Control system for composite methanol evaporator

Abstract

A control system (100) for a composite methanol evaporator. First, temperature values at a plurality of predetermined time points within a predetermined time period that are collected by a plurality of temperature sensors deployed at a plurality of positions of a heat conduction pipe are acquired; then, temperature time-series features of the temperature values at the plurality of predetermined time points within the predetermined time period that are collected by the temperature sensors are extracted to obtain a plurality of temperature time-series feature vectors; next, spatial topological features of the plurality of positions are extracted to obtain a spatial topological feature matrix; and finally, on the basis of the plurality of temperature time-series feature vectors and the spatial topological feature matrix, whether to start an electric motor to drive an evaporation mechanism to rotate is determined. Thus, when to start an electric motor of an evaporation mechanism can be rationally determined, thereby reducing unnecessary energy consumption, and lowering production costs.

Description

A control system for a composite methanol evaporator Technical Field

[0001] The present application relates to the field of methanol evaporators, and more specifically, to a control system of a composite methanol evaporator. Background Art

[0002] Methanol is a colorless, transparent liquid with a pungent odor. It is one of the main raw materials for the production of formaldehyde. In the formaldehyde production system, methanol evaporates from liquid into gas, and then is used as one of the gases for preparing formaldehyde. Other mixed gases are added to formaldehyde.

[0003] In the existing methanol evaporator, the heat pipe cannot move during evaporation, making it difficult to change the contact surface between the heat pipe and the methanol. In addition, the heat on the surface of the heat pipe is required to evaporate the methanol, so the surface temperature of the heat pipe decreases after a long period of contact with the methanol, thereby affecting the practicality of the heat pipe, thereby reducing the evaporation efficiency of the methanol and increasing production costs.

[0004] To address this issue, patent CN113521773B provides a solution. This solution uses a motor to drive the evaporation mechanism to rotate during use, thereby raising the temperature. This prevents liquid methanol from continuously spraying onto one part of the evaporator, preventing the heat pipe from recovering in time. However, if the evaporation mechanism is activated too frequently, unnecessary energy consumption will increase. Therefore, determining when to activate the evaporation mechanism is a pressing technical issue in the control system. Technical issues

[0005] To address the above-mentioned technical problems, the present application is proposed. The embodiments of the present application provide a control system for a composite methanol evaporator. The control system can reasonably determine when to start the motor of the evaporation mechanism, thereby reducing unnecessary energy consumption and lowering production costs. Technical Solutions

[0006] According to one aspect of the present application, a control system for a composite methanol evaporator is provided, comprising:

[0007] a temperature value acquisition module, configured to acquire temperature values ​​at multiple predetermined time points within a predetermined time period collected by multiple temperature sensors deployed at multiple locations of the heat pipe;

[0008] a temperature time series feature extraction module, configured to extract temperature time series features of temperature values ​​at a plurality of predetermined time points within a predetermined time period collected by each temperature sensor to obtain a plurality of temperature time series feature vectors;

[0009] a spatial topological feature extraction module, configured to extract the spatial topological features of the plurality of positions to obtain a spatial topological feature matrix; and

[0010] The motor control module is configured to determine whether to start the motor to drive the evaporation mechanism to rotate based on the multiple temperature time series feature vectors and the spatial topology feature matrix. Beneficial effects

[0011] Compared to the prior art, the control system for the composite methanol evaporator provided in this application first obtains temperature values ​​at multiple predetermined time points within a predetermined time period, collected by multiple temperature sensors deployed at multiple locations on the heat pipe. Next, the control system extracts the temperature time series characteristics of the temperature values ​​collected by each temperature sensor at multiple predetermined time points within the predetermined time period to obtain multiple temperature time series feature vectors. Finally, the control system extracts the spatial topological features of the multiple locations to obtain a spatial topological feature matrix. Finally, based on the multiple temperature time series feature vectors and the spatial topological feature matrix, the control system determines whether to start the motor to drive the evaporation mechanism to rotate. This allows for a reasonable determination of when to start the evaporation mechanism's motor, reducing unnecessary energy consumption and lowering production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0013] FIG1 is a block diagram of a control system of a composite methanol evaporator according to an embodiment of the present application.

[0014] FIG2 is a block diagram of the temperature time series feature extraction module in the control system of the compound methanol evaporator according to an embodiment of the present application.

[0015] FIG3 is a block diagram of the spatial topological feature extraction module in the control system of the compound methanol evaporator according to an embodiment of the present application.

[0016] FIG4 is a block diagram of the motor control module in the control system of the compound methanol evaporator according to an embodiment of the present application.

[0017] FIG5 is a flow chart of a control method for a composite methanol evaporator according to an embodiment of the present application.

[0018] FIG6 is a schematic diagram of a system architecture of a control method for a composite methanol evaporator according to an embodiment of the present application.

[0019] FIG7 is an application scenario diagram of a control system of a composite methanol evaporator according to an embodiment of the present application. Modes for Carrying Out the Invention

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings. Obviously, the embodiments described 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 making creative efforts also fall within the scope of protection of this application.

[0021] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0022] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0023] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0024] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0025] To address the aforementioned technical issues, the present application's technical concept is to capture the time-series temperature information at each location on the heat pipe and, combined with the spatial topological distribution characteristics of each location, automatically determine whether to activate the motor to drive the evaporation mechanism. This approach rationally determines when to activate the evaporation mechanism's motor, reducing unnecessary energy consumption and lowering production costs.

