A vacuum pump cooling device control method, device and electronic equipment

Abstract

The present invention provides a control method for a vacuum pump cooling device, comprising: during the operation of the vacuum pump, collecting the first temperature distribution of the outer surface of the pump body through a temperature sensor; collecting the second temperature distribution of the cooling liquid pipeline of the intelligent valve and the valve opening distribution of the cooling injection pipeline; determining the opening control parameters of each intelligent valve based on the first temperature distribution, the second temperature distribution and the valve opening distribution; sending the opening control parameters to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameters. The intelligent valve adjusts its own opening according to the opening control parameters, and can control different openings for different heat areas, thereby improving the heat dissipation effect, thereby solving the problem that the existing traditional control method has poor heat dissipation effect and reduces the life of the vacuum pump.

Description

Technical Field

[0001] The present invention relates to the technical field of vacuum pumps, and in particular to a control method and device for a vacuum pump cooling device, and electronic equipment. Background Art

[0002] In recent years, my country's pump industry has maintained a state of booming production and sales and rapid development. A vacuum pump refers to a device or equipment that uses mechanical, physical, chemical or physicochemical methods to evacuate the container to obtain a vacuum. The vacuum pump generates a lot of heat during operation, causing the temperature of the pump body to rise rapidly. Therefore, when the vacuum pump is working, cooling it down becomes a critical problem that needs to be solved. The traditional control method uses a cold water pipe to cool the vacuum pump. However, due to the poor controllability of the existing cold water pipe, when the vacuum pump has uneven heat, the heat dissipation effect will be poor, resulting in a reduced life of the vacuum pump. Summary of the invention

[0003] The embodiment of the present invention provides a control method for a vacuum pump cooling device, aiming to solve the problem that the existing traditional control method has poor heat dissipation effect, resulting in a reduced life of the vacuum pump. During the operation of the vacuum pump, the first temperature distribution of the outer surface of the pump body is collected by a temperature sensor, as well as the second temperature distribution of the coolant pipeline of the intelligent valve and the valve opening distribution of the cooling injection pipeline, and the opening control parameters of each intelligent valve are determined according to the first temperature distribution, the second temperature distribution and the valve opening distribution, and the opening control parameters are sent to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameters, and can be controlled to different openings for different heat areas, thereby solving the problem that the existing traditional control method has poor heat dissipation effect, resulting in a reduced life of the vacuum pump.

[0004] In a first aspect, an embodiment of the present invention provides a control method for a vacuum pump cooling device, wherein the vacuum pump cooling device comprises a coolant pipeline arranged on the outer surface of a pump body of a vacuum pump and an intelligent valve arranged at each branch port in the coolant pipeline, wherein the intelligent valve is communicatively connected to a host computer, and the intelligent valve is used to collect the coolant temperature and control the valve opening. The control method for the vacuum pump cooling device is applied to the host computer, and the control method for the vacuum pump cooling device comprises the following steps:

[0005] During the operation of the vacuum pump, a first temperature distribution of the outer surface of the pump body is collected by a temperature sensor, and the temperature sensor array is arranged on the outer surface of the pump body; and a second temperature distribution of the coolant pipeline and a valve opening distribution of the cooling injection pipeline are collected by the smart valve, wherein the first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, and each of the first sampling points corresponds to a temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, and each of the second sampling points corresponds to a smart valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point;

[0006] Determining an opening control parameter of each of the smart valves based on the first temperature distribution, the second temperature distribution, and the valve opening distribution;

[0007] The opening control parameter is sent to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameter.

[0008] Optionally, the step of determining the opening control parameters of each of the intelligent valves based on the first temperature distribution, the second temperature distribution and the valve opening distribution includes:

[0009] Based on the first temperature distribution, the second temperature distribution, the valve opening distribution and the unfolded plane of the outer surface of the pump body, a first temperature matrix, a second temperature matrix and an opening matrix are constructed, wherein the first temperature matrix and the second temperature matrix have the same spatiotemporal resolution, and the opening matrix and the second temperature matrix have the same spatial resolution;

[0010] Based on the first temperature matrix, the second temperature matrix and the opening matrix, the opening control parameters of each of the smart valves are determined.

