next time point should increase, decrease or remain unchanged.

[0110] Specifically, the S21 is to obtain the time queue of the ultrasonic power value of the ultrasonic device. It should be understood that ultrasonic power is one of the key parameters affecting the dissolution efficiency and the quality of catalyst preparation. By obtaining the change data of the ultrasonic power value over time, the power change trend in the ultrasonic dissolution process can be analyzed, and then the ultrasonic device can be adaptively adjusted to ensure the stability of the dissolution process and the preparation quality of the catalyst.

[0111] Optionally, in one embodiment of the present application, the time queue of the ultrasonic power value of the ultrasonic device is obtained in the following manner. Specifically, first, a monitoring device with a data acquisition function, such as a sensor or a data acquisition card, is connected to the ultrasonic device to monitor the ultrasonic power in real time. Secondly, the ultrasonic power value is collected at a certain time interval (for example, every second, every millisecond) by the monitoring device. These values ​​reflect the power output of the ultrasonic device at a specific time point.

[0112] Specifically, the S22, obtains the time queue of the real-time temperature of the solution collected by the temperature sensor. It should be understood that the solution recorded here is a mixture composed of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent. Optionally, in one embodiment of the present application, the time queue of the real-time temperature of the solution collected by the temperature sensor is obtained in the following manner. First, a temperature sensor, such as a thermocouple or a resistance temperature detector (RTD), is installed in the reaction vessel to measure the temperature of the solution in real time. Then, a data acquisition system (such as a data acquisition card or a data recorder) is used to connect the temperature sensor to read the temperature data regularly. Then, the specific time of each data point is recorded to form a corresponding relationship between temperature and time.

[0113] Specifically, the S23, the time queue of the ultrasonic power value and the time queue of the real-time temperature of the solution are input into the timing encoder based on the 1D-CNN model to obtain the sequence of the local timing correlation feature vector of the ultrasonic power and the sequence of the local timing correlation feature vector of the real-time temperature of the solution. It should be understood that the change of the ultrasonic power and the solution temperature over time may have an important impact on the preparation process of the catalyst. Timing coding can help identify the trend, periodicity and abnormal pattern of these parameters over time. Therefore, the time queue of the ultrasonic power value and the time queue of the real-time temperature of the solution are input into the timing encoder based on the 1D-CNN model for timing coding, so as to extract the local timing implicit correlation features of the ultrasonic power value and the real-time temperature of the solution in each local time period on the time dimension respectively, so as to obtain the sequence of the local timing correlation feature vector of the ultrasonic power and the sequence of the local timing correlation feature vector of the real-time temperature of the solution.

[0114] Specifically, in S24, the sequence of the local time series associated feature vectors of the ultrasonic power and the sequence of the local time series associated feature vectors of the real-time temperature of the solution are respectively input into the node time series aggregation network guided by the feature significance to obtain the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector. It should be understood that since the ultrasonic power and the real-time temperature of the solution have a correlation relationship between the local time series features in the time dimension, the local time series features of the ultrasonic power and the real-time temperature of the solution in different local time periods will have different contributions in the process of aggregating their respective time series information. Based on this, in order to better understand and analyze the dynamic changes of the ultrasonic power and the real-time temperature of the solution during the dissolution process, and at the same time highlight the local time series features that have the greatest impact on the dissolution process during the aggregation of their respective time series features, and provide a basis for the subsequent ultrasonic power adaptive optimization control, in the technical solution of the present application, the sequence of the local time series associated feature vectors of the ultrasonic power and the sequence of the local time series associated feature vectors of the real-time temperature of the solution are further input into the node time series aggregation network guided by the feature significance to obtain the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector.

[0115] Optionally, in one embodiment of the present application, the sequence of the ultrasonic power local time series association feature vectors and the sequence of the solution real-time temperature local time series association feature vectors are respectively input into the feature significance guided node time series aggregation network to obtain the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector, including: calculating the feature significance description factor of each ultrasonic power local time series association feature vector in the sequence of the ultrasonic power local time series association feature vector, wherein the feature significance description factor is related to the mean and variance of each ultrasonic power local time series association feature vector; based on the relationship between each ultrasonic power local time series association feature vector in the sequence of the ultrasonic power local time series association feature vector and the current The distance span between the ultrasonic power local time series correlation feature vectors is used to construct the feature significance attenuation factor of each ultrasonic power local time series correlation feature vector; based on the feature significance attenuation factor of each ultrasonic power local time series correlation feature vector, the feature significance description factor of each ultrasonic power local time series correlation feature vector is modulated to obtain a sequence of feature significance attenuation description factors; the sequence of feature significance attenuation description factors is input into a gated mask module to obtain a sequence of mask feature significance attenuation weight factors; the sequence of mask feature significance attenuation weight factors is used as a sequence of weights to calculate the weighted sum of the sequence of ultrasonic power local time series correlation feature vectors to obtain the ultrasonic power time series significant aggregation representation vector.

[0116] Optionally, in one embodiment of the present application, the characteristic significance description factor of each ultrasonic power local timing association feature vector is modulated based on the characteristic significance attenuation factor of each ultrasonic power local timing association feature vector to obtain a sequence of characteristic significance attenuation description factors, including: calculating the positional product between the characteristic significance attenuation factor of each ultrasonic power local timing association feature vector and the characteristic significance description factor of each ultrasonic power local timing association feature vector to obtain a sequence of characteristic significance attenuation description factors.