[0026] Based on this, FIG1 is a block diagram of a control system for a compound methanol evaporator according to an embodiment of the present application. As shown in FIG1 , the control system 100 for a compound methanol evaporator according to an embodiment of the present application includes: a temperature value acquisition module 110 for acquiring temperature values ​​at multiple predetermined time points within a predetermined time period, collected by multiple temperature sensors deployed at multiple locations on the heat pipe; a temperature time series feature extraction module 120 for extracting temperature time series features of the temperature values ​​at multiple predetermined time points within the predetermined time period, collected by each temperature sensor, to obtain multiple temperature time series feature vectors; a spatial topology feature extraction module 130 for extracting spatial topology features of the multiple locations to obtain a spatial topology feature matrix; and a motor control module 140 for determining whether to start the motor to drive the evaporation mechanism to rotate based on the multiple temperature time series feature vectors and the spatial topology feature matrix.

[0027] Specifically, in the technical solution of the present application, first, temperature values ​​at multiple predetermined time points within a predetermined time period are acquired by multiple temperature sensors deployed at multiple positions of the heat pipe.

[0028] Next, the temperature time series characteristics of the temperature values ​​collected by each temperature sensor at multiple predetermined time points within a predetermined time period are extracted to obtain multiple temperature time series feature vectors. In other words, characteristic information about the temperature time series variation at each location is extracted. Specifically, this temperature time series variation characteristic information reflects the temperature variation at different locations where the heat pipe contacts methanol, providing information about the uniformity and stability of the contact between the heat pipe and methanol.

[0029] In a specific example of the present application, the encoding process of extracting the temperature time series features of the temperature values ​​at multiple predetermined time points within a predetermined time period collected by each temperature sensor to obtain multiple temperature time series feature vectors includes: first arranging the temperature values ​​at multiple predetermined time points within a predetermined time period collected by each temperature sensor into temperature time series input vectors according to the time dimension to obtain multiple temperature time series input vectors; and then passing the multiple temperature time series input vectors through a temperature time series feature extractor based on a one-dimensional convolutional layer to obtain multiple temperature time series feature vectors.

[0030] Next, the spatial topological features of these multiple locations are extracted to obtain a spatial topological feature matrix. Here, spatial topological relationships describe the spatial distances and interactions between different locations. Constructing a spatial topological matrix captures the correlations and spatial topological distribution characteristics between the heat pipes. These spatial topological features provide information about the heat conduction paths and heat exchange between the heat pipes, further revealing the overall characteristics of the heat pipe temperature distribution.

[0031] In a specific example of the present application, the encoding process of extracting the spatial topological features of the multiple positions to obtain a spatial topological feature matrix includes: first constructing a spatial topological matrix of the multiple positions, wherein the eigenvalue of each position on the non-diagonal position of the spatial topological matrix is ​​the spatial distance between the corresponding two positions; and then passing the spatial topological matrix through a spatial topological feature extractor based on a convolutional neural network model to obtain a spatial topological feature matrix.

[0032] Given that graph neural network models have the ability to process graph-structured data, they can model and learn the relationships between nodes. In the application scenario of this application, the locations of heat pipes can be regarded as nodes in the graph data, and the spatial relationships between them can be represented as edges in the graph data.

[0033] Specifically, in the technical solution of this application, the multiple temperature time-series feature vectors and the spatial topology feature matrix are passed through a graph neural network model to obtain a global temperature time-series feature matrix of the spatial topology heat pipe. In other words, by inputting the temperature time-series features and spatial topology features into the graph neural network model, the model can be used to learn the interactions and propagation patterns between nodes to capture the temperature time-series feature distribution information of the heat pipe, which contains spatial topology information.

[0034] Accordingly, as shown in Figure 2, the temperature time series feature extraction module 120 includes: a vectorization unit 121 for arranging the temperature values ​​collected by each temperature sensor at multiple predetermined time points within a predetermined time period into temperature time series input vectors according to the time dimension to obtain multiple temperature time series input vectors; and a time series feature extraction unit 122 for passing the multiple temperature time series input vectors through a temperature time series feature extractor based on a one-dimensional convolutional layer to obtain the multiple temperature time series feature vectors. It is worth noting that the one-dimensional convolutional layer is a neural network layer commonly used in deep learning for processing sequence data. It performs a sliding window operation along the time dimension and extracts features of the input sequence by learning a trainable convolution kernel (filter). The function of the one-dimensional convolutional layer in the temperature time series feature extraction module is to extract features from the multiple temperature time series input vectors. It can capture local patterns, trends, and correlation information in the input sequence and encode them into feature vectors. The input of the one-dimensional convolutional layer is a one-dimensional vector sequence, such as time series data. It performs a convolution operation on the input sequence through a sliding window, convolving the local features within the window with a trainable convolution kernel to obtain a new feature representation. This operation can be performed on the entire input sequence to extract the overall time series features. The output of the one-dimensional convolution layer can be multiple feature maps, each corresponding to a convolution kernel. Each feature map contains the feature representation of the input sequence at different positions and can capture patterns and association information at different scales. These feature maps can be further used for subsequent tasks such as classification, regression or other data analysis tasks. In other words, the role of the one-dimensional convolution layer in the temperature time series feature extraction module is to extract the features of multiple temperature time series input vectors through convolution operations for subsequent analysis and processing. It can help the model understand the patterns, trends and association information in the temperature series, thereby improving the understanding and prediction capabilities of temperature data.