[0011] Optionally, the step of constructing a first temperature matrix, a second temperature matrix and an opening matrix based on the first temperature distribution, the second temperature distribution, the valve opening distribution and the unfolded plane of the outer surface of the pump body includes:

[0012] Performing a planar unfolding on the outer surface of the pump body to obtain an unfolded plane of the outer surface of the pump body;

[0013] Mapping the first sampling point of the first temperature distribution to the unfolded plane according to the position of the corresponding temperature sensor on the outer surface of the pump body, and mapping the second sampling point of the second temperature distribution to the unfolded plane according to the projection position of the corresponding smart valve on the outer surface of the pump body, to obtain a first mapping plane, wherein the first mapping plane includes the first sampling point and the second sampling point;

[0014] Performing Delaunay triangulation on the first mapping plane according to the sampling points in the first mapping plane to obtain a first mapping plane, wherein the second mapping plane includes a plurality of Delaunay triangles, and the circumscribed circle of each of the Delaunay triangles does not include any other sampling points in the face region where the Delaunay triangle is located;

[0015] Based on the second mapping plane, a first temperature matrix, a second temperature matrix and an opening matrix are constructed.

[0016] Optionally, the step of constructing a first temperature matrix, a second temperature matrix and an opening matrix based on the second mapping plane includes:

[0017] Determine the Delaunay triangle with the smallest sum of distances from its vertices to its centroid in the second mapping plane as the target Delaunay triangle;

[0018] determining a grid size of the second mapping plane based on the target Delaunay triangle;

[0019] Dividing the second mapping plane based on the grid size to obtain a third mapping plane, wherein the third mapping plane includes M×N grid units;

[0020] Mapping the first sampling point of the first temperature distribution and the corresponding first temperature value sequence into corresponding grid cells in the third mapping plane to obtain a first temperature matrix; and

[0021] Mapping the second sampling points of the second temperature distribution and the corresponding second temperature value sequence into corresponding grid cells in the third mapping plane to obtain a second temperature matrix; and

[0022] The third sampling point of the opening matrix and the corresponding valve opening value are mapped to the corresponding grid unit in the third mapping plane to obtain the opening matrix.

[0023] Optionally, the step of determining the opening control parameters of each of the intelligent valves based on the first temperature matrix, the second temperature matrix and the opening matrix includes:

[0024] Obtain the trained opening prediction model;

[0025] Based on the trained opening prediction model, prediction processing is performed on the first temperature matrix, the second temperature matrix and the opening matrix to obtain an opening prediction result;

[0026] The opening control parameters of each of the intelligent valves are determined based on the opening prediction result.

[0027] Optionally, the trained opening prediction model includes a first feature extraction module, a second feature extraction module, a third feature extraction module, a first fusion module, a second fusion module and an output module. The first feature extraction module, the second feature extraction module and the third feature extraction module are arranged in parallel. The input end of the first fusion module is connected to the output end of the first feature extraction module and the second feature extraction module at the same time. The input end of the second fusion module is connected to the output end of the first fusion module and the third feature extraction module at the same time. The input end of the output module is connected to the output end of the second fusion module. The output module includes M×N grid units, each output unit corresponds to an output of a predicted opening value. The step of performing prediction processing on the first temperature matrix, the second temperature matrix and the opening matrix based on the trained opening prediction model to obtain an opening prediction result includes:

[0028] Performing a first feature extraction process on the first temperature matrix by the first feature extraction module to obtain a first spatiotemporal feature of the first temperature matrix; and

[0029] Performing a second feature extraction process on the second temperature matrix by the second feature extraction module to obtain a second spatiotemporal feature of the second temperature matrix;

[0030] The first fusion module fuses the first spatiotemporal feature and the second spatiotemporal feature in the time dimension to obtain a first fused spatial feature, where the fused spatial feature includes a first eigenvalue in the spatial dimension;

[0031] Performing a third feature extraction process on the aperture matrix by the third feature extraction module to obtain a spatial feature of the aperture matrix, wherein the spatial feature includes a second eigenvalue in a spatial dimension;

[0032] The fused spatial feature and the spatial feature are fused in the spatial dimension by the second fusion module to obtain a fused feature;

[0033] The output module performs classification and regression processing on the fusion features to obtain an opening prediction result.

[0034] Optionally, the step of obtaining a trained opening prediction model includes:

[0035] Acquire a model to be trained and a training data set, wherein the model to be trained is a neural network model, and the training data set includes a plurality of four-tuple sample data, each of which includes a first sample temperature matrix, a second sample temperature matrix, a sample opening matrix, and a label matrix, wherein the first sample temperature matrix, the second sample temperature matrix, and the sample opening matrix are collected from the same sample vacuum pump cooling device and the sample vacuum pump under different working conditions or time periods, and the label matrix is ​​an ideal opening matrix annotated by an expert;

[0036] The model to be trained is subjected to supervised training using the training data set to obtain a trained opening prediction model.