[0117] Optionally, in one embodiment of the present application, a feature significance description factor of each ultrasonic power local time series association feature vector in the sequence of the ultrasonic power local time series association feature vector is calculated, wherein the feature significance description factor is related to the mean and variance of each ultrasonic power local time series association feature vector, including: respectively calculating the mean and variance of the ultrasonic power local time series association feature vector to obtain the ultrasonic power local time series association feature mean and the ultrasonic power local time series association feature variance; calculating the fourth power of the positional difference between the ultrasonic power local time series association feature vector and the ultrasonic power local time series association feature mean, and then calculating the expected value of the fourth power modulated semantic deviation vector to obtain the ultrasonic power local time series association feature expectation factor; calculating the positional division between the ultrasonic power local time series association feature expectation factor and the square of the ultrasonic power local time series association feature variance to obtain the feature significance description factor of the ultrasonic power local time series association feature vector.

[0118] Optionally, in one embodiment of the present application, based on the distance span between each ultrasonic power local timing association feature vector in the sequence of the ultrasonic power local timing association feature vector and the current ultrasonic power local timing association feature vector, a characteristic significance attenuation factor of each ultrasonic power local timing association feature vector is constructed, including: calculating the distance span between other ultrasonic power local timing association feature vectors in the sequence of the ultrasonic power local timing association feature vector and the current ultrasonic power local timing association feature vector to obtain a characteristic value distance span; extracting the maximum value in the current ultrasonic power local timing association feature vector to obtain the current ultrasonic power local timing association feature maximum value; extracting the maximum value in the other ultrasonic power local timing association feature vectors to obtain other ultrasonic power local timing association feature maximum values; calculating the difference between the current ultrasonic power local timing association feature maximum value and the other ultrasonic power local timing association feature maximum values ​​and dividing the difference by the characteristic value distance span to obtain the characteristic significance attenuation factor.

[0119] Optionally, in one embodiment of the present application, the sequence of feature significance attenuation descriptors is input into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors, including: taking the negative number of each feature significance attenuation descriptor in the sequence of feature significance attenuation descriptors as an exponential power, calculating an exponential function value with the natural constant e as the base to obtain a sequence of feature significance attenuation class support descriptors; calculating the reciprocal of the sum of each feature significance attenuation class support descriptor in the sequence of feature significance attenuation class support descriptors and a constant one to obtain a sequence of feature significance attenuation weight factors; based on a predetermined threshold, masking the sequence of feature significance attenuation weight factors to obtain a sequence of masked feature significance attenuation weight factors, wherein, in response to each feature significance attenuation weight factor in the sequence of feature significance attenuation weight factors being greater than the predetermined threshold, the feature significance attenuation weight factor greater than the predetermined threshold is taken to have the original value, otherwise it is set to zero.

[0120] Optionally, in one embodiment of the present application, the sequence of the mask feature significance attenuation weight factors is used as a sequence of weights, and the weighted sum of the sequence of the ultrasonic power local timing association feature vectors is calculated to obtain the ultrasonic power timing significant aggregation representation vector, including: using the sequence of the mask feature significance attenuation weight factors as a sequence of weights, calculating the weighted sum between each other ultrasonic power local timing association feature vector in the sequence of the ultrasonic power local timing association feature vector to obtain a historical ultrasonic power local timing association feature vector; calculating the positional sum between the historical ultrasonic power local timing association feature vector and the current ultrasonic power local timing association feature vector to obtain the ultrasonic power timing significant aggregation representation vector.

[0121] Specifically, the sequence of the ultrasonic power local time series associated feature vectors is input into the feature significance guided node time series aggregation network and processed by the following feature aggregation formula to obtain the ultrasonic power time series significant aggregation representation vector;

[0122] Among them, the feature aggregation formula is:

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129]

[0130] in, is the sequence of local temporal correlation feature vectors of the ultrasonic power, is the first in the sequence of the local temporal correlation feature vector of the ultrasonic power The local temporal correlation feature vector of ultrasonic power, and are respectively the first and The local temporal correlation feature vector of ultrasonic power, Indicates taking The maximum value in Indicates taking The maximum value in express and The eigenvalue distance span between For the The characteristic significance attenuation factor corresponding to the local time series correlation eigenvector of ultrasonic power, For the The mean of the local temporal correlation eigenvectors of ultrasonic power, For the The variance of the local temporal correlation eigenvector of the ultrasonic power, For the The first The eigenvalues ​​at the positions, represents the computational expectation, For the The characteristic significance attenuation description factor corresponding to the local temporal correlation eigenvector of ultrasonic power, For the The characteristic significance attenuation weight factor of the local temporal correlation characteristic vector of ultrasound power, For masking operation, is the predetermined threshold, For the The mask feature significance attenuation weight factor of the local temporal correlation feature vector of ultrasound power, is the number of feature vectors in the sequence of the ultrasonic power local time series associated feature vectors, A significant aggregation representation vector for the ultrasound power time series.