[0035] Accordingly, as shown in Figure 3, the spatial topological feature extraction module 130 includes: a matrixing unit 131 for constructing a spatial topological matrix for the multiple locations, wherein the eigenvalue of each off-diagonal location in the spatial topological matrix represents the spatial distance between the corresponding two locations; and a topological feature extraction unit 132 for passing the spatial topological matrix through a spatial topological feature extractor based on a convolutional neural network model to obtain the spatial topological feature matrix. It is worth noting that the convolutional neural network (CNN) is a deep learning model particularly suitable for processing data with a grid structure. The core concept of a convolutional neural network is to extract local features from the input data through convolution operations, reduce the spatial dimension of the features through pooling operations, and ultimately input the extracted features into a fully connected layer for classification, regression, or other tasks. The role of the convolutional neural network model in the spatial topological feature extraction module is to extract spatial topological features by processing the spatial topological matrix. A spatial topological matrix is ​​a matrix that represents the spatial relationship between locations, wherein the eigenvalues ​​at off-diagonal locations represent the spatial distance between the corresponding two locations. The topological feature extraction unit processes the spatial topological matrix through a spatial topological feature extractor based on a convolutional neural network model to obtain a spatial topological feature matrix. In spatial topological feature extraction, the convolutional neural network model can learn local patterns and structural information in the input data. Through convolution operations, the model can capture the spatial relationships and local features between different positions, and gradually extract higher-level feature representations through the stacking of multiple convolutional layers and pooling layers. These feature representations can be used for subsequent tasks such as classification, segmentation, and target detection. In other words, the role of the convolutional neural network model in the spatial topological feature extraction module is to extract features from the spatial topological matrix through convolution operations, helping the model understand the spatial relationships and local structures between positions, thereby improving the ability to understand and analyze spatial data.

[0036] Then, the spatial topological heat pipe temperature time series global feature matrix is ​​passed through a classifier to obtain a classification result, and the classification result is used to indicate whether to start the motor to drive the evaporation mechanism to rotate.

[0037] Accordingly, as shown in FIG4 , the motor control module 140 includes: a joint encoding unit 141 for jointly encoding the multiple temperature time series feature vectors and the spatial topological feature matrix to obtain a spatial topological heat pipe temperature time series global feature matrix; and a start confirmation unit 142 for determining whether to start the motor to drive the evaporation mechanism to rotate based on the spatial topological heat pipe temperature time series global feature matrix.

[0038] More specifically, the joint encoding unit 141 is further configured to: apply the multiple temperature time series feature vectors and the spatial topological feature matrix to a graph neural network model to obtain the spatial topological heat pipe temperature time series global feature matrix. It is worth noting that a graph neural network (GNN) is a deep learning model for processing graph-structured data. It can learn and represent node and edge features, and capture the relationships and global structure between nodes through local connections and aggregation operations within the graph. The GNN model in the joint encoding unit takes the multiple temperature time series feature vectors and the spatial topological feature matrix as input and generates the spatial topological heat pipe temperature time series global feature matrix by learning the features of the nodes and edges in the graph structure. In a GNN, each node represents a sample or data point, and each edge represents a relationship or connection between nodes. The model iteratively updates node features, taking into account both the node's own features and those of its neighboring nodes. In this way, the node representation gradually incorporates local and global information, enabling a better understanding of the relationships and overall structure between nodes. A GNN model typically includes two main operations: node update and message passing. Node updates generate updated node representations by aggregating the node's own features and the features of neighboring nodes. Message passing is the process of transferring the node's representation information to neighboring nodes and performing aggregation operations on neighboring nodes. Through multiple rounds of node updates and message passing, the model can gradually extract higher-level graph feature representations. The advantage of the graph neural network model is that it can process data with complex topological structures, such as social networks, molecular structures, recommendation systems, etc. It can learn the relationships and global structures between nodes, thereby performing tasks such as classification, regression, and clustering on nodes. In this application, the graph neural network model is used to combine the temperature time series feature vectors and the spatial topological feature matrix to learn the relationships and global structures between nodes (positions) in the heat pipe system, and generate a global feature matrix of the heat pipe temperature time series, so that the temperature changes in the heat pipe system can be more comprehensively understood and analyzed.

[0039] More specifically, the start confirmation unit 142 is further configured to: pass the spatial topological heat pipe temperature time series global feature matrix through a classifier to obtain a classification result, and the classification result is used to indicate whether to start the motor to drive the evaporation mechanism to rotate.

[0040] More specifically, the startup confirmation unit 142 is further used to: expand the spatial topological heat pipe temperature time series global feature matrix into a classification feature vector according to a row vector or a column vector; use the fully connected layer of the classifier to fully connect encode the classification feature vector to obtain an encoded classification feature vector; and input the encoded classification feature vector into the Softmax classification function of the classifier to obtain the classification result.

