A method for optimizing the hydrothermal carbonization process of seaweed based on machine learning

Abstract

The present invention relates to the technical field of seaweed self-water hydrothermal carbonization optimization, and specifically to a seaweed self-water hydrothermal carbonization process optimization method based on machine learning, comprising: preparing hydrothermal carbon using seaweed as raw material, collecting reaction temperature, reaction time, and corresponding physicochemical parameters of hydrothermal carbon products, and using reaction temperature, reaction time and physicochemical parameters of hydrothermal carbon products as training sets. Establishing a machine learning model, and using the training set to train the machine learning model. Determine the corresponding required physicochemical parameters according to the purchase needs of buyers, and use the machine learning model to match the required physicochemical parameters with the corresponding required reaction temperature and required reaction time. Using seaweed as raw material, hydrothermal carbon is produced according to the required reaction temperature and required reaction time. It can help manufacturers realize customized production for buyers, so that the products are highly matched with the use needs, improve the accuracy of the use of hydrothermal carbon products in subsequent applications, and reduce unnecessary waste.

Description

Technical Field

[0001] The present invention relates to the technical field of seaweed self-water hydrothermal carbonization optimization, and in particular to a seaweed self-water hydrothermal carbonization process optimization method based on machine learning. Background Art

[0002] Hydrothermal carbonization is a black solid product with carbon as the main component and rich oxygen-containing functional groups, which is obtained by hydrothermal reaction at a certain reaction temperature and autogenous pressure using biomass or its components as raw materials and water as solvent and reaction medium. Hydrothermal carbonization technology has attracted more and more attention in realizing the efficient conversion and application of biomass resources due to its economy and environmental friendliness.

[0003] In the application process of the existing hydrothermal carbonization process, manufacturers independently produce hydrothermal carbon products, independently formulate product lines and product types, and buyers choose from the hydrothermal carbon products on the market according to their own purchasing needs. For manufacturers, their production and sales processes are passive and blind, and they cannot effectively customize hydrothermal carbon products according to the actual needs of buyers. This makes it possible that the products purchased by buyers may not meet their actual needs, which directly affects the accuracy of the use of hydrothermal carbon products in subsequent applications, and also causes unnecessary waste of hydrothermal carbon products, and cannot provide accurate services to buyers.

[0004] In view of this, this application is hereby filed. Summary of the invention

[0005] The purpose of the present invention is to provide a method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning, which can help manufacturers realize customized production for buyers, so that the products purchased by buyers are highly matched with their actual use needs, thereby improving the use accuracy of hydrothermal carbonization products in subsequent applications, reducing unnecessary waste of hydrothermal carbonization products, improving the production accuracy of manufacturers, and realizing precise services for buyers.

[0006] The embodiment of the present invention is achieved as follows:

[0007] A method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning, comprising the following steps:

[0008] S1. Prepare hydrochar using seaweed as raw material, collect reaction temperature, reaction time, and corresponding physical and chemical parameters of hydrochar products, and use reaction temperature, reaction time and physical and chemical parameters of hydrochar products as training sets.

[0009] S2. Establish a machine learning model and train the machine learning model using the training set.

[0010] S3. Determine the corresponding physical and chemical parameters according to the buyer's purchasing needs, and use the machine learning model to match the corresponding physical and chemical parameters with the corresponding response temperature and response time.

[0011] S4. Using seaweed as raw material, hydrothermal charcoal is produced according to the required reaction temperature and required reaction time.

[0012] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0013] S5. A technical information database is established for each buyer, in which the buyer's "required physical and chemical parameters - hydrothermal charcoal usage" data is stored.

[0014] In S3, based on the "demand physical and chemical parameters - hydrothermal charcoal usage" data, the corresponding demand physical and chemical parameters are determined according to the buyer's purchasing needs.

[0015] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0016] S6. Find the intersection of the "demand physical and chemical parameters - hydrothermal carbon use" data of all buyers, and use the number of data in the intersection to represent the popularity of the technology. The more data in the intersection, the higher the popularity of the technology.

[0017] S7. Calculate the union of the "demand physical and chemical parameters - hydrothermal carbon use" data of all buyers, and use the number of data in the union to represent the scope of technical application. The more data in the union, the wider the scope of technical application.