[0131] Specifically, for the process of aggregating the time series information of the sequence of the local time series associated feature vectors of the real-time temperature of the solution, first, the node time series aggregation network guided by the feature significance will perform a significance analysis on the sequence of the local time series associated feature vectors of the real-time temperature of the solution, and determine its representativeness and discrimination in a statistical sense by calculating the mean and variance of each local time series associated feature vector of the real-time temperature of the solution. The mean reflects the central trend of the local time series characteristics of the real-time temperature of the solution, and the variance reflects the degree of fluctuation of the characteristic value, and the two jointly determine the importance of the local time series characteristics of the real-time temperature of the solution. According to the above-mentioned significance analysis results, the model generates the feature significance description factor of each node of the local time series associated feature vector of the real-time temperature of the solution, and these factors provide a basis for the subsequent feature selection. By identifying which local time series characteristics of the real-time temperature of the solution are the most significant, the performance of the real-time temperature of the solution at different time points can be better understood. Next, the model constructs a feature significance attenuation factor, which adjusts the influence of each local time series associated feature vector of the real-time temperature of the solution by calculating the distance between the local time series associated feature vectors of the real-time temperature of the solution. The local time series features of the real-time temperature of the solution that are closer to the local time series associated feature vector of the real-time temperature of the current node will obtain higher weights, which means that they occupy a more important position in the correlation of local features. Then, by applying the significance attenuation factor to the significance descriptor, the modulated descriptor sequence more accurately reflects the importance of each local time series feature of the real-time temperature of the solution. Subsequently, through the gated mask module, the model further screens and optimizes these features, generates a sequence of mask feature significance attenuation weight factors, and realizes the dynamic adjustment of the weights of each local time series feature of the real-time temperature of the solution. Finally, through the weighted summation method, the model generates a significant aggregation representation vector of the real-time temperature of the solution. These vectors integrate the feature information of all local time series nodes of the real-time temperature of the solution in the sequence of the local time series associated feature vector of the real-time temperature of the solution, and highlight the most significant features while suppressing unimportant features. In this way, it can be ensured that the extracted real-time temperature of the solution time series aggregation feature representation can reflect the dynamic changes of the real-time temperature of the solution during the dissolution process, and highlight the features that have the greatest impact on the dissolution process. This is crucial for the subsequent ultrasonic power adaptive optimization control, because it can help the system better understand and predict the behavior of the dissolution process and make optimal power adjustment decisions accordingly, thereby ensuring that the ultrasonic dissolution process is in the best state.

[0132] Specifically, in S25, the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are input into the feature local-global interactive response module based on feature decoupling to obtain the ultrasonic power-solution temperature local-global interactive response matrix. It should be understood that since the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector respectively contain the time series significant aggregation representation information of each local time series feature of the ultrasonic power and the solution real-time temperature at the time node, they are used to respectively reflect the ultrasonic intensity and dissolution rate time series feature information in the ultrasonic dissolution process. And since there is a time series response correlation relationship and complex interactive influence between the ultrasonic intensity and the dissolution rate. Therefore, in order to explore and characterize the feature response interactive information and correlation relationship between the ultrasonic intensity time series aggregation feature and the dissolution rate time series aggregation feature, in the technical solution of the present application, the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are further input into the feature local-global interactive response module based on feature decoupling to obtain the ultrasonic power-solution temperature local-global interactive response matrix.

[0133] Optionally, in one embodiment of the present application, the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are input into a feature local-global interactive response module based on feature decoupling to obtain an ultrasonic power-solution temperature local-global interactive response matrix, including: feature decoupling the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector to obtain a sequence of ultrasonic power local time series semantic representation vectors and a sequence of solution real-time temperature local time series semantic representation vectors; calculating the local semantic association interaction vector between each group of corresponding ultrasonic power local time series semantic representation vectors and the solution real-time temperature local time series semantic representation vector in the sequence of the ultrasonic power local time series semantic representation vector and the sequence of the solution real-time temperature local time series semantic representation vector to obtain a sequence of ultrasonic power-solution temperature local time series semantic association interaction vectors; calculating each ultrasonic power-solution temperature local time series semantic association interaction vector in the sequence of the ultrasonic power-solution temperature local time series semantic association interaction vectors. The method comprises the following steps: performing weighted optimization on the sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights; inputting the sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights into a random inactivation module to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights after inactivation; taking the sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights after inactivation as weight distribution, weightedly optimizing the sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors to obtain an optimized sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors; matrixing the optimized sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors to obtain an ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix; and performing autocorrelation interaction perception analysis on the ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix to obtain the ultrasonic power-solution temperature local-global interaction response matrix.

[0134] Optionally, in one embodiment of the present application, the fine-grained attention weights of each ultrasonic power-solution temperature local temporal semantic association interaction vector in the sequence of the ultrasonic power-solution temperature local temporal semantic association interaction vectors are calculated to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights, including: matrix multiplying the ultrasonic power-solution temperature local temporal semantic association interaction vector and the weight matrix, and then adding them positionally with the bias vector to obtain a linearly transformed ultrasonic power-solution temperature local temporal semantic association interaction vector; using a tanh function to activate the linearly transformed ultrasonic power-solution temperature local temporal semantic association interaction vector, and then vector multiplying it with the weight coefficient vector to obtain the ultrasonic power-solution temperature local temporal fine-grained attention weight.

[0135] Optionally, in one embodiment of the present application, the local semantic association interaction vector between each group of corresponding ultrasonic power local temporal semantic representation vectors and the local temporal semantic representation vector of the real-time temperature of the solution in the sequence of the ultrasonic power local temporal semantic representation vector and the sequence of the local temporal semantic representation vector of the real-time temperature of the solution is calculated to obtain a sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors, including: performing positional difference, positional dot multiplication and positional addition processing on each group of corresponding ultrasonic power local temporal semantic representation vectors and the local temporal semantic representation vector of the real-time temperature of the solution to obtain an ultrasonic power-solution temperature local temporal semantic difference table. The ultrasonic power-solution temperature local temporal semantics difference representation vector, the ultrasonic power-solution temperature local temporal semantics dot product representation vector and the ultrasonic power-solution temperature local temporal semantics sum representation vector are cascaded to obtain an ultrasonic power-solution temperature local temporal semantics association multidimensional representation fusion vector; the ultrasonic power-solution temperature local temporal semantics association multidimensional representation fusion vector is one-dimensionally convolutionally encoded and then passed through a maximum pooling layer to obtain an ultrasonic power-solution temperature local temporal semantics association interaction vector.