[0041] That is, in the technical solution disclosed herein, the labels of the classifier include starting the motor to drive the evaporation mechanism to rotate (first label), and not starting the motor to drive the evaporation mechanism to rotate (second label), wherein the classifier determines which classification label the spatial topological heat pipe temperature time series global feature matrix belongs to by using a soft maximum function. It is worth noting that the first label p1 and the second label p2 here do not contain artificially set concepts. In fact, during the training process, the computer model does not have the concept of "whether to start the motor to drive the evaporation mechanism to rotate". It only has two classification labels and the probability of the output feature under these two classification labels, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether to start the motor to drive the evaporation mechanism to rotate is actually converted from the classification label to a binary classification class probability distribution that conforms to natural laws. In essence, what is used is the physical meaning of the natural probability distribution of the label, rather than the linguistic text meaning of "whether to start the motor to drive the evaporation mechanism to rotate".

[0042] It's easy to understand that a classifier uses given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are commonly used to solve binary classification problems. For multi-class classification, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-class classification. However, this approach is error-prone and inefficient. A commonly used multi-classification method is the Softmax classification function.

[0043] Furthermore, in the technical solution of the present application, the control system of the composite methanol evaporator is characterized in that it also includes a training module for training the temperature time series feature extractor based on the one-dimensional convolution layer, the spatial topology feature extractor based on the convolutional neural network model, the graph neural network model and the classifier.

[0044] Specifically, the training module includes: a training data acquisition unit for acquiring training data, wherein the training data includes training temperature values ​​at multiple predetermined time points within a predetermined time period collected by multiple temperature sensors deployed at multiple training positions of the heat pipe, and a real value of whether to start the motor to drive the evaporation mechanism to rotate; a training vectorization unit for arranging the training temperature values ​​at multiple predetermined time points within a predetermined time period collected by each temperature sensor into training temperature time series input vectors according to the time dimension to obtain multiple training temperature time series input vectors; a training temperature time series feature extraction unit for passing the multiple training temperature time series input vectors through the temperature time series feature extractor based on the one-dimensional convolution layer to obtain multiple training temperature time series feature vectors; a training matrixization unit for constructing a training space topology matrix for the multiple training positions, wherein the eigenvalue of each position on the non-diagonal position of the training space topology matrix is ​​the spatial distance between the corresponding two training positions; a training space topology feature extraction unit for extracting the training temperature values ​​at multiple predetermined time points within a predetermined time period collected by each temperature sensor A unit for passing the training spatial topology matrix through the spatial topology feature extractor based on the convolutional neural network model to obtain a training spatial topology feature matrix; a training graph neural network encoding unit for passing the multiple training temperature time series feature vectors and the training spatial topology feature matrix through the graph neural network model to obtain a training spatial topology heat pipe temperature time series global feature matrix; a training classification unit for passing the training spatial topology heat pipe temperature time series global feature matrix through a classifier to obtain a classification loss function value; and a loss training unit for training the temperature time series feature extractor based on the one-dimensional convolution layer, the spatial topology feature extractor based on the convolutional neural network model, the graph neural network model and the classifier with the classification loss function value, wherein in each round of the training iteration, the training spatial topology global temperature time series feature vector obtained after expanding the training spatial topology heat pipe temperature time series global feature matrix is ​​subjected to weight space exploration constraint iteration based on class matrix regularization.

[0045] In the technical solution of the present application, each training temperature time series feature vector represents the local time series correlation feature of the training temperature values ​​at each position of the heat pipe collected by the corresponding temperature sensor. Therefore, after passing it through the graph neural network model together with the training space topology feature matrix, topological correlation based on the spatial topology of multiple temperature sensors can be further performed. That is, the training space topology heat pipe temperature time series global feature matrix also includes the feature representation of the time-space mixed correlation dimension of the training temperature values ​​in the diverse time-space mixed space context. In this way, while improving the diversified feature distribution representation of the training space topology heat pipe temperature time series global feature matrix in the full time-space feature domain, it will also cause the training space topology heat pipe temperature time series global feature matrix to be enriched in the probability distribution domain of the classification result corresponding to the diversified feature distribution when classified by the classifier, thereby affecting the convergence effect of the weight matrix of the classifier during the classification process.

[0046] Based on this, when the applicant of this application classifies the training space topology heat pipe temperature time series global feature matrix through a classifier, at each iteration of the weight matrix, the training space topology global temperature time series feature vector obtained after expanding the training space topology heat pipe temperature time series global feature matrix is ​​subjected to weight space exploration constraints based on class matrix regularization.

[0047] Accordingly, in a specific example, in each round of iteration of the training, the training space topology global temperature time series feature vector obtained after the global feature matrix of the temperature time series of the training space topology heat pipe is expanded is subjected to weight space exploration constraint iteration based on class degree matrix regularization, including: performing weight space exploration constraint iteration based on class degree matrix regularization on the training space topology global temperature time series feature vector using the following optimization formula to obtain an optimized training space topology global temperature time series feature vector; wherein, the optimization formula is: in, is the global temperature time series feature vector of the training space topology, and is a column vector, represents matrix multiplication, represents the weight matrix of the previous iteration, represents the global temperature time series feature vector of the optimization training space topology, is a row vector, is a learnable domain transfer matrix, for example, it can be initially set to the weight matrix of the previous iteration The diagonal matrix of the diagonal elements of Represents the weight matrix after iteration.