[0018] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0019] S8. Determine the buyer's current demand physical and chemical parameters based on the buyer's "demand physical and chemical parameters - hydrothermal charcoal uses" data, and use them as matching keywords. Search for content matching the matching keywords in the technical forum posts and / or technical literature in the buyer's industry, and extract the new uses involved in the matching keywords mentioned therein. Supplement the buyer's "demand physical and chemical parameters - hydrothermal charcoal uses" data based on the new use, and mark the data corresponding to the new use as data to be verified. Send the data to be verified to the corresponding buyer, and push the corresponding technical forum posts and / or technical literature to the buyer. If the push receives an affirmative response from the buyer, the data to be verified will be formally included in the "demand physical and chemical parameters - hydrothermal charcoal uses" data.

[0020] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0021] S9. For each buyer, a visual view is constructed according to the corresponding "demand physical and chemical parameters - hydrothermal charcoal use" data. In the visual view, a variable axis is constructed for each data in the demand physical and chemical parameters, and the variable axes are arranged in parallel and evenly spaced. The lengths of the variable axes are equal, and the two end points of the variable axes correspond to the lower limit and upper limit of the corresponding data.

[0022] S10. For each data in the data of "demand physical and chemical parameters - hydrothermal charcoal use", determine the value segment of the corresponding data on each variable axis. For adjacent variable axes, construct connecting lines between the two lower limits of the two corresponding value segments and between the two upper limits. The area enclosed by all connecting lines and the two variable axes at the two ends is used as the evaluation area.

[0023] S11. The area of ​​a single evaluation area is used to represent the breadth of a buyer's single-purpose technology in the corresponding hydrothermal carbon application, wherein the larger the area of ​​a single evaluation area, the higher the breadth of the single-purpose technology. The sum of the areas of all the buyer's evaluation areas is used to represent the breadth of the buyer's all-purpose technology in the overall application of hydrothermal carbon, wherein the larger the sum of the areas of the evaluation areas, the higher the breadth of the all-purpose technology.

[0024] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0025] S12. For two buyers, the evaluation areas of the two buyers for the same hydrothermal charcoal use are placed in the same visual view, and the overlap of the evaluation areas of the two buyers is determined. The overlapping areas with at least two intersections with the same variable axis are taken as the first overlapping areas, and the remaining overlapping areas are taken as the second overlapping areas.

[0026] S13: Count the sum of the areas of all first overlapping regions and the sum of the areas of all second overlapping regions. The sum of the areas of the first overlapping regions is used to represent the technical relevance of the two buyers, wherein the larger the sum of the areas of the first overlapping regions is, the higher the technical relevance of the two buyers is.

[0027] The sum of the areas of the second overlapping regions is used to represent the technical selection preferences of the two buyers, wherein the larger the sum of the areas of the second overlapping regions is, the greater the difference in the technical selection preferences of the two buyers is.

[0028] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0029] S14, setting a technology relevance threshold, and determining the first overlapping region area and threshold corresponding thereto. Setting a technology selection tendency threshold, and determining the second overlapping region area and threshold corresponding thereto.

[0030] S15. For any buyer, determine the sum of the areas of the first overlapping areas and the sum of the areas of the second overlapping areas between the buyer and other buyers, and screen out other buyers whose sum of the areas of the first overlapping areas is greater than or equal to a threshold of the sum of the areas of the first overlapping areas and whose sum of the areas of the second overlapping areas is less than the threshold of the sum of the areas of the second overlapping areas as similar buyers.

[0031] S16. Establish a technical association relationship between the buyer and similar buyers, and notify the manager of the buyers with the technical association relationship.

[0032] The beneficial effects of the technical solution of the embodiment of the present invention include:

[0033] In the seaweed self-water hydrothermal carbonization process optimization method based on machine learning provided in an embodiment of the present invention, the water content of fresh algae is above 95%. Even if there is water loss during transportation, its final water content can usually be maintained at above 50%, which fully meets the water requirement of the hydrothermal carbonization process. By using seaweed as the raw material for the hydrothermal carbonization process, the water contained in the seaweed itself (i.e., self-water) can be used to complete the hydrothermal carbonization without adding additional water. Based on this, in the process of seaweed self-water hydrothermal carbonization, only the reaction temperature and reaction time need to be controlled.

[0034] It is the use of seaweed as raw material for the hydrothermal carbonization process that simplifies the production variables that need to be controlled (reaction temperature and reaction time), reduces the complexity of the machine learning model, and helps improve the accuracy of the machine learning model.