[0136] Optionally, in one embodiment of the present application, an autocorrelation interaction perception analysis is performed on the ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix to obtain the ultrasonic power-solution temperature local-global interaction response matrix, including: performing matrix multiplication on the ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix and the transposed matrix of the ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix, and then dividing the matrix by the scale of the ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix by position to obtain an ultrasonic power-solution temperature temporal fine-grained interaction feature global dependency matrix; using a Relu function to activate the ultrasonic power-solution temperature temporal fine-grained interaction feature global dependency matrix to obtain the ultrasonic power-solution temperature local-global interaction response matrix.

[0137] Specifically, the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are input into the feature local-global interactive response module based on feature decoupling for processing using the following feature decoupling interactive response formula to obtain the ultrasonic power-solution temperature local-global interactive response matrix;

[0138] Among them, the feature decoupling interaction response formula is:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147] in, is the significant aggregation representation vector of the ultrasonic power time series, is the significant aggregation representation vector of the solution real-time temperature time series, For feature decoupling operations, are the first, second, and third vectors in the sequence of the local temporal semantic representation vector of ultrasonic power. , The local temporal semantic representation vector of ultrasound power, are the first, second, and third elements in the sequence of the local temporal semantic representation vector of the real-time temperature of the solution. , The local temporal semantic representation vector of the real-time temperature of the solution, For cascade operation, is a one-dimensional convolution operation, For positional subtraction, For point-by-position multiplication, To add by position, For maximum pooling processing, is the local temporal semantic association interaction vector of ultrasonic power and solution temperature, and are the weight matrix and bias vector respectively, for function, is the weight coefficient vector, is the fine-grained attention weight of the local temporal relationship between ultrasonic power and solution temperature, For inactivation treatment, is the fine-grained attention weight of the local temporal relationship between ultrasonic power and solution temperature after inactivation treatment, is the first in the sequence of optimized ultrasonic power-solution temperature local temporal semantic association interaction vectors An optimized ultrasound power-solution temperature local temporal semantic association interaction vector, represents a matrix arrangement of a vector sequence, is the global semantic representation matrix of fine-grained interaction between ultrasonic power and solution temperature, is the scale of the global semantic representation matrix of the fine-grained interaction between ultrasound power and solution temperature, for Activation function, is the ultrasonic power-solution temperature local-global interaction response matrix.

[0148] Specifically, first, the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are feature decoupled. This process decomposes the original ultrasonic power and solution real-time temperature time series significant aggregation representation vectors into multiple local time series semantic representation vectors, so that the model can independently analyze and process the details of each local time series feature of the ultrasonic power and solution real-time temperature in the time dimension, so as to more carefully analyze the fine-grained time series correlation characteristics between the ultrasonic power and the solution real-time temperature in different local time periods. This decoupling can enhance the model's ability to capture feature dynamics and help better understand and express the intrinsic properties and interactive semantics between the real-time data of the two. Then, the sequence of ultrasonic power-solution temperature local time series semantic association interaction vectors between the sequence of decoupled ultrasonic power local time series semantic representation vectors and the sequence of solution real-time temperature local time series semantic representation vectors is calculated. The ultrasonic power-solution temperature local time series semantic association interaction vector reveals the potential correlation between the ultrasonic power local time series feature semantics and the solution real-time temperature local time series feature semantics, and provides a quantitative perspective for understanding how the two features interact at the local level, which is crucial for capturing the local pattern and structure of the interactive information between ultrasonic power and solution real-time temperature. Further, a fine-grained attention weight is calculated for each ultrasonic power-solution temperature local time series semantic association interaction vector, and the fine-grained attention weight highlights the key elements in this interaction vector, namely, the ultrasonic power and solution real-time temperature local time series semantic association features, so that the model can focus on the most important feature interaction information. It should be understood that by giving more important elements a greater weight, the fine-grained attention weight enables the model to pay more attention to the most critical feature parts for the ultrasonic power optimization control task, thereby improving the quality of feature representation and the prediction accuracy of the model. Further, the sequence of the ultrasonic power-solution temperature local time series fine-grained attention weights is input into a random dropout module to achieve weight sparsification and enhance the generalization ability of the model. The random inactivation process reduces the risk of overfitting of the model by randomly "turning off" a part of the weights, while retaining the most important local temporal feature interaction information between ultrasonic power and real-time temperature of the solution. This enables the model to generalize better when facing new data and adapt to different environmental conditions of the ultrasonic dissolution process. Then, the sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights after inactivation is used as the weight distribution to weight-optimize the sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors, and the optimized sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors is matrixed to obtain the ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix.The ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix integrates the local interaction information and attention weights of ultrasonic power and solution real-time temperature and the temporal interaction representation information. It integrates the interaction information and attention weights between each ultrasonic power-solution temperature local temporal semantic feature, so that the model can understand the relationship between features more comprehensively. Finally, the ultrasonic power-solution temperature temporal fine-grained interaction global semantic representation matrix is ​​subjected to autocorrelation interaction perception analysis to obtain the ultrasonic power-solution temperature local-global interaction response matrix. Through the autocorrelation interaction perception analysis, the global fine-grained interaction semantics between ultrasonic power and solution real-time temperature can be captured by autocorrelation features, so as to use the global perspective to more comprehensively perceive the temporal interaction and responsiveness relationship between ultrasonic power and solution temperature. In this way, the local temporal fine-grained interaction and global semantic interaction response relationship between ultrasonic intensity and ultrasonic dissolution rate can be used to improve the accuracy and reliability of subsequent ultrasonic power optimization control tasks.