[0048] Here, considering the weight space domain of the weight matrix and the training space topology global temperature time series feature vector The domain gap between the probability distribution domains of the classification results is calculated by the weight matrix Relative to the training space topology global temperature time series feature vector The regularized representation of the degree matrix is ​​used as an inter-domain transferring agent to transfer the probability distribution of valuable label constraints to the weight space. This prevents the rich labeled probability distribution domain from over-exploiting the weight distribution in the weight space during the classification process based on the weight space. This improves the convergence of the weight matrix and the training effect of the global feature matrix of the topological heat pipe temperature time series in the training space when classified by the classifier.

[0049] In summary, the control system 100 of the composite methanol evaporator according to the embodiment of the present application is explained, which can reasonably determine when to start the motor of the evaporation mechanism, thereby reducing unnecessary energy consumption and lowering production costs.

[0050] As described above, the control system 100 for the compound methanol evaporator according to the embodiment of the present application can be implemented in various terminal devices, such as a server equipped with a control algorithm for the compound methanol evaporator according to the embodiment of the present application. In one example, the control system 100 for the compound methanol evaporator according to the embodiment of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the control system 100 for the compound methanol evaporator according to the embodiment of the present application can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the control system 100 for the compound methanol evaporator according to the embodiment of the present application can also be one of the many hardware modules of the terminal device.

[0051] Alternatively, in another example, the control system 100 of the compound methanol evaporator based on the embodiment of the present application and the terminal device may also be separate devices, and the control system 100 of the compound methanol evaporator may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0052] FIG5 is a flow chart of a control method for a compound methanol evaporator according to an embodiment of the present application. FIG6 is a schematic diagram of a system architecture of a control method for a compound methanol evaporator according to an embodiment of the present application. As shown in FIG5 and FIG6, the control method for a compound methanol evaporator according to an embodiment of the present application includes: S110, obtaining temperature values ​​at multiple predetermined time points within a predetermined time period collected by multiple temperature sensors deployed at multiple positions of a heat pipe; S120, extracting temperature time series characteristics of the temperature values ​​at multiple predetermined time points within a predetermined time period collected by each temperature sensor to obtain multiple temperature time series characteristic vectors; S130, extracting spatial topological characteristics of the multiple positions to obtain a spatial topological characteristic matrix; and, S140, determining whether to start a motor to drive the evaporation mechanism to rotate based on the multiple temperature time series characteristic vectors and the spatial topological characteristic matrix.

[0053] In a specific example, in the control method of the above-mentioned composite methanol evaporator, the temperature time series characteristics of the temperature values ​​of multiple predetermined time points within a predetermined time period collected by each temperature sensor are extracted to obtain multiple temperature time series feature vectors, including: arranging the temperature values ​​of multiple predetermined time points within a predetermined time period collected by each temperature sensor into temperature time series input vectors according to the time dimension to obtain multiple temperature time series input vectors; and, passing the multiple temperature time series input vectors through a temperature time series feature extractor based on a one-dimensional convolutional layer to obtain the multiple temperature time series feature vectors.

[0054] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned control method of the compound methanol evaporator have been described in detail in the description of the control system 100 of the compound methanol evaporator with reference to Figures 1 to 4 above, and therefore, repeated description thereof will be omitted.

[0055] FIG7 is a diagram illustrating an application scenario of a control system for a composite methanol evaporator according to an embodiment of the present application. As shown in FIG7 , in this application scenario, temperature values ​​(e.g., D1 shown in FIG7 ) collected by multiple temperature sensors deployed at multiple locations on the heat pipe at multiple predetermined time points within a predetermined time period, as well as multiple locations of the multiple temperature sensors (e.g., D2 shown in FIG7 ), are first acquired. The behavior monitoring video is then input into a server (e.g., S shown in FIG7 ) equipped with a control algorithm for the composite methanol evaporator. The server then uses the control algorithm to process the temperature values ​​collected by each temperature sensor at multiple predetermined time points within the predetermined time period and the multiple locations of the multiple temperature sensors to obtain a classification result indicating whether to activate the motor to drive the evaporation mechanism to rotate.

[0056] Further, the composite methanol evaporator is described. In a specific embodiment of the present application, it includes a box body, a filter screen is provided below the inner wall of the box body, two layers of evaporation mechanisms are provided above the filter screen, the two layers of evaporation mechanisms are staggered and evenly distributed, an exhaust pipe is provided on the top inner wall of the box body and in the middle position, a plurality of groups of circular tubes are provided on the inner wall of the box body near the exhaust pipe, the two ends of the plurality of groups of circular tubes are fixedly connected to the two sides of the inner wall of the box body respectively, a plurality of groups of atomizing nozzles are provided below the circular tubes, the liquid inlet ends of the plurality of groups of atomizing nozzles are connected to the inner wall of the circular tubes, and the left side of the circular tubes passes through the inner wall of the box body and extends to the box body. A T-tube is fixedly connected to the outside of the box, and the other end of the T-tube is fixedly connected to a liquid pump, the lower end of the liquid pump is fixedly connected to a liquid outlet pipe on the side close to the box body, and the lower end of the liquid pump is fixedly connected to a liquid inlet pipe on the side away from the box body. A fixed plate is fixedly connected to the outer surface of the lower end of the fixed plate and extends to the bottom inner wall of the box body, and the other end of the liquid inlet pipe passes through the outer surface of the lower end of the fixed plate and extends to the bottom of the fixed plate. The front outer surface of the box body is fixedly connected to a protective cover, the right outer surface of the box body is fixedly connected to a fixing bracket, and the upper end of the fixing bracket is fixedly connected to a motor.