[0035] By adding machine learning models, customized production of hydrothermal charcoal products can be achieved based on the actual needs of buyers, reducing the blindness of production by manufacturers and eliminating the need for buyers to choose from the numerous hydrothermal charcoal products on the market. While improving the supply accuracy of hydrothermal charcoal products, it also reduces the purchasing burden of buyers, greatly improving the efficiency and accuracy of the hydrothermal charcoal product supply chain, reducing the waste of hydrothermal charcoal products due to the mismatch between products and uses, and improving resource utilization.

[0036] In general, the seaweed self-water hydrothermal carbonization process optimization method based on machine learning provided in the embodiments of the present invention can help manufacturers realize customized production for buyers, so that the products purchased by buyers are highly matched with their actual use needs, improve the use accuracy of hydrothermal carbonization products in subsequent applications, reduce unnecessary waste of hydrothermal carbonization products, improve the production accuracy of manufacturers, and realize accurate service to buyers. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 A schematic diagram of the variable axes in the visual view;

[0039] Figure 2 It is a schematic diagram of the value range of physical and chemical parameters;

[0040] Figure 3 is a schematic diagram of the evaluation area;

[0041] Figure 4 is a schematic diagram of the first overlapping area;

[0042] Figure 5 is a schematic diagram of the second overlapping area. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0044] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] The terms “first”, “second”, etc. are only used for distinguishing descriptions and should not be understood as indicating or implying relative importance.

[0046] As shown in this specification and claims, unless the context clearly provides an example, the words "a", "an", "the", etc. do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0047] The flowcharts used in this specification are used to illustrate the operations performed by the system according to the embodiments of this specification. It is understood that the operations of each step are not necessarily performed precisely in order. On the contrary, the steps can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0048] In order to overcome the defects in the prior art, this embodiment provides a method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning, which comprises the following steps:

[0049] S1. Prepare hydrochar using seaweed as raw material, collect reaction temperature, reaction time, and corresponding physical and chemical parameters of hydrochar products, and use reaction temperature, reaction time, and physical and chemical parameters of hydrochar products as training sets. That is, the physical and chemical parameters of hydrochar products prepared at a certain reaction temperature and reaction time are matched one by one to form a training set. Among them, the physical and chemical parameters corresponding to the hydrochar products include but are not limited to: surface composition, pore size, porosity, microstructure, pH value (acidity), etc. The types of physical and chemical parameters actually required can be flexibly selected and adjusted according to specific circumstances.

[0050] S2. Establish a machine learning model and train the machine learning model using the training set.

[0051] S3. Determine the corresponding physical and chemical parameters according to the buyer's purchase needs, and use the machine learning model to match the corresponding physical and chemical parameters with the corresponding reaction temperature and reaction time. That is to say, when the buyer expresses his willingness to buy, the physical and chemical parameters of the hydrothermal charcoal product he needs are determined according to the buyer's purchase needs (i.e., the physical and chemical parameters of the demand), and the machine learning model is used to match the hydrothermal charcoal product corresponding to the "physical and chemical parameters of the demand" with the corresponding reaction temperature (i.e., the reaction temperature of the demand) and reaction time (i.e., the reaction time of the demand) in the preparation process.

[0052] S4. Using seaweed as raw material, hydrothermal charcoal is produced according to the required reaction temperature and required reaction time. In this way, the hydrothermal charcoal products actually required by the buyer (matching the buyer's purchase demand) can be accurately produced.

[0053] In this scheme, the water content of fresh algae is above 95%. Even if there is water loss during transportation, its final water content can usually be maintained above 50%, which fully meets the water requirement of the hydrothermal carbonization process. Using seaweed as the raw material for the hydrothermal carbonization process, the water contained in the seaweed itself (i.e., self-water) can be used to complete the hydrothermal carbonization without adding additional water. Based on this, in the process of hydrothermal carbonization of seaweed self-water, only the reaction temperature and reaction time need to be controlled.

[0054] It is the use of seaweed as raw material for the hydrothermal carbonization process that simplifies the production variables that need to be controlled (reaction temperature and reaction time), reduces the complexity of the machine learning model, and helps improve the accuracy of the machine learning model.