[0149] Specifically, the S26, the ultrasonic power-solution temperature local-global interactive response matrix is ​​input into the ultrasonic power adaptive optimization module based on the classifier to obtain an optimization instruction, and the optimization instruction is used to indicate that the ultrasonic power value at the next time point should increase, decrease or remain unchanged. In other words, the ultrasonic power and the fusion temperature are used to perform classification processing, so as to decode the ultrasonic power value at the next time point, thereby automatically optimizing the ultrasonic power. In this way, the power of the ultrasonic wave can be adaptively controlled based on the dynamic change of the dissolved state during the ultrasonic dissolution process, so as to realize more intelligent ultrasonic dissolution process regulation, and provide a basis for the preparation of highly selective and highly active zinc pyridine carboxylate complex composite catalysts.

[0150] In particular, in the technical solution of the present application, the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector respectively represent the time series significant aggregation features of the local time series implicit correlation features of the ultrasonic power determined based on 1D convolutional coding and the time series significant aggregation features of the local time series implicit correlation features of the solution real-time temperature determined based on 1D convolutional coding. It should be understood that in the technical solution of the present application, the temperature time series correlation pattern feature of the solution real-time temperature represents the rate of boring ultrasonic dissolution in the high-dimensional feature domain, because the material will generate heat during ultrasonic dissolution, so there is a high-dimensional implicit correlation between heat and ultrasonic dissolution rate. However, considering that the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector have local time series features and aggregation features of local time series features based on the entire time domain, therefore, when the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are input into the feature local-global interaction response based on feature decoupling, due to the granularity difference of the feature receptive field during the response interaction and the feature modal difference, the ultrasonic power-solution temperature local-global interaction response matrix will have a diversified set expression distribution of multi-modal temporal semantic features. Therefore, it is expected to improve the balance between the regression mapping accuracy and completeness of the ultrasonic power-solution temperature local-global interaction response matrix when performing class regression through the classifier-based ultrasonic power adaptive optimization module, thereby improving the accuracy of the obtained optimization instructions.

[0151] In a preferred example, the ultrasonic power-solution temperature local-global interaction response matrix is ​​passed through a classifier-based ultrasonic power adaptive optimization module to obtain optimization instructions, including:

[0152] Determine the maximum eigenvalue of the ultrasonic power-solution temperature local-global interaction response and the minimum eigenvalue of the ultrasonic power-solution temperature local-global interaction response of the ultrasonic power-solution temperature local-global interaction response matrix, and calculate the ultrasonic power-solution temperature local-global interaction response mean and ultrasonic power-solution temperature local-global interaction response standard deviation of the feature set of the ultrasonic power-solution temperature local-global interaction response matrix;

[0153] Calculating the quotient of the ultrasonic power-solution temperature local-global interaction response mean and the ultrasonic power-solution temperature local-global interaction response standard deviation to obtain a statistical normalized value of the ultrasonic power-solution temperature local-global interaction response;

[0154] Calculating the inverse of each eigenvalue of the ultrasonic power-solution temperature local-global interaction response matrix, multiplying it by the difference between the maximum eigenvalue of the ultrasonic power-solution temperature local-global interaction response and the minimum eigenvalue of the ultrasonic power-solution temperature local-global interaction response, and then subtracting it from the ultrasonic power-solution temperature local-global interaction response statistical normalization value to obtain an ultrasonic power-solution temperature local-global interaction response distribution approximation matrix;

[0155] Calculate an exponential function with natural constants as the base, where each eigenvalue of the ultrasonic power-solution temperature local-global interaction response distribution approximation matrix is ​​an exponential function, and perform point addition with the ultrasonic power-solution temperature local-global interaction response statistical normalization value to obtain an ultrasonic power-solution temperature local-global interaction response probability approximation matrix;

[0156] Calculating the base-two logarithm of the absolute value of each eigenvalue of the ultrasonic power-solution temperature local-global interactive response probability approximation matrix to obtain an optimized ultrasonic power-solution temperature local-global interactive response matrix;

[0157] The optimized ultrasonic power-solution temperature local-global interaction response matrix is ​​passed through the classifier-based ultrasonic power adaptive optimization module to obtain the optimization instruction.

[0158] Among them, the local-global interaction response matrix of ultrasonic power-solution temperature is recorded as The optimization expression is:

[0159]

[0160] in, represents the ultrasonic power-solution temperature local-global interaction response matrix, , , and are the characteristic sets of the ultrasonic power-solution temperature local-global interaction response matrix The mean, standard deviation, minimum and maximum values ​​of It means point multiplication by position. It means adding by position. It means subtracting by position. Indicates the calculation of the inverse of each eigenvalue of the feature map, Represents the optimized ultrasound power-solution temperature local-global interaction response matrix.