[0057] By adopting the above technical solution, when the user needs to evaporate liquid methanol, the liquid methanol is first pumped from the liquid inlet pipe into the T-tube through the liquid pump, and then diverted to each circular tube through the T-tube, and then the liquid methanol is sprayed on the evaporation mechanism in the form of mist through the atomizing nozzle, and finally the liquid methanol is evaporated into gas through the evaporation mechanism, and then the gas is collected into the collection device through the exhaust pipe. The liquid methanol that has not evaporated will be stored at the bottom of the box after filtering impurities through the filter net, and then the liquid methanol that has not evaporated will continue to be circulated and processed through the liquid pump. By setting a motor to drive the evaporation mechanism to rotate, the evaporation mechanism can be heated up during use, avoiding the liquid methanol being sprayed towards one place of the evaporation mechanism, resulting in the temperature of the heat transfer pipe not being able to recover in time and reducing the evaporation efficiency of the methanol.

[0058] As an embodiment of the present application, the evaporation mechanism includes a first fixed block, a second fixed block, a heat conduction pipe, a rotating shaft, a connecting rod, a first-stage gear, a second-stage gear, a first-stage pulley and a first-stage belt, the rear outer surface of the box body is fixedly connected to the first fixed block, the front outer surface of the box body is fixedly connected to the second fixed block, the first fixed block and the second fixed block are both provided with multiple grooves on one side of the box body, the multiple grooves are annularly distributed on one side of the fixed block, the first fixed block is provided with an air inlet hole on the side away from the box body, the other end of the air inlet hole is respectively connected to the inner wall of the groove, the inner wall of the groove is rotatably connected to the heat conduction pipe, one end of the heat conduction pipe close to the second fixed block is fixedly connected to the connecting rod, the other end of the connecting rod passes through the outer surface of the second fixed block and extends to the outside of the second fixed block and is fixedly connected to the first-stage gear, the outer surface of the connecting rod is rotatably connected to the inner wall of the second fixed block, and a through hole is provided on the side of the connecting rod away from the second fixed block, and the other end of the through hole is connected to the inner wall of the heat conduction pipe. By adopting the above technical solution, when the evaporation mechanism needs to work, water vapor is introduced into the heat pipe through the air inlet, and then the motor is started. The motor drives the primary pulley to rotate through the belt drive, the primary pulley drives the rotating shaft to rotate, the rotating shaft drives the secondary gear to rotate, and the secondary gear is engaged with the primary gear to drive the connecting rod to rotate. The connecting rod drives the heat pipe to rotate so that the contact surface between the heat pipe and the misted methanol is always changing, thereby ensuring that the heated surface of the heat pipe has a certain buffer time for the temperature to recover, thereby improving the evaporation efficiency of the misted methanol.

[0059] As an embodiment of the present application, a first bearing groove is provided on a side of the first fixed block close to the housing, a first bearing is fixedly connected to the inner wall of the first bearing groove, a rotating shaft is fixedly connected to the inner wall of the first bearing, a second bearing groove is provided on a side of the second fixed block close to the housing, a second bearing is fixedly connected to the inner wall of the second bearing groove, the inner wall of the second bearing is fixedly connected to the outer surface of the rotating shaft, the other end of the rotating shaft passes through the outer surface of the second fixed block and extends to the outside of the second fixed block and is fixedly connected to a secondary gear, a primary pulley is fixedly connected to the outer surface of the rotating shaft and is located on the right side of the secondary gear, and a primary belt is sleeved between the outer surfaces of two adjacent primary pulleys. By adopting the above technical solution, the wear of the rotating shaft can be reduced and the transmission efficiency can be enhanced by setting the first bearing and the second bearing.

[0060] As one embodiment of the present application, the inner wall of the heat pipe is provided with multiple barrier sheets, which are staggered and distributed on both sides of the heat pipe. By adopting the above technical solution, by providing multiple barrier sheets on the inner wall of the heat pipe, which are staggered and distributed on both sides of the heat pipe, the flow rate of water vapor in the heat pipe can be reduced, thereby improving the utilization rate of water vapor, reducing production costs, and conserving resources.

[0061] As one embodiment of the present application, a protective cover is fixedly connected to the outer surface of the housing and located outside the second fixed block. The protective cover has multiple annular grooves on the side away from the housing, each of which contains fan blades. The inner walls of the fan blades are fixedly connected to the outer surface of the rotating shaft. By adopting the above technical solution, the provision of a protective cover can reduce safety hazards and prevent users from being in danger during use. The provision of fan blades can quickly discharge exhausted water vapor from the protective cover, preventing water vapor from damaging components within the protective cover.

[0062] In one embodiment of the present application, a secondary pulley is fixedly connected to the outer surface of the motor's output shaft. A secondary belt is sleeved around the outer surface of both the secondary pulley and the outer surface of the primary pulley near the lower right corner of the housing. By employing this technical solution, the secondary pulley and secondary belt enable the motor to provide power to the evaporation mechanism during rotation.