[0037] In a second aspect, an embodiment of the present invention further provides a vacuum pump cooling device control device, the vacuum pump cooling device comprising a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each branch port in the coolant pipeline, the intelligent valve is communicatively connected to a host computer, the intelligent valve is used to collect the coolant temperature and control the valve opening, the vacuum pump cooling device control system is applied to the host computer, and the vacuum pump cooling device control device comprises:

[0038] A collection module, used for collecting a first temperature distribution of the outer surface of the pump body through a temperature sensor during the operation of the vacuum pump, wherein the temperature sensor array is arranged on the outer surface of the pump body; and a second temperature distribution of the coolant pipeline through the smart valve and a valve opening distribution of the cooling injection pipeline, wherein the first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, each of the first sampling points corresponds to one temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, each of the second sampling points corresponds to one smart valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point;

[0039] a determination module, configured to determine an opening control parameter of each of the intelligent valves based on the first temperature distribution, the second temperature distribution, and the valve opening distribution;

[0040] The adjustment module is used to send the opening control parameter to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameter.

[0041] In the third aspect, an embodiment of the present invention further provides a vacuum pump cooling device, which includes a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each branch port in the coolant pipeline, wherein the intelligent valve is communicatively connected to a host computer, and the intelligent valve is used to collect the coolant temperature and control the valve opening, and the vacuum pump cooling device is used to implement the steps in the vacuum pump cooling device control method provided in the embodiment of the present invention.

[0042] In a fourth aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps in the vacuum pump cooling device control method provided in the embodiment of the present invention are implemented.

[0043] In the embodiment of the present invention, during the operation of the vacuum pump, the first temperature distribution of the outer surface of the pump body is collected by the temperature sensor, as well as the second temperature distribution of the coolant pipeline of the intelligent valve and the valve opening distribution of the cooling injection pipeline, and the opening control parameters of each intelligent valve are determined according to the first temperature distribution, the second temperature distribution and the valve opening distribution, and the opening control parameters are sent to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameters. The present invention can determine the opening control parameters of each intelligent valve according to the first temperature distribution, the second temperature distribution and the valve opening distribution, and send the opening control parameters to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameters, and can control different openings for different heat areas, thereby solving the problem that the existing traditional control method has poor heat dissipation effect and reduces the life of the vacuum pump. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 is a flow chart of a vacuum pump cooling device control method provided by an embodiment of the present invention;

[0046] Figure 2 It is a structural schematic diagram of a vacuum pump cooling device control device provided by an embodiment of the present invention;

[0047] Figure 3 It is a structural schematic diagram of a control system of a vacuum pump cooling device provided by an embodiment of the present invention;

[0048] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] like Figure 1 As shown, Figure 1 1 is a flow chart of a vacuum pump cooling device control method provided by an embodiment of the present invention. The vacuum pump cooling device control method comprises the following steps:

[0051] 101. During the operation of the vacuum pump, the first temperature distribution of the outer surface of the pump body is collected through the temperature sensor; and the second temperature distribution of the coolant pipeline of the intelligent valve and the valve opening distribution of the cooling injection pipeline are collected.

[0052] In an embodiment of the present invention, the vacuum pump cooling device includes a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each diversion port in the coolant pipeline, the intelligent valve is connected to the host computer for communication, the intelligent valve is used to collect the coolant temperature and control the valve opening, and the vacuum pump cooling device control method is applied to the host computer. The coolant pipeline is a pipeline for transmitting coolant, which mainly transports the coolant to the outer surface of the vacuum pump that needs to be cooled to achieve a cooling effect. The intelligent valve can monitor the coolant temperature and control the valve opening in real time.

[0053] The first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, each first sampling point corresponds to a temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, each second sampling point corresponds to an intelligent valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point.

[0054] The temperature sensor is a device that can sense temperature and convert temperature information into a usable output signal. The temperature sensor is used to collect real-time temperature data on the surface of the pump body.

[0055] The above valve opening distribution can be understood as controlling the flow of coolant by adjusting the opening of the valve, so as to meet the cooling needs of different parts. For example, when the coolant temperature rises to a certain value, the smart valve will increase the opening to allow more coolant to enter the coolant pipeline, so that the pump surface cools down faster; when the coolant temperature is lower than a certain value, the smart valve will reduce the opening and reduce the flow of coolant.

[0056] 102. Determine an opening control parameter of each intelligent valve based on the first temperature distribution, the second temperature distribution, and the valve opening distribution.

[0057] In the embodiment of the present invention, the opening control parameters of each intelligent valve may be determined according to the first temperature distribution, the second temperature distribution, and the valve opening distribution.

[0058] The first temperature matrix, the second temperature matrix and the opening matrix can be constructed through the first temperature distribution, the second temperature distribution, the valve opening distribution and the unfolded plane of the outer surface of the pump body; and the opening control parameters of each intelligent valve can be determined based on the first temperature matrix, the second temperature matrix and the opening matrix.