[0055] By adding machine learning models, customized production of hydrothermal charcoal products can be achieved based on the actual needs of buyers, reducing the blindness of production by manufacturers and eliminating the need for buyers to choose from the numerous hydrothermal charcoal products on the market. While improving the supply accuracy of hydrothermal charcoal products, it also reduces the purchasing burden of buyers, greatly improving the efficiency and accuracy of the hydrothermal charcoal product supply chain, reducing the waste of hydrothermal charcoal products due to the mismatch between products and uses, and improving resource utilization.

[0056] In general, the seaweed self-hydraulic hydrothermal carbonization process optimization method based on machine learning provided in this embodiment can help manufacturers achieve customized production for buyers, so that the products purchased by buyers are highly matched with their actual use needs, improve the use accuracy of hydrothermal carbonization products in subsequent applications, reduce unnecessary waste of hydrothermal carbonization products, improve the production accuracy of manufacturers, and achieve accurate services for buyers.

[0057] In this embodiment, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0058] S5. A technical information database is established for each buyer, in which the buyer's "required physical and chemical parameters - hydrothermal charcoal usage" data is stored.

[0059] There are several data items in the "Required physical and chemical parameters - hydrothermal charcoal use" data, each of which is a specific / range of physical and chemical parameters and the specific use of hydrothermal charcoal products with these physical and chemical parameters. The "Required physical and chemical parameters - hydrothermal charcoal use" data records the different uses of hydrothermal charcoal products with different physical and chemical parameters by the buyer. A separate technical information database is established for each buyer to facilitate separate statistics on the actual needs of each buyer and accurate analysis of the purchase needs of each buyer. This is because in the actual transaction process, different buyers may directly provide the required physical and chemical parameters of the hydrothermal charcoal when proposing purchase needs, or they may directly propose usage needs.

[0060] In S3, based on the data of “demand physical and chemical parameters - hydrothermal charcoal usage”, the corresponding demand physical and chemical parameters are determined according to the buyer’s purchasing needs (which may be clear physical and chemical parameters or usage requirements).

[0061] Based on this, the seaweed self-water hydrothermal carbonization process optimization method based on machine learning also includes the following steps:

[0062] S6. Find the intersection of the "demand physical and chemical parameters - hydrothermal charcoal use" data of all buyers, and use the number of data in the intersection to represent the popularity of the technology. The more data in the intersection, the higher the popularity of the technology. Through this operation, we can know how many of the physical and chemical parameters of hydrothermal charcoal products corresponding to all buyers have exactly the same specific uses, that is, how many hydrothermal charcoal product application technologies are used by these buyers, which can reflect the general application scale of hydrothermal charcoal product application technologies among buyers to a certain extent, that is, the popularity of the technology.

[0063] S7. Calculate the union of the "demand physical and chemical parameters - hydrothermal carbon use" data of all buyers, and use the number of data in the union to represent the scope of technical application. The more data in the union, the wider the scope of technical application. Through this operation, we can know the scale of the application of hydrothermal carbon products among all buyers, which can reflect the actual application of hydrothermal carbon products to a certain extent, that is, it reflects the scope of technical application.

[0064] Through this design, statistics and census of hydrothermal carbon product application technology can be completed at the same time, so as to more intuitively reflect the actual application of hydrothermal carbon products in related industries, which has positive significance for evaluating the application of industry technology. In addition, the data items other than the intersection are concentrated, that is, the hydrothermal carbon application technology that is not widely used in the industry, which also provides guidance for promoting the popularization of hydrothermal carbon product application technology in the industry.

[0065] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0066] S8. Determine the physical and chemical parameters that the buyer is using (i.e., the physical and chemical parameters that are being used) based on the buyer's "demand physical and chemical parameters - hydrothermal charcoal usage" data, and use them as matching keywords.

[0067] Search for content matching the matching keywords in technical forum posts and / or technical literature in the buyer's industry, and extract new uses involving the matching keywords mentioned therein. Semantic analysis methods may be used for retrieval and extraction, but are not limited thereto.

[0068] According to the new use, the buyer's "Required Physical and Chemical Parameters - Hydrothermal Charcoal Use" data is supplemented (i.e. additional data entries are added), and the data corresponding to the new use is marked as data to be verified.

[0069] The data to be verified will be sent to the corresponding buyer, and the corresponding technical forum post and / or technical literature will be pushed to the buyer. If the push receives a positive response from the buyer (i.e. the buyer determines that the new use is actually feasible), the data to be verified will be formally included in the "Required Physical and Chemical Parameters - Hydrothermal Carbon Use" data.