[0063] In one embodiment of the present application, the heat pipe and the barrier are both made of aluminum, and the wall thickness of the heat pipe is 0.5-0.8 mm. By adopting the above technical solution, the heat exchange efficiency of the heat pipe can be improved by setting the heat pipe and the barrier to be both made of aluminum and the wall thickness of the heat pipe to be 0.5-0.8 mm.

[0064] As one embodiment of the present application, the primary gear and the secondary gear located on the right side of the same second fixed block are meshed with each other. By adopting the above technical solution, by setting the primary gear and the secondary gear located on the right side of the same second fixed block to mesh with each other, the heat pipes within the fixed block can be rotated simultaneously, ensuring that the heated surface of the heat pipes has a certain buffer time, allowing the heated surface to recover in temperature and improving the evaporation effect.

[0065] In one embodiment of the present application, an elastic sleeve is provided on the outer surface of the heat pipe. The outer surface of the elastic sleeve is slidably connected to the inner wall of the housing and the inner wall of the groove, respectively. The elastic sleeve is made of a wear-resistant material. By adopting this technical solution, the provision of the elastic sleeve can reduce wear on the outer surface of the heat pipe, thereby improving the utilization rate of the heat pipe.

[0066] As an embodiment of the present application, an opening and closing door is provided at the lower front end of the box, a handle is fixedly connected to the left side of the opening and closing door, and a sealing gasket is provided on the inner wall of the opening and closing door. By adopting the above technical solution, the opening and closing door facilitates cleaning of the interior of the box, the handle facilitates opening and closing of the opening and closing door, and the sealing gasket enhances the sealing of the box.

[0067] Working principle: When the user needs to evaporate liquid methanol, first pump the liquid methanol from the liquid inlet pipe into the T-tube through the liquid pump, then divert it to each round tube through the T-tube, and then spray the liquid methanol in the form of mist on the evaporation mechanism through the atomizing nozzle, and then introduce water vapor into the heat conduction pipe through the air inlet, and then start the motor, the output shaft of the motor drives the secondary pulley to rotate, the secondary pulley drives the primary pulley to rotate through the secondary belt, the primary pulley drives the rotating shaft to rotate, the rotating shaft drives the secondary gear to rotate, and then the secondary gear and the primary gear engage with each other to drive the connecting rod to rotate. The connecting rod drives the heat pipe to rotate so that the contact surface between the heat pipe and the misted methanol is always changing. The misted methanol is vaporized and converted into gas after being heated, and then the gas is collected into the collection device through the exhaust pipe. The liquid methanol that has not evaporated will be stored at the bottom of the box after filtering out impurities through the filter net, and then the unevaporated liquid methanol will continue to circulate through the liquid pump, thereby ensuring that the heated surface of the heat pipe has a certain buffer time, so that the heat pipe can be heated up in time during use, avoiding the liquid methanol from being sprayed towards one place of the heat pipe, resulting in the heat pipe temperature not being able to recover in time and reducing the evaporation efficiency of the methanol.

[0068] In the present application, when the user needs to evaporate liquid methanol, the liquid methanol is first pumped from the liquid inlet pipe into the T-tube through the liquid pump, and then diverted to each circular tube through the T-tube, and then the liquid methanol is sprayed on the evaporation mechanism in the form of mist through the atomizing nozzle, and then water vapor is introduced into the heat conduction tube through the air inlet hole, and then the motor is started, the output shaft of the motor drives the secondary pulley to rotate, the secondary pulley drives the primary pulley to rotate through the secondary belt, the primary pulley drives the rotating shaft to rotate, the rotating shaft drives the secondary gear to rotate, and then the secondary gear and the primary gear are engaged with each other to drive the connecting rod to rotate. The connecting rod drives the heat pipe to rotate so that the contact surface between the heat pipe and the misted methanol is always changing. The misted methanol is vaporized and converted into gas after being heated, and then the gas is collected into the collection device through the exhaust pipe. The liquid methanol that has not evaporated will be stored at the bottom of the box, and then the liquid methanol that has not evaporated will continue to circulate through the liquid pump, thereby ensuring that the heated surface of the heat pipe has a certain buffer time, so that the heat pipe can be heated up in time during use, avoiding the liquid methanol from being sprayed towards one place of the heat pipe, resulting in the heat pipe temperature not being able to recover in time and reducing the evaporation efficiency of the methanol, thereby reducing production costs.

[0069] This application uses specific terms to describe the embodiments of this application. For example, "first / second embodiment", "one embodiment", and / or "some embodiments" refer to a certain feature, structure, or characteristic related to at least one embodiment of this application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or multiple times in different places in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of this application may be appropriately combined.

[0070] In addition, it will be understood by those skilled in the art that various aspects of the present application can be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application can be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0071] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or highly formal sense, unless explicitly defined as such herein.

[0072] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application are described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all such modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A control system for a composite methanol evaporator, characterized in that: include: A temperature value acquisition module, used to acquire temperature values ​​at multiple predetermined time points within a predetermined time period collected by multiple temperature sensors deployed at multiple positions of the heat pipe; A temperature time series feature extraction module, used to extract temperature time series features of temperature values ​​at multiple predetermined time points within a predetermined time period collected by each temperature sensor to obtain multiple temperature time series feature vectors; A spatial topological feature extraction module, used to extract the spatial topological features of the plurality of positions to obtain a spatial topological feature matrix; as well as The motor control module is used to determine whether to start the motor to drive the evaporation mechanism to rotate based on the multiple temperature time series feature vectors and the spatial topological feature matrix.