[0161] Therefore, by performing random statistical normalization on the diversified feature set of the ultrasonic power-solution temperature local-global interaction response matrix, the confidence space constructed based on the overall eigenvalue of the ultrasonic power-solution temperature local-global interaction response matrix is ​​approximated by the standardized continuous probability density distribution relative to the response hypothesis test of each eigenvalue of the ultrasonic power-solution temperature local-global interaction response matrix, thereby establishing the target reachability of the discretized feature distribution of the ultrasonic power-solution temperature local-global interaction response matrix to the unified regression target, so as to achieve the balanced executableness between mapping accuracy and mapping completeness in the class regression process based on the discretized feature distribution of the ultrasonic power-solution temperature local-global interaction response matrix, and improve the accuracy of the optimization instructions obtained by the ultrasonic power-solution temperature local-global interaction response matrix through the classifier-based ultrasonic power adaptive optimization module. In this way, the power of the ultrasonic wave can be more accurately and adaptively controlled based on the dynamic change of the dissolution state during the ultrasonic dissolution process, thereby realizing a more intelligent ultrasonic dissolution process regulation, and providing a basis for the preparation of highly selective and highly active zinc pyridine carboxylate complex composite catalysts.

[0162] In particular, the dissolution time of the composition using a fixed ultrasonic power and the effect of the ultrasonic dissolution strategy of the present application are compared in the following table:

[0163] Table 1 Process parameters for preparing composite catalysts in Schemes 1-4 and Comparative Schemes 1-4

[0164]

[0165] It can be seen from the data in Table 1 that no matter how the formula components of the zinc pyridinecarboxylate complex composite catalyst are, the dissolution time is optimized by the adaptive ultrasonic power control strategy provided by the present application relative to the fixed power control strategy, wherein the dissolution time of Scheme 1 is 8 minutes, the dissolution time of Scheme 1 is 20 minutes, the dissolution time of Scheme 2 is 10 minutes, the dissolution time of Scheme 2 is 20 minutes, the dissolution time of Scheme 3 is 16 minutes, the dissolution time of Scheme 3 is 22 minutes, the dissolution time of Scheme 4 is 15 minutes, and the dissolution time of Scheme 4 is 21 minutes; and the specific surface area of ​​the prepared composite catalyst is relatively increased, and the average volume particle size is reduced, indicating that the composite catalyst prepared by the adaptive ultrasonic power control strategy provided by the present application has a smaller particle size and a larger specific surface area, and the catalyst activity and selectivity are improved relative to the fixed power control strategy.

[0166] Application examples and comparative application examples

[0167] The composite catalysts prepared by the above schemes 1-4 and comparative schemes 1-4 were used to prepare propylene oxide-carbon dioxide copolymers according to the following synthesis method, which were respectively recorded as application examples 1-4 and comparative application examples 1-4. The specific parameters are described in Table 2 below.

[0168] A rotor and 0.3 g of a composite catalyst (the composite catalyst is derived from the corresponding embodiment and the corresponding comparative example) were added to a 100 mL high-pressure reactor, dried at 40°C for 3 h under vacuum conditions, cooled to room temperature, 30 mL of purified propylene oxide was added, and carbon dioxide was charged to make the pressure in the reactor reach 4.0 MPa. The reaction was continued at a relatively low temperature of 30 to 75°C and a stirring speed of 600 rpm. After the reaction was completed, the reactor was removed, cooled to room temperature, unreacted carbon dioxide was discharged, the product in the reactor was taken out, unreacted monomers were removed by vacuum, and the obtained product was washed with a mixed solution of dichloromethane and methanol, and dried to obtain a copolymer of carbon dioxide and propylene oxide. The conversion rate of propylene oxide and the catalytic activity of the catalyst can be calculated by weighing and testing.

[0169] Specific method and calculation formula for propylene oxide conversion rate: propylene oxide conversion rate % = copolymer mass * 0.569 / propylene oxide mass * 100%

[0170] Carbonate group content detection and calculation method: The molar fraction of carbonate structural units in epoxy compound-carbon dioxide copolymer resin is determined by nuclear magnetic resonance hydrogen spectrum 1 H NMR, according to the analytical method described in CN111349224A

[0057] -

[0068] . According to the method, 1 The molar fraction of the carbonate group content was determined by H-NMR (Bruker, DPX400, 400 MHz; pulse program zg30, waiting time d1:10 s, 64 scans). In each case, the sample was dissolved in deuterated chloroform. 1 The relevant resonances in H-NMR (based on TMS=0 ppm) are as follows: 5.0 ppm and 4.2 ppm belong to the proton peaks of the methine and methylene groups on the polycarbonate group, 4.9 ppm, 4.5 ppm and 4.1 ppm belong to the proton peaks of the methine and methylene groups in the five-membered ring carbonate, and 3.5-3.8 ppm belong to the proton peaks of the ether group. The capital letter A plus a numerical subscript represents the integrated area of ​​the peak at a certain ppm on the nuclear magnetic hydrogen spectrum. A is the abbreviation of Area in English, for example, A 5.0 represents the integrated area of ​​the peak at 5.0 ppm. 1 The integrated area of ​​the H-NMR spectrum and its related proton peaks defines the ratio (molar ratio) of carbonate groups in the copolymerization reaction (FCO2 ) is calculated as follows:

[0171] F CO2 =(A 5.0 +A 4.2 -2×A 4.5 ) / [(A 5.0 +A 4.2 -2×A 4.5 )+A 3.5 ]×100%.

[0172] By-product content detection and calculation method: The by-product is a cyclic small molecule propylene carbonate, and its weight is expressed by Wcc%. In the carbonate group content detection method, the chemical environment corresponding to each peak is known, so Wcc%=102A 4.5 / [102(A 5.0 +A 4.2 -2A 4.5 +A 1.5 )+58A 3.5 ], where 102 is the molecular weight of propylene carbonate and polycarbonate groups, and 58 is the chemical molecular weight of the polyether structure.