[0059] The first temperature matrix and the second temperature matrix have the same temporal and spatial resolution, and the opening matrix and the second temperature matrix have the same spatial resolution. Each element in the first temperature matrix represents the temperature value of the corresponding position on the outer surface of the pump body at a certain point in time; and each element in the second temperature matrix represents the temperature value of the corresponding position of the coolant in the pipeline at a certain point in time. The first temperature matrix is ​​used to describe the temperature distribution on the surface of the pump body, and the second temperature matrix is ​​used to describe the temperature distribution of the coolant in the pipeline. Therefore, they need to have the same temporal and spatial resolution, so that they can accurately reflect the temperature changes at various positions on the outer surface of the pump body and at various positions of the coolant in the pipeline during the calculation process. Each element in the opening matrix represents the opening size of each diversion port valve in the pipeline. Therefore, the second temperature matrix and the opening matrix need to have the same spatial resolution, so that the coolant temperature at various positions in the pipeline and the opening size of each diversion port valve can be accurately reflected during the calculation process.

[0060] The above opening control parameters can be understood as the opening size of each diversion port valve that needs to be adjusted by the intelligent valve during the control process, which is used to control the flow and pressure of the coolant to ensure that the equipment operates at an appropriate temperature. By controlling the opening of each intelligent valve through the opening control parameters, the flow and pressure of the coolant can be effectively controlled to ensure that the equipment operates at an appropriate temperature.

[0061] 103. Send the opening control parameter to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameter.

[0062] In the embodiment of the present invention, the above-mentioned intelligent valve is used to collect the temperature of the coolant and control the valve opening. Each intelligent valve corresponds to a coolant pipeline diversion port.

[0063] The opening of each intelligent valve can be controlled by the opening control parameters, so that the coolant in the coolant pipeline cools and flushes the surface of the pump body, thereby reducing the temperature of the pump body surface and improving the cooling efficiency.

[0064] In the embodiment of the present invention, during the operation of the vacuum pump, the first temperature distribution of the outer surface of the pump body is collected by the temperature sensor, and the temperature sensor array is arranged on the outer surface of the pump body; through the second temperature distribution of the intelligent valve coolant pipeline and the valve opening distribution of the cooling injection pipeline, the first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, each first sampling point corresponds to a temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, each second sampling point corresponds to an intelligent valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point; based on the first temperature distribution, the second temperature distribution and the valve opening distribution, the opening control parameters of each intelligent valve are determined; the opening control parameters are sent to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameters. The present invention can determine the opening control parameters of each intelligent valve according to the first temperature distribution, the second temperature distribution and the valve opening distribution, and send the opening control parameters to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameters, and can control different openings for different heat areas, thereby solving the problem that the existing traditional control method has poor heat dissipation effect and reduces the life of the vacuum pump.

[0065] Optionally, in the step of determining the opening control parameters of each intelligent valve based on the first temperature distribution, the second temperature distribution and the valve opening distribution, a first temperature matrix, a second temperature matrix and an opening matrix can be constructed based on the first temperature distribution, the second temperature distribution, the valve opening distribution and the unfolded plane of the outer surface of the pump body, the first temperature matrix and the second temperature matrix have the same temporal and spatial resolution, and the opening matrix and the second temperature matrix have the same spatial resolution; based on the first temperature matrix, the second temperature matrix and the opening matrix, the opening control parameters of each intelligent valve are determined.

[0066] In the embodiment of the present invention, each element in the first temperature matrix represents the temperature value of the corresponding position on the outer surface of the pump body at a certain time point; and each element in the second temperature matrix represents the temperature value of the corresponding position of the coolant in the pipeline at a certain time point. The first temperature matrix is ​​used to describe the temperature distribution on the surface of the pump body, and the second temperature matrix is ​​used to describe the temperature distribution of the coolant in the pipeline. Therefore, they need to have the same temporal and spatial resolution so that the temperature changes at various positions on the outer surface of the pump body and at various positions of the coolant in the pipeline can be accurately reflected during the calculation process.

[0067] Each element in the opening matrix represents the opening size of each diversion valve in the pipeline. Therefore, the second temperature matrix and the opening matrix need to have the same spatial resolution so that the coolant temperature at each position in the pipeline and the opening size of each diversion valve can be accurately reflected during the calculation process.