2. The control system of the composite methanol evaporator according to claim 1, characterized in that: The temperature time series feature extraction module comprises: a vectorization unit, configured to arrange the temperature values ​​at a plurality of predetermined time points within a predetermined time period collected by each temperature sensor into temperature time series input vectors according to a time dimension to obtain a plurality of temperature time series input vectors; and The time series feature extraction unit is used to pass the multiple temperature time series input vectors through a temperature time series feature extractor based on a one-dimensional convolution layer to obtain the multiple temperature time series feature vectors.

3. The control system of the composite methanol evaporator according to claim 2, characterized in that: The spatial topological feature extraction module comprises: A matrixing unit, configured to construct a spatial topological matrix of the plurality of positions, wherein the eigenvalue of each position at a non-diagonal position of the spatial topological matrix is ​​a spatial distance between two corresponding positions; and A topological feature extraction unit is used to obtain the spatial topological feature matrix by passing the spatial topological matrix through a spatial topological feature extractor based on a convolutional neural network model.

4. The control system of the composite methanol evaporator according to claim 3, characterized in that: The motor control module comprises: A joint encoding unit, used for jointly encoding the multiple temperature time series feature vectors and the space topology feature matrix to obtain a space topology heat pipe temperature time series global feature matrix; and A start confirmation unit is used to determine whether to start the motor to drive the evaporation mechanism to rotate based on the spatial topological heat pipe temperature time series global characteristic matrix.

5. The control system of the composite methanol evaporator according to claim 4, characterized in that: The joint coding unit is further configured to: The multiple temperature time series feature vectors and the spatial topological feature matrix are passed through a graph neural network model to obtain the spatial topological heat pipe temperature time series global feature matrix.

6. The control system of the composite methanol evaporator according to claim 5, characterized in that: The startup confirmation unit is further used for: The spatial topological heat pipe temperature time series global feature matrix is ​​passed through a classifier to obtain a classification result, and the classification result is used to indicate whether to start the motor to drive the evaporation mechanism to rotate.

7. The control system of the composite methanol evaporator according to claim 6, characterized in that: It also includes a training module for training the temperature time series feature extractor based on the one-dimensional convolution layer, the spatial topology feature extractor based on the convolutional neural network model, the graph neural network model and the classifier; Wherein, the training module includes: a training data acquisition unit, configured to acquire training data, wherein the training data includes training temperature values ​​at multiple predetermined time points within a predetermined time period collected by multiple temperature sensors disposed at multiple training positions of the heat pipe, and a real value of whether to start the motor to drive the evaporation mechanism to rotate; A training vectorization unit, used for arranging the training temperature values ​​at a plurality of predetermined time points within a predetermined time period collected by each temperature sensor into training temperature time series input vectors according to the time dimension to obtain a plurality of training temperature time series input vectors; A training temperature time series feature extraction unit, used for respectively passing the plurality of training temperature time series input vectors through the temperature time series feature extractor based on the one-dimensional convolution layer to obtain a plurality of training temperature time series feature vectors; A training matrixing unit, used for constructing a training space topology matrix of the plurality of training positions, wherein the eigenvalue of each position on the non-diagonal position of the training space topology matrix is ​​the spatial distance between the corresponding two training positions; A training space topology feature extraction unit, used for passing the training space topology matrix through the space topology feature extractor based on the convolutional neural network model to obtain a training space topology feature matrix; A training graph neural network encoding unit is used to pass the multiple training temperature time series feature vectors and the training space topology feature matrix through the graph neural network model to obtain a training space topology heat pipe temperature time series global feature matrix; A training classification unit, used for passing the training spatial topological heat pipe temperature time series global feature matrix through a classifier to obtain a classification loss function value; and A loss training unit is used to train the temperature timing feature extractor based on the one-dimensional convolution layer, the spatial topology feature extractor based on the convolutional neural network model, the graph neural network model and the classifier with the classification loss function value, wherein in each round of iteration of the training, the training spatial topology global temperature timing feature vector obtained by expanding the training spatial topology heat pipe temperature timing global feature matrix is ​​subjected to weight space exploration constraint iteration based on class degree matrix regularization.

8. The control system of the compound methanol evaporator according to claim 7, characterized in that: In each round of iteration of the training, a weight space exploration constraint iteration based on class degree matrix regularization is performed on the training space topology global temperature time series feature vector obtained by expanding the training space topology heat pipe temperature time series global feature matrix, including: The weight space exploration constraint iteration based on class degree matrix regularization is performed on the training space topology global temperature time series feature vector using the following optimization formula to obtain an optimized training space topology global temperature time series feature vector; Wherein, the optimization formula is: in, is the global temperature time series feature vector of the training space topology, and is a column vector, represents matrix multiplication, represents the weight matrix of the previous iteration, represents the global temperature time series feature vector of the optimization training space topology, is a row vector, is the learnable domain transfer matrix, Represents the weight matrix after iteration.

Images