[0070] When the data to be verified is formally included in the "Required Physical and Chemical Parameters - Hydrothermal Charcoal Use" data: (1) If the newly included data is a new use for a certain specific / range of physical and chemical parameters that already existed, you can confirm with the buyer whether to retain the original use. If it is retained, the new use will be added to the use corresponding to the physical and chemical parameter. If it is not retained, the new use will replace the original use. (2) If the newly included data is a completely new range and a completely new use, add a new entry directly. (3) If the newly included data is a further division of the original range of physical and chemical parameters, the physical and chemical parameters of the original range will be split according to the numerical endpoints of the newly included data and split into multiple range values. For the new use corresponding to the newly included data, you can confirm with the buyer whether to retain the original use of the range. If it is retained, the new use will be added to the use of the corresponding range value. If it is not retained, the new use will replace the original use of the corresponding range value.

[0071] Through this design, the hydrothermal charcoal application technology is actively pushed and updated to buyers, which helps promote the application and popularization of technology in this field.

[0072] In order to further promote technical exchanges between different buyers in this field, the seaweed self-water hydrothermal carbonization process optimization method based on machine learning also includes the following steps:

[0073] S9. For each buyer, a visual view is constructed according to the corresponding "demand physical and chemical parameters-hydrothermal charcoal use" data. In the visual view, a variable axis is constructed for each type of data in the demand physical and chemical parameters, and the variable axes of different types of data are arranged in parallel and evenly spaced. The lengths of the variable axes are equal, and the end points of the variable axes correspond to the lower and upper limits of the corresponding data. Among them, in the visual view of each buyer, the lengths of the variable axes are equal, and the lower and upper limits of the variable axes of the same type are also the same. Let's take the pore size, porosity and pH value in the physical and chemical parameters as an example. Figure 1 As shown, variable axes are constructed for pore size, porosity and pH value respectively. The variable axis can be a horizontal axis or a vertical axis, which can be flexibly selected. We take the vertical axis as an example for demonstration and explanation, but it is not limited to this. Among them, the pore size corresponds to the variable axis Y1, the lower limit of the pore size is Y1s1 (corresponding to the lower end point of the variable axis Y1), and the upper limit of the pore size is Y1s2 (corresponding to the upper end point of the variable axis Y1); the porosity corresponds to the variable axis Y2, the lower limit of the porosity is Y2s1 (corresponding to the lower end point of the variable axis Y2), and the upper limit of the porosity is Y2s2 (corresponding to the upper end point of the variable axis Y2); the pH value corresponds to the variable axis Y3, the lower limit of the pH value is Y3s1 (corresponding to the lower end point of the variable axis Y3), and the upper limit of the pH value is Y3s2 (corresponding to the upper end point of the variable axis Y3).

[0074] S10. For each data in the data of "Required physical and chemical parameters - hydrothermal charcoal use", determine the value range of the corresponding data on each variable axis. If the value is a fixed value, it is a point. For example, for a certain use, the value range of pore size is d1~d2, the value range of porosity is p1~p2, and the value range of pH value is a1~a2. Figure 2 As shown, the corresponding variable axes are marked. For adjacent variable axes, connecting lines are constructed between the two lower limits of the two corresponding value segments and between the two upper limits. The area surrounded by all connecting lines and the two variable axes at the two ends is used as the evaluation area. Figure 3 As shown, connecting lines are constructed between d1 and p1, between p1 and a1, between d2 and p2, and between p2 and a2. If the variable axis Y1 and the variable axis Y3 are located at the two sides, the evaluation area enclosed is area Q.

[0075] S11. The area of ​​a single evaluation area is used to represent the breadth of a buyer's single-purpose technology in the corresponding hydrothermal carbon application. The larger the area of ​​a single evaluation area, the higher the breadth of the single-purpose technology, which means that the buyer has a higher technology breadth in this application. The sum of the areas of all evaluation areas of the buyer (i.e., the evaluation areas of all data items in the buyer's "demand physical and chemical parameters-hydrothermal carbon application" data) is used to represent the breadth of the buyer's all-purpose technology in the overall application of hydrothermal carbon. The larger the sum of the areas of the evaluation areas, the higher the breadth of the all-purpose technology.