[0173] Number average molecular weight detection method and molecular weight distribution detection method: measured by GPC, the specific test conditions are: gel chromatography column is PLgelMixed-C column (300mm×4.5mm, 5μm); column temperature is 40℃; mobile phase is tetrahydrofuran, flow rate is 1mL / min; injection volume is 100μL; analysis time is 20min. The number average molecular weight and molecular weight distribution data can be obtained.

[0174] Table 2 The present invention scheme 1-4 and the comparative scheme 1-4

[0175] Process parameters and performance parameters of composite catalyst for preparing propylene oxide-carbon dioxide copolymer

[0176]

[0177] It can be seen from the data of the application examples and comparative application examples in Table 2 that no matter how the formula components of the zinc pyridinecarboxylate complex composite catalyst are, the performance of the polymerization product obtained by using the adaptive ultrasonic power control strategy provided by the present application for the copolymerization reaction of carbon dioxide and propylene oxide is better than that of the fixed power control strategy. It can be seen from the performance of the above composite catalyst that a smaller average volume particle size and a larger specific surface area are beneficial to the activity and selectivity of the catalyst, and the catalytic activity can reach 510.97 g / cm2 for 4 h. polyment / g catIt can achieve extremely high propylene oxide conversion rate in the low temperature (30-75°C) catalytic carbon dioxide copolymerization reaction, significantly shorten the polymerization reaction time, and make the product have high carbonate content, high molecular weight, high conversion rate and low by-product content, which is suitable for industrial production.

[0178] In summary, in the technical solution of the present application, by introducing a sensor-based data acquisition system and combining it with an advanced machine learning algorithm, real-time monitoring of the dissolution process is achieved, and the output of the ultrasonic power is adaptively regulated based on the actual dissolution state to ensure the reliability and stability of the dissolution process. The catalyst preparation can adaptively control the power of the ultrasonic wave based on the dynamic changes of the dissolution state during the ultrasonic dissolution process, thereby realizing a more intelligent regulation of the ultrasonic dissolution process, and providing a basis for the preparation of a highly selective and highly active zinc pyridine carboxylate complex composite catalyst. A method for preparing a zinc pyridine carboxylate complex composite catalyst is provided, wherein the composite catalyst comprises a zinc pyridine carboxylate complex and a double metal cyanide catalyst (DMC) of a specific ratio and structure, wherein the zinc pyridine carboxylate complex has both an electron-donating group and an electron-withdrawing group, and the zinc pyridine carboxylate complex and DMC play a synergistic role after being composited, and have the characteristics of high selectivity and high activity in the catalytic carbon dioxide copolymerization reaction at low temperature (30-75°C), wherein the high activity is reflected in the catalytic activity reaching 510.97 g in 4 h. polyment / g cat The selectivity is reflected in the polycarbonate content being above 98.0% and having a small molecular weight distribution, which is less than 1.8. On the basis of ensuring a high carbonate content, a high propylene oxide conversion rate can be achieved, and the conversion rate is above 95%. The polymerization reaction time is significantly shortened, so that the product has high carbonate content, high molecular weight, high conversion rate and low by-product content at the same time, which solves the problem that the prior art polypropylene carbonate cannot take into account both high conversion rate and high carbonate content, and solves the problem that the production process time is too long, and is very suitable for industrial production.

[0179] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A zinc pyridine carboxylate complex composite catalyst, characterized in that: The structural formula of the zinc pyridine carboxylate complex is shown in Formula 1: Formula 1, Wherein, R1 is an electron withdrawing group selected from -CF3, -CN, -F, -Cl, -Br, -I; R2 is an electron donating group selected from -O-, -NHR3, -NH2, -OH, -OR4, -R5, -Ph, wherein R3, R4, R5 are any alkyl groups; The method for preparing the zinc pyridinecarboxylate complex composite catalyst comprises the following steps: Adding a zinc pyridine carboxylate complex, a double metal cyanide catalyst and a mixed solvent into a high-pressure reactor at one time, wherein the mixed solvent is a mixed solvent comprising a polar solvent and a non-polar solvent, wherein the polar solvent is selected from one or more of water, acetonitrile, methanol, ethanol, acetone, isopropanol, acetic acid, and ether, and the non-polar solvent is selected from one or more of toluene, benzene, carbon tetrachloride, hexane, and dichloromethane; Using an ultrasonic device to ultrasonically dissolve the mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent; The zinc pyridinecarboxylate complex composite catalyst is prepared by reacting in an inert gas atmosphere; The zinc pyridine carboxylate complex composite catalyst has the following elemental composition by mass percentage: Zn 10% to 30%, Co 5% to 20%, C 30% to 40%, N 4% to 20%, O 10% to 20%, and other substituent elements 1% to 20% in total; The method of using an ultrasonic device to ultrasonically dissolve the mixture consisting of the zinc pyridine carboxylate complex, the double metal cyanide catalyst and the mixed solvent comprises: Obtaining a time queue of ultrasonic power values ​​of the ultrasonic device; A time queue for obtaining the real-time temperature of the solution collected by the temperature sensor; A coordinated response analysis is performed on the time queue of the ultrasonic power value and the time queue of the real-time temperature of the solution to obtain an optimization instruction, wherein the optimization instruction is used to indicate whether the ultrasonic power value at the next time point should increase, decrease, or remain unchanged.

2. The zinc pyridinecarboxylate complex composite catalyst according to claim 1, characterized in that: The average volume particle size of the zinc pyridine carboxylate complex composite catalyst is 10 nm to 16 nm, and the specific surface area of ​​the zinc pyridine carboxylate complex composite catalyst is 16 m 2 / g~170 m 2 / g.

