Methanol coal blending algorithm based on gating circulation unit
By using deep learning algorithm based on gated cycle units in the coal-to-methanol industry to build a methanol coal mixing model, the problem of insufficient reliability of coal mixing models in the industry is solved, the scientificity and accuracy of the ratio of coal in furnaces is achieved, and the production cost is reduced.
Patent Information
- Application Number
- CN202510010583.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-03
- Publication Date
- 2025-05-06
AI Technical Summary
The coal-to-methanol industry lacks scientific coal mixing methods, which leads to waste of resources and high production costs of coal entering the furnace. The existing coal mixing models have not been fully integrated into the experience of professional coal mixing engineers, resulting in insufficient model reliability.
A deep learning algorithm framework based on gated cycle units is used to construct a methanol coal mixing model, combining coal inventory, coal index information and boiler design indicators, integrating the professional knowledge and production experience of coal mixing engineers, and self-learning through historical data and continuous production data to optimize model performance.
It achieves the highest cost-effective coal ratio in furnaces under the conditions of meeting the production indicator requirements, which improves the accuracy and scientificity of coals, enhances the adaptability and reliability of the model, and reduces the comprehensive cost of in furnaces.
Smart Images
Figure CN119943188A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent coal blending, and in particular to a methanol coal blending algorithm based on a gated circulation unit. Background Art
[0002] Coal-to-methanol plays a connecting role in the coal chemical industry. The production of high-quality methanol can significantly improve the quality of subsequent olefins and fine chemicals, and can achieve energy conservation and environmental protection from the source. Therefore, under the premise of ensuring the quality of methanol production, it is crucial to reduce the production cost of coal entering the furnace. The factors affecting methanol production mainly include production process, boiler design, and coal characteristics. The main indicators of coal types include calorific value, moisture, ash content, volatile matter, sulfur content, alkali metal content, silicon-aluminum ratio, etc., and according to the differences in furnace types, some furnace types need to additionally consider the softening temperature, flow temperature, deformation temperature and Hastelloy grindability index of the coal type. Because there are many indicators and conditions that need to be considered, it is very difficult to find the optimal ratio based solely on manual experience.
[0003] In the modern coal chemical industry, various coal blending algorithms have been widely used in power plants and coking plants to improve the resource utilization conversion rate of coal and achieve the goal of reducing costs and increasing efficiency. However, there is a lack of relevant applications in the coal-to-methanol industry. Most methanol plants still rely entirely on the work experience of coal blending engineers to complete coal blending, which leads to unscientific and human factors that waste resources.
[0004] There is no relevant coal blending method in the coal-to-methanol industry, but it is relatively mature in the power coal blending and coking coal blending industries. There are some differences between them and they cannot be directly applied. The existing technology uses a mathematical model for power coal blending that minimizes the proportion of high-quality coal, maximizes the proportion of inferior coal, and achieves the lowest unit price of raw coal. On this basis, it uses a generalized regression neural network for prediction and uses an exhaustive method to screen out the most cost-effective blending scheme; the model does not fully consider the actual operating conditions of the boiler and the deviation problems in the production process, resulting in it usually only being able to obtain a local optimal solution. In addition, there is also an artificial intelligence coal blending system for coking based on a neural network model. The system mainly constructs a coke quality prediction model by analyzing existing coking data and exploring the relationship between coal blending ratio and coke quality. It does not incorporate the production experience of coal blending experts, and its reliability still needs further verification. Summary of the invention
[0005] The present invention provides a methanol coal blending algorithm based on a gated circulation unit, and constructs a methanol coal blending model using a deep learning algorithm framework. Under the constraints of coal inventory, coal index information, and boiler design indicators, the coal blending ratio that meets production requirements and has the highest cost-effectiveness is obtained, which effectively solves the problem that the coal blending models of other types of existing coal blending have failed to fully incorporate the experience of professional coal blending engineers, resulting in insufficient model reliability.