[0076] Through this design, the technical aspects of different buyers in the actual application of hydrothermal charcoal products can be intuitively evaluated and reflected. It can be used as a basis for evaluating the buyer's technical mastery, as well as a reference for evaluating the scope of the buyer's technical coverage. It can help to intuitively analyze the technical application of different merchants in this field.

[0077] Furthermore, the method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps:

[0078] S12. For two buyers, the evaluation areas of the two buyers for the same hydrothermal charcoal use (i.e., for the same use) are placed in the same visual view, and the overlap of the evaluation areas of the two buyers is determined. The overlapping area with at least two intersections with the same variable axis is taken as the first overlapping area (e.g., Figure 4 As shown in the middle area C1, but not limited thereto), the remaining overlapping area is used as the second overlapping area (such as Figure 5 As shown in the middle area C2 and C3, but not limited to this).

[0079] S13, counting the sum of the areas of all first overlapping areas (i.e., counting the sum of the areas of all first overlapping areas with the same purpose) and the sum of the areas of all second overlapping areas. The sum of the areas of the first overlapping areas is used to represent the technical relevance of the two buyers, wherein the larger the sum of the areas of the first overlapping areas is, the higher the technical relevance of the two buyers is.

[0080] The sum of the areas of the second overlapping regions is used to represent the technical selection preferences of the two buyers. The larger the sum of the areas of the second overlapping regions is, the greater the difference in the technical selection preferences of the two buyers is. In particular, if the sum of the areas of the first overlapping regions and the sum of the areas of the second overlapping regions are both 0, it means that the technical relevance of the two buyers is not high, and the difference in the technical selection preferences is also large.

[0081] This design can help analyze the technology usage and relationships between different buyers, and facilitate statistics and analysis of technology usage within the industry.

[0082] Based on this, the seaweed self-water hydrothermal carbonization process optimization method based on machine learning also includes the following steps:

[0083] S14, setting a technology relevance threshold, and determining the first overlapping region area and threshold corresponding thereto. Setting a technology selection tendency threshold, and determining the second overlapping region area and threshold corresponding thereto.

[0084] S15. For any buyer, determine the sum of the areas of the first overlapping areas and the sum of the areas of the second overlapping areas between the buyer and other buyers, and screen out other buyers whose sum of the areas of the first overlapping areas is greater than or equal to a threshold of the sum of the areas of the first overlapping areas and whose sum of the areas of the second overlapping areas is less than the threshold of the sum of the areas of the second overlapping areas as similar buyers.

[0085] S16. Establish a technical association relationship between the buyer and similar buyers, and notify the manager of the buyers with the technical association relationship.

[0086] This design can help screen out buyers with high technical correlations with each other, making it easier for managers to establish technical communication channels between these buyers based on actual conditions, promote in-depth technical exchanges between these buyers, and thus promote the in-depth development of hydrothermal carbon product application technology, which helps to improve the technology circulation and technological development speed in this field.

[0087] In summary, the seaweed self-water hydrothermal carbonization process optimization method based on machine learning provided in the embodiments of the present invention can help manufacturers realize customized production for buyers, so that the products purchased by buyers are highly matched with their actual use needs, thereby improving the use accuracy of hydrothermal carbonization products in subsequent applications, reducing unnecessary waste of hydrothermal carbonization products, improving the production accuracy of manufacturers, and realizing accurate services for buyers.

[0088] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning, characterized in that: The steps include: S1. Prepare hydrochar using seaweed as raw material, collect reaction temperature, reaction time, and corresponding physical and chemical parameters of the hydrochar product, and use the reaction temperature, reaction time, and physical and chemical parameters of the hydrochar product as a training set; S2. Establish a machine learning model, and train the machine learning model using the training set; S3. Determine the corresponding physical and chemical parameters according to the buyer's purchase demand, and use the machine learning model to match the corresponding demand reaction temperature and demand reaction time for the physical and chemical parameters; S4, using seaweed as raw material, producing hydrothermal charcoal according to the required reaction temperature and the required reaction time; S5. Establish a technical information database for each buyer, wherein the technical information database stores the buyer's "required physical and chemical parameters - hydrothermal charcoal usage" data; S6. Calculate the intersection of the "demand physical and chemical parameters - hydrothermal charcoal usage" data of all buyers, and use the number of data items in the intersection to represent the popularity of the technology; wherein, the more data items in the intersection, the higher the popularity of the technology; S7, obtaining a union of the "demand physical and chemical parameters - hydrothermal charcoal usage" data of all buyers, and using the number of data items in the union to represent the scope of technical application; wherein, the more data items in the union, the wider the scope of technical application; Among them, in S3, based on the "demand physical and chemical parameters - hydrothermal charcoal usage" data, the corresponding demand physical and chemical parameters are determined according to the buyer's purchasing needs.