3. The zinc pyridinecarboxylate complex composite catalyst according to claim 1, characterized in that: The zinc pyridinecarboxylate complex composite catalyst has a Zn-Zn bond with a bond length of 0.01Å to 3.0Å and a Zn-Co bond with a bond length of 0.1Å to 5.0Å.

4. The zinc pyridinecarboxylate complex composite catalyst according to claim 1, characterized in that: The reaction temperature of the inert gas atmosphere reaction is 20°C-150°C, and the reaction time is 12 h-38 h; the pressure of the inert gas atmosphere is 0.2-1 MPa.

5. The zinc pyridinecarboxylate complex composite catalyst according to claim 1, characterized in that: Performing a coordinated response analysis on the time queue of the ultrasonic power value and the time queue of the real-time temperature of the solution to obtain an optimization instruction includes: Inputting the time queue of the ultrasonic power value and the time queue of the real-time temperature of the solution into a temporal encoder based on a 1D-CNN model to obtain a sequence of local temporal correlation feature vectors of the ultrasonic power and a sequence of local temporal correlation feature vectors of the real-time temperature of the solution; Inputting the sequence of the local time series associated feature vectors of the ultrasonic power and the sequence of the local time series associated feature vectors of the real-time temperature of the solution into the node time series aggregation network guided by feature significance respectively to obtain the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector; Inputting the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector into a feature local-global interactive response module based on feature decoupling to obtain an ultrasonic power-solution temperature local-global interactive response matrix; The ultrasonic power-solution temperature local-global interaction response matrix is ​​input into a classifier-based ultrasonic power adaptive optimization module to obtain an optimization instruction, wherein the optimization instruction is used to indicate whether the ultrasonic power value at the next time point should increase, decrease, or remain unchanged.

6. The zinc pyridinecarboxylate complex composite catalyst according to claim 5, characterized in that: The sequence of the local time series associated feature vectors of the ultrasonic power and the sequence of the local time series associated feature vectors of the solution real-time temperature are respectively input into the node time series aggregation network guided by feature significance to obtain the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector, including: Calculating a characteristic significance description factor of each ultrasonic power local time series correlation feature vector in the sequence of the ultrasonic power local time series correlation feature vectors, wherein the characteristic significance description factor is related to the mean and variance of each ultrasonic power local time series correlation feature vector; Constructing a characteristic significance attenuation factor of each ultrasonic power local time series association feature vector based on the distance span between each ultrasonic power local time series association feature vector in the sequence of ultrasonic power local time series association feature vectors and the current ultrasonic power local time series association feature vector; Modulating the characteristic significance description factors of the respective ultrasonic power local time series associated characteristic vectors based on the characteristic significance attenuation factors of the respective ultrasonic power local time series associated characteristic vectors to obtain a sequence of characteristic significance attenuation description factors; Inputting the sequence of feature significance attenuation description factors into a gated mask module to obtain a sequence of masked feature significance attenuation weight factors; The sequence of the mask feature significance attenuation weight factors is used as the sequence of weights, and the weighted sum of the sequence of the ultrasound power local time series associated feature vectors is calculated to obtain the ultrasound power time series significant aggregation representation vector.

7. The zinc pyridinecarboxylate complex composite catalyst according to claim 6, characterized in that: The ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector are input into a feature local-global interactive response module based on feature decoupling to obtain an ultrasonic power-solution temperature local-global interactive response matrix, including: Performing feature decoupling on the ultrasonic power time series significant aggregation representation vector and the solution real-time temperature time series significant aggregation representation vector to obtain a sequence of ultrasonic power local time series semantic representation vectors and a sequence of solution real-time temperature local time series semantic representation vectors; Calculating the local semantic association interaction vectors between each corresponding group of the sequence of the ultrasonic power local temporal semantic representation vectors and the sequence of the solution real-time temperature local temporal semantic representation vectors to obtain a sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors; Calculating the fine-grained attention weight of each ultrasonic power-solution temperature local temporal semantic association interaction vector in the sequence of ultrasonic power-solution temperature local temporal semantic association interaction vectors to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights; Inputting the sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights into a random deactivation module to obtain a sequence of ultrasonic power-solution temperature local temporal fine-grained attention weights after deactivation processing; Taking the sequence of fine-grained attention weights of the local temporal sequence of ultrasonic power-solution temperature after the inactivation process as the weight distribution, weighted optimization is performed on the sequence of the local temporal sequence semantic association interaction vectors of ultrasonic power-solution temperature to obtain an optimized sequence of the local temporal sequence semantic association interaction vectors of ultrasonic power-solution temperature; Matrixing the optimized sequence of ultrasound power-solution temperature local time series semantic association interaction vectors to obtain an ultrasound power-solution temperature time series fine-grained interaction global semantic representation matrix; An autocorrelation interaction perception analysis is performed on the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix to obtain the ultrasonic power-solution temperature local-global interaction response matrix.

8. The zinc pyridinecarboxylate complex composite catalyst according to claim 7, characterized in that: The ultrasonic power-solution temperature time series fine-grained interactive global semantic representation matrix is ​​subjected to autocorrelation interactive perception analysis to obtain the ultrasonic power-solution temperature local-global interactive response matrix, including: Performing matrix multiplication of the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix and the transposed matrix of the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix, and then dividing by the scale of the ultrasonic power-solution temperature time series fine-grained interaction global semantic representation matrix by position to obtain the ultrasonic power-solution temperature time series fine-grained interaction feature global dependency matrix; The Relu function is used to activate the global dependency matrix of the ultrasonic power-solution temperature time series fine-grained interaction features to obtain the ultrasonic power-solution temperature local-global interaction response matrix.

Images