[0006] The present invention provides a methanol coal blending algorithm based on a gated circulation unit, comprising:
[0007] Collecting historical actual production coal blending data provided by the methanol plant, and performing data preprocessing on the historical actual production coal blending data;
[0008] The processed historical actual production coal blending data is labeled and divided into a training set and a validation set, so as to train a preset GRU model using the training set and the validation set to obtain a target model;
[0009] The raw coal information is collected as the input of the target model, and after calculation by the target model, a coal type ratio scheme with the lowest input and output coal cost is output; wherein the raw coal information is inventory information or purchasable information.
[0010] Furthermore, in the step of collecting the historical actual production coal blending data provided by the methanol plant and performing data preprocessing on the historical actual production coal blending data, the data preprocessing method is to process the missing values, that is, to process them by deletion, interpolation or filling with the mean / median to ensure the uniformity of the data in format, unit and range.
[0011] Furthermore, in the step of annotating the processed historical actual production coal blending data and dividing it into a training set and a validation set, so as to use the training set and the validation set to train the preset GRU model and obtain the target model, a coal blending expert annotates the processed historical actual production coal blending data to identify the data with excellent performance in the historical actual production coal blending data; after cleaning and annotating the historical actual production coal blending data, a data set for training the preset GRU model is formed, and it is divided into a training set and a validation set in a ratio of 7:3.
[0012] Furthermore, the processed historical actual production coal blending data is labeled and divided into a training set and a validation set, and the preset GRU model is trained using the training set and the validation set. In the step of obtaining the target model, the preset GRU model has a built-in update gate and a reset gate so that the model can effectively control the retention of valid information. By analyzing the historical actual production coal blending data, effective features are extracted and features that have a great impact on the model performance are screened out to analyze the contribution of the features to the model prediction results.
[0013] Furthermore, in the preset GRU model, according to the actual demand of coal blending in the methanol plant, in order to obtain the lowest cost ratio of coal entering the furnace, the model objective function is set as:
[0014] minW=P T X+B
[0015] Among them, W is the comprehensive cost of coal entering the furnace, X is the quality of a single type of coal, P T is the unit price of a single type of coal, and B is the adjustment parameter of the coal blending model, which includes the business experience of coal blending engineers, historical coal blending data learning experience and model self-learning experience.
[0016] Furthermore, in the preset GRU model, the update gate is responsible for evaluating the importance of the input information to determine the information to be integrated and passed to the next state, thereby ensuring that the network can effectively retain the core data. The specific formula of the update gate is:
[0017] z t =σ(A (z) x t +U (z) h t―1 )
[0018] Among them, x t Indicates the current input, A (z) Its weight, h t―1 is the hidden information passed down from the previous node, U (z) is its weight, σ represents the activation function, and the final result is shown as z t ;
[0019] The hidden information h of the current node t The update method is as follows. After the update, the input is input to the reset gate r t middle:
[0020] h t =z t ⊙h t―1 +(1―z t )⊙h t
[0021] r t =σ(A (r) x t +U (r) h t―1 )
[0022] Reset Gate t Calculate the amount of data to be deleted, and keep z except resetting unnecessary data t With h t The core factors affecting coal blending are extracted from the results of the calculation in the formula:
[0023] h t =tanh(Ax t +r t ⊙Uh t―1 )
[0024] After completing the construction of the core algorithm of the above model, the target model is obtained by adjusting the hyperparameters, learning rate, and activation function, and after multiple iterative training of the training set data and verification on the verification set data.
[0025] Furthermore, after the step of collecting raw coal information as input of the target model and outputting the coal type ratio scheme with the lowest input and output coal cost after calculation by the target model, the step further includes:
[0026] The actual production data is collected and compared with the predicted data output by the target model, and the results are fed back to the target model to optimize and update the parameters of the target model.
[0027] The present invention also provides a methanol coal blending device based on a gate-controlled circulation unit, comprising:
[0028] A collection module, used to collect historical actual production coal blending data provided by the methanol plant, and perform data preprocessing on the historical actual production coal blending data;
[0029] A labeling module, used for labeling the processed historical actual production coal blending data, and dividing it into a training set and a validation set, so as to train a preset GRU model using the training set and the validation set to obtain a target model;
[0030] The output module is used to collect raw coal information as the input of the target model, and output the coal type ratio scheme with the lowest input and output coal cost after calculation by the target model; wherein the raw coal information is inventory information or purchasable information.