2. The method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning according to claim 1, characterized in that: The method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps: S8. Determine the required physical and chemical parameters that the buyer is using based on the buyer's "required physical and chemical parameters - hydrothermal charcoal uses" data, and use them as matching keywords; search for content matching the matching keywords in technical forum posts and / or technical literature in the buyer's industry, and extract new uses involved in the matching keywords mentioned therein; supplement the buyer's "required physical and chemical parameters - hydrothermal charcoal uses" data based on the new uses, and mark the data corresponding to the new uses as data to be verified; send the data to be verified to the corresponding buyer, and push the corresponding technical forum posts and / or technical literature to the buyer; if the push receives an affirmative reply from the buyer, the data to be verified will be formally included in the "required physical and chemical parameters - hydrothermal charcoal uses" data.

3. The method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning according to claim 1, characterized in that: The method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps: S9. For each buyer, a visual view is constructed according to the corresponding "demand physical and chemical parameters - hydrothermal charcoal use" data; in the visual view, a variable axis is constructed for each data in the demand physical and chemical parameters, and the variable axes are arranged in parallel and evenly spaced; the lengths of the variable axes are equal, and the two end points of the variable axes correspond to the lower limit and upper limit of the corresponding data respectively; S10, for each piece of data in the "demand physical and chemical parameters - hydrothermal charcoal use" data, determine the value segment of the corresponding data on each variable axis; for adjacent variable axes, construct connecting lines between the two lower limit values ​​of the two corresponding value segments and between the two upper limit values; and use the area enclosed by all the connecting lines and the two variable axes at the two ends as the evaluation area; S11. The area of ​​a single evaluation area is used to represent the breadth of the buyer's single-purpose technology in the corresponding hydrothermal charcoal application, wherein the larger the area of ​​a single evaluation area is, the higher the breadth of the single-purpose technology is; the sum of the areas of all the evaluation areas of the buyer is used to represent the breadth of all-purpose technology of the buyer in the overall application of hydrothermal charcoal, wherein the larger the sum of the areas of the evaluation areas is, the higher the breadth of the all-purpose technology is.

4. The method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning according to claim 3 is characterized in that: The method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps: S12, for two buyers, placing the evaluation areas of the two buyers for the same hydrothermal charcoal use in the same visual view, determining the overlap of the evaluation areas of the two buyers, taking the overlapping areas having at least two intersections with the same variable axis as the first overlapping areas, and taking the remaining overlapping areas as the second overlapping areas; S13, calculating the sum of the areas of all the first overlapping regions and the sum of the areas of all the second overlapping regions; using the sum of the areas of the first overlapping regions to represent the technical relevance of the two buyers, wherein the larger the sum of the areas of the first overlapping regions is, the higher the technical relevance of the two buyers is; The sum of the areas of the second overlapping regions is used to represent the technical selection preferences of the two buyers, wherein the larger the sum of the areas of the second overlapping regions is, the greater the difference in the technical selection preferences of the two buyers is.

5. The method for optimizing the process of seaweed self-water hydrothermal carbonization based on machine learning according to claim 4 is characterized in that: The method for optimizing the seaweed self-water hydrothermal carbonization process based on machine learning also includes the following steps: S14, setting a technology relevance threshold, and determining a first overlapping region area and a threshold corresponding thereto; setting a technology selection tendency threshold, and determining a second overlapping region area and a threshold corresponding thereto; S15. For any buyer, determine the sum of the areas of the first overlapping areas and the sum of the areas of the second overlapping areas between the buyer and other buyers, and select other buyers whose sum of the areas of the first overlapping areas is greater than or equal to a threshold of the sum of the areas of the first overlapping areas and whose sum of the areas of the second overlapping areas is less than a threshold of the sum of the areas of the second overlapping areas as similar buyers; S16: Establish a technical association relationship between the buyer and the similar buyers, and notify the manager of the buyers with the technical association relationship.

Images