[0031] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.
[0032] The present invention also provides a computer-readable storage medium on which a computer program is stored, and the computer program implements the steps of the above method when executed by a processor.
[0033] The beneficial effects of the present invention are:
[0034] The present invention aims to reduce the comprehensive cost of coal for coal-to-methanol production, proposes a coal blending algorithm for coal-to-methanol production, and uses a deep learning algorithm framework to build a methanol coal blending model. Under the constraints of coal inventory, coal index information, and boiler design indicators, the coal blending ratio that meets various production index requirements and has the highest cost-effectiveness is obtained. In the process of model construction, the professional knowledge and production experience of coal blending engineers are incorporated to improve the accuracy and scientificity of coal blending. In addition, the model takes into account the characteristics of different gasifiers and can use historical data and continuous production data for self-learning to enhance the adaptability of the model. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 The figure is a schematic diagram of a method flow according to an embodiment of the present invention.
[0036] Figure 2 FIG. 1 is a schematic diagram of a device structure according to an embodiment of the present invention.
[0037] Figure 3 The figure is a schematic diagram of the internal structure of a computer device according to an embodiment of the present invention.
[0038] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0039] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.
[0040] The present invention relates to the process of using multiple types of coal to improve energy conversion rate and increase economic benefits in the field of coal-to-methanol production, and particularly designs a methanol-coal blending algorithm based on a gated circulation unit.
[0041] The coal-to-methanol blending algorithm based on the gated recurrent unit (GRU) provided by the present invention achieves the lowest cost of coal blending ratio for the furnace under the condition of meeting the requirements of various production indicators. It effectively solves the problem that the coal blending model in other existing inventions related to coal blending fails to fully incorporate the experience of professional coal blending engineers, resulting in insufficient model reliability. In addition, it overcomes the problem of large hardware resource consumption and difficulty in obtaining the global optimal solution caused by the exhaustive method used in some technologies.
[0042] like Figure 1 As shown, the present invention provides a methanol coal blending algorithm based on a gated circulation unit, comprising:
[0043] S1. Collect the historical actual production coal blending data provided by some large methanol plants, and perform data preprocessing on the historical actual production coal blending data, that is, perform data cleaning work, including but not limited to supplementing missing values, deleting dirty data, etc.
[0044] The original data comes from coal-to-methanol plants of several large listed companies. The actual production data in recent years is used as the data source, which is highly representative of the industry. The original data is purified, irrelevant and abnormal data are eliminated, and missing values are processed by deletion, interpolation, or filling with mean / median. Ensure the uniformity of data in format, unit and range to prevent analysis errors caused by data inconsistency, and coal blending experts will perform data labeling.
[0045] S2. Annotate the processed historical actual production coal blending data and divide it into a training set and a validation set, so as to train a preset GRU model using the training set and the validation set to obtain a target model.
[0046] The coal blending experts will annotate the processed historical actual production coal blending data to identify the data with excellent performance in the historical actual production coal blending data; after cleaning and annotating the historical actual production coal blending data, a data set for the preset GRU model training is formed, and it is divided into a training set and a validation set in a ratio of 7:3. The training set is used to perform preliminary training on the model to obtain a basic model, and then the validation set is used to evaluate and fine-tune the basic model, and finally the target model is established.
[0047] GRU (Gated Recurrent Unit) is a variant of RNN (Recurrent Neural Network) and is good at processing sequence data. GRU has built-in update gate and reset gate. These gate mechanisms enable GRU to effectively control the retention of effective information. It can extract effective features and filter out features that have a greater impact on model performance by analyzing historical coal blending data, and analyze the contribution of features to model prediction results.
[0048] According to the actual needs of the methanol plant coal blending work, in order to obtain the lowest cost ratio of coal entering the furnace, the model objective function can be abstracted as:
[0049] minW=P T X+B (1)
[0050] Among them, W is the comprehensive cost of coal entering the furnace, X is the quality of a single type of coal, P T is the unit price of a single type of coal, and B is the adjustment parameter of the coal blending model, which includes the business experience of coal blending engineers, historical coal blending data learning experience and model self-learning experience.
[0051] The update gate in GRU is responsible for evaluating the importance of input information to determine which information should be integrated and passed to the next state, thereby ensuring that the network can effectively retain the core data. The specific formula of the update gate is:
[0052] z t =σ(A (z) x t +U (z) h t―1 ) (2)
[0053] Among them, x t Indicates the current input, A (z) Its weight, h t―1 is the hidden information passed down from the previous node, U (z)is its weight, σ represents the activation function, and the final result is shown as z t ;
[0054] h t =z t ⊙h t―1 +(1―z t )⊙h t (3)
[0055] The hidden information of this node h t The update method is shown in formula (4). After the update, the input is input to the reset gate r t middle:
[0056] r t =σ(A (r) x t +U (r) h t―1 ) (4)
[0057] In (5), the amount of data to be deleted is calculated. In addition to resetting unnecessary data, the results calculated in (3) and (4) can be retained to extract the core factors affecting coal blending, as shown in (6):
[0058] h t =tanh(Ax t +r t ⊙Uh t―1 ) (6)
[0059] After completing the construction of the core algorithm of the above model, the model achieved a high accuracy on the validation set by carefully adjusting the hyperparameters, learning rate, activation function, etc., and after multiple iterative training of the training set data.
[0060] S3. After the model training is completed, the raw coal information (which may be inventory information or purchase information) is collected as the input of the target model, and the target model calculates and outputs the coal type ratio scheme with the lowest input and output coal cost.
[0061] S4. After the production process is completed, actual production data is collected and compared with the predicted data, and this information is fed back to the model. The model self-learns based on actual production information and continuously optimizes its own performance to better meet the production needs of specific furnace types, thereby reducing the prediction deviation caused by furnace type differences.
[0062] Based on the above core technology, the present invention provides a coal-to-methanol coal blending algorithm based on a gated circulation unit, which can obtain the coal type blending ratio with the lowest comprehensive coal input cost. It has been put into use in the production of a large methanol plant and has achieved considerable benefits in cost control. It uses the gated circulation unit technology to develop a coal-to-methanol coal blending model, which is trained based on the historical coal blending production data of some large methanol plants. In addition, the model has the ability to continuously learn and can self-optimize according to actual production data, thereby effectively correcting the prediction deviation caused by different furnace types; the coal blending model is used to guide the coal blending production work of the methanol plant. While ensuring that the various indicators of the coal input are within the specified range, the model can output the coal blending plan with the lowest coal input cost.
[0063] The present invention aims to reduce the comprehensive cost of coal for coal-to-methanol production, proposes a coal blending algorithm for coal-to-methanol production, and uses a deep learning algorithm framework to build a methanol coal blending model. Under the constraints of coal inventory, coal index information, and boiler design indicators, the coal blending ratio that meets various production index requirements and has the highest cost-effectiveness is obtained. In the process of model construction, the professional knowledge and production experience of coal blending engineers are incorporated to improve the accuracy and scientificity of coal blending. In addition, the model takes into account the characteristics of different gasifiers and can use historical data and continuous production data for self-learning to enhance the adaptability of the model.
[0064] like Figure 2 As shown, the present invention also provides a methanol coal blending device based on a gate-controlled circulation unit, comprising:
[0065] Collection module 1, used to collect historical actual production coal blending data provided by the methanol plant, and perform data preprocessing on the historical actual production coal blending data;
[0066] The labeling module 2 is used to label the processed historical actual production coal blending data and divide it into a training set and a verification set, so as to train the preset GRU model using the training set and the verification set to obtain a target model;
[0067] The output module 3 is used to collect raw coal information as the input of the target model, and output the coal type ratio scheme with the lowest input and output coal cost after calculation by the target model; wherein the raw coal information is inventory information or purchasable information.
[0068] In one embodiment, in the collection module 1, the data preprocessing method is to process the missing values, that is, to process them by deletion, interpolation or filling with mean / median to ensure the uniformity of the data in format, unit and range.
[0069] In one embodiment, in the labeling module 2, a coal blending expert labels the processed historical actual production coal blending data to identify the data with excellent performance in the historical actual production coal blending data; after cleaning and labeling the historical actual production coal blending data, a data set for training a preset GRU model is formed, and the data set is divided into a training set and a validation set in a ratio of 7:3.
[0070] In one embodiment, in the labeling module 2, the preset GRU model has built-in update gates and reset gates so that the model can effectively control the retention of valid information, extract effective features by analyzing the historical actual production coal blending data, and screen out features that have a great impact on the model performance, so as to analyze the contribution of the features to the model prediction results.
[0071] In one embodiment, in the preset GRU model of the marking module 2, according to the actual demand of the methanol plant coal blending work, in order to obtain the lowest cost ratio of the incoming coal, the model objective function is set to:
[0072] minW=P T X+B
[0073] Among them, W is the comprehensive cost of coal entering the furnace, X is the quality of a single type of coal, P T is the unit price of a single type of coal, and B is the adjustment parameter of the coal blending model, which includes the business experience of coal blending engineers, historical coal blending data learning experience and model self-learning experience.
[0074] In one embodiment, in the preset GRU model of the labeling module 2, the update gate is responsible for evaluating the importance of the input information to determine the information to be integrated and passed to the next state, thereby ensuring that the network can effectively retain the core data. The specific formula of the update gate is:
[0075] z t =σ(A (z) x t +U (z) h t―1 )
[0076] Among them, x t Indicates the current input, A (z) Its weight, h t―1 is the hidden information passed down from the previous node, U (z) is its weight, σ represents the activation function, and the final result is shown as z t ;
[0077] The hidden information h of the current node t The update method is as follows. After the update, the input is input to the reset gate r t middle:
[0078] h t =zt ⊙h t―1 +(1―z t )⊙h t
[0079] r t =σ(A (r) x t +U (r) h t―1 )
[0080] Reset Gate t Calculate the amount of data to be deleted, and keep z except resetting unnecessary data t With h t The core factors affecting coal blending are extracted from the results of the calculation in the formula:
[0081] h t =tanh(Ax t +r t ⊙Uh t―1 )
[0082] After completing the construction of the core algorithm of the above model, the target model is obtained by adjusting the hyperparameters, learning rate, and activation function, and after multiple iterative training of the training set data and verification on the verification set data.
[0083] In one embodiment, it further includes:
[0084] The optimization module is used to collect actual production data and compare and analyze the predicted data output by the target model, and feed the results back to the target model to optimize and update the parameters of the target model.
[0085] The above modules are used to execute the corresponding steps in the above-mentioned methanol coal blending algorithm based on the gated cycle unit. The specific implementation method thereof is described in the above-mentioned method embodiment, which will not be described in detail here.
[0086] like Figure 3 As shown, the present invention also provides a computer device, which can be a server, and its internal structure can be as shown in Figure 3As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store all data required for the process of the methanol coal blending algorithm based on the gated circulation unit. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the methanol coal blending algorithm based on the gated circulation unit is implemented.
[0087] Those skilled in the art will understand that Figure 3 The structure shown in is merely a block diagram of a portion of the structure related to the present application solution and does not constitute a limitation on the computer device to which the present application solution is applied.
[0088] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, any one of the above-mentioned methanol-coal blending algorithms based on a gated cycle unit is implemented.
[0089] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0090] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.
[0091] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. A methanol coal blending algorithm based on a gated cycle unit, characterized in that: include: Collecting historical actual production coal blending data provided by the methanol plant, and performing data preprocessing on the historical actual production coal blending data; The processed historical actual production coal blending data is labeled and divided into a training set and a validation set, so as to train a preset GRU model using the training set and the validation set to obtain a target model; The raw coal information is collected as the input of the target model, and after calculation by the target model, a coal type ratio scheme with the lowest input and output coal cost is output; wherein the raw coal information is inventory information or purchasable information.
2. The methanol coal blending algorithm based on gated cycle unit according to claim 1 is characterized in that: In the step of collecting the historical actual production coal blending data provided by the methanol plant and performing data preprocessing on the historical actual production coal blending data, the data preprocessing method is to process the missing values, that is, to process by deletion, interpolation or filling with the mean / median to ensure the uniformity of the data in format, unit and range.
3. The methanol coal blending algorithm based on gated cycle unit according to claim 1 is characterized in that: In the step of labeling the processed historical actual production coal blending data and dividing it into a training set and a validation set, so as to use the training set and the validation set to train the preset GRU model and obtain the target model, a coal blending expert labels the processed historical actual production coal blending data to identify the data with excellent performance in the historical actual production coal blending data; after cleaning and labeling the historical actual production coal blending data, a data set for training the preset GRU model is formed, and it is divided into a training set and a validation set in a ratio of 7:
3.
4. The methanol coal blending algorithm based on gated cycle unit according to claim 3 is characterized in that: The processed historical actual production coal blending data is labeled and divided into a training set and a validation set, and the preset GRU model is trained using the training set and the validation set to obtain the step of the target model. The preset GRU model has built-in update gates and reset gates so that the model can effectively control the retention of valid information. By analyzing the historical actual production coal blending data, effective features are extracted and features that have a great impact on the model performance are screened out to analyze the contribution of the features to the model prediction results.
5. The methanol coal blending algorithm based on gated cycle unit according to claim 4 is characterized in that: In the preset GRU model, according to the actual needs of the methanol plant coal blending work, in order to obtain the lowest cost ratio of coal entering the furnace, the model objective function is set to: minW=P T X+B Among them, W is the comprehensive cost of coal entering the furnace, X is the quality of a single type of coal, P T is the unit price of a single type of coal, and B is the adjustment parameter of the coal blending model, which includes the business experience of coal blending engineers, historical coal blending data learning experience and model self-learning experience.
6. The methanol coal blending algorithm based on gated cycle unit according to claim 5, characterized in that: In the preset GRU model, the update gate is responsible for evaluating the importance of the input information to determine the information to be integrated and passed to the next state, thereby ensuring that the network can effectively retain the core data. The specific formula of the update gate is: z t =σ(A (z) x t +U (z) h t―1 ) Among them, x t Indicates the current input, A (z) Its weight, h t―1 is the hidden information passed down from the previous node, U (z) is its weight, σ represents the activation function, and the final result is shown as z t ; The hidden information h of the current node t The update method is as follows. After the update, the input is input to the reset gate r t middle: h t =z t ⊙h t―1 +(1―z t )⊙h t r t =σ(A (r) x t +U (r) h t―1 ) Reset Gate t Calculate the amount of data to be deleted, and keep z except resetting unnecessary data t With h t The core factors affecting coal blending are extracted from the results of the calculation in the formula: h t =tanh(Ax t +r t ⊙Uh t―1 ) After completing the construction of the core algorithm of the above model, the target model is obtained by adjusting the hyperparameters, learning rate, and activation function, and after multiple iterative training of the training set data and verification on the verification set data.
7. The methanol coal blending algorithm based on gated cycle unit according to claim 1, characterized in that: After the step of collecting raw coal information as the input of the target model and outputting the coal type ratio scheme with the lowest input and output coal cost after calculation by the target model, the method further includes: The actual production data is collected and compared with the predicted data output by the target model, and the results are fed back to the target model to optimize and update the parameters of the target model.
8. A methanol coal blending device based on a gated circulation unit, characterized in that: include: A collection module, used to collect historical actual production coal blending data provided by the methanol plant, and perform data preprocessing on the historical actual production coal blending data; A labeling module, used for labeling the processed historical actual production coal blending data, and dividing it into a training set and a validation set, so as to train a preset GRU model using the training set and the validation set to obtain a target model; The output module is used to collect raw coal information as the input of the target model, and output the coal type ratio scheme with the lowest input and output coal cost after calculation by the target model; wherein the raw coal information is inventory information or purchasable information.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.