[0068] The above opening control parameters can be understood as the opening size of each diversion port valve that needs to be adjusted by the intelligent valve during the control process, which is used to control the flow and pressure of the coolant to ensure that the equipment operates at an appropriate temperature. By controlling the opening of each intelligent valve through the opening control parameters, the flow and pressure of the coolant can be effectively controlled to ensure that the equipment operates at an appropriate temperature.

[0069] Optionally, in the step of constructing a first temperature matrix, a second temperature matrix and an opening matrix based on the first temperature distribution, the second temperature distribution, the valve opening distribution and the unfolded plane of the outer surface of the pump body, the outer surface of the pump body can be plane unfolded to obtain the unfolded plane of the outer surface of the pump body; the first sampling point of the first temperature distribution is mapped to the unfolded plane according to the position of the corresponding temperature sensor on the outer surface of the pump body, and the second sampling point of the second temperature distribution is mapped to the unfolded plane according to the projection position of the corresponding intelligent valve on the outer surface of the pump body to obtain a first mapping plane, and the first mapping plane includes the first sampling point and the second sampling point; the first mapping plane is Delaunay triangulated according to the sampling points in the first mapping plane to obtain a second mapping plane, and the second mapping plane includes multiple Delaunay triangles, and the circumscribed circle of each Delaunay triangle does not contain any other sampling points in the face area where the Delaunay triangle is located; based on the second mapping plane, the first temperature matrix, the second temperature matrix and the opening matrix are constructed.

[0070] In the embodiment of the present invention, the above-mentioned plane unfolding can be understood as a process of flattening a three-dimensional object into a two-dimensional plane image.

[0071] The above-mentioned first sampling point can be understood as the first sampling position of the first temperature distribution on the outer surface of the pump body.

[0072] The second sampling point mentioned above can be understood as a second sampling position of the second temperature distribution of the coolant pipeline.

[0073] The mapping process is to combine the location information and temperature data of the sampling points and unfold them on a plane. This allows for a more intuitive observation and analysis of the temperature distribution on the outer surface of the pump body and the valve opening at each diversion point.

[0074] The above-mentioned Delaunay triangulation is an algorithm for triangulating a planar point set. Delaunay triangulation connects all points with lines between the centers of the minimum circumscribed circles so that these lines do not intersect, forming a triangular mesh. Delaunay triangulation divides polygons into the minimum number of triangles for more accurate calculations while ensuring the quality and fidelity of the mesh.

[0075] It should be noted that, through Delaunay triangulation, the sampling points in the first mapping plane are connected to form a plurality of Delaunay triangles, and the circumscribed circle of each Delaunay triangle does not contain any other sampling points in the face area where the Delaunay triangle is located.

[0076] Optionally, in the step of constructing the first temperature matrix, the second temperature matrix and the opening matrix based on the second mapping plane, a Delaunay triangle with the smallest sum of distances from vertices to the center of mass can be determined in the second mapping plane as a target Delaunay triangle; the grid size of the second mapping plane is determined based on the target Delaunay triangle; the second mapping plane is divided based on the grid size to obtain a third mapping plane, and the third mapping plane includes M×N grid units; the first sampling point of the first temperature distribution and the corresponding first temperature value sequence are mapped to the corresponding grid cells in the third mapping plane to obtain the first temperature matrix; and the second sampling point of the second temperature distribution and the corresponding second temperature value sequence are mapped to the corresponding grid cells in the third mapping plane to obtain the second temperature matrix; and the third sampling point of the opening matrix and the corresponding valve opening value are mapped to the corresponding grid cells in the third mapping plane to obtain the opening matrix.

[0077] In an embodiment of the present invention, a Delaunay triangle with the smallest sum of distances from its vertices to its center is determined on the second mapping plane as a target Delaunay triangle; and the grid size on the second mapping plane is determined according to the band of the target Delaunay triangle; the second mapping plane is divided into M×N grid units according to the grid size on the second mapping plane, i.e., a third mapping plane; the sampling points of the first temperature distribution and the corresponding temperature value sequence are mapped to the corresponding grid units on the third mapping plane to obtain a matrix containing M×N elements, i.e., a first temperature matrix; the sampling points of the second temperature distribution and the corresponding temperature value sequence are mapped to the corresponding grid units on the third mapping plane to obtain a matrix containing M×N elements, i.e., a second temperature matrix; the sampling points of the opening matrix and the corresponding valve opening values ​​are mapped to the corresponding grid units on the third mapping plane to obtain a matrix containing M×N elements, i.e., an opening matrix.

[0078] The above grid size can be understood as the side length of the grid unit. The above temperature value sequence can be understood as a set of temperature values ​​arranged in chronological order.

[0079] Optionally, in the step of determining the opening control parameters of each intelligent valve based on the first temperature matrix, the second temperature matrix and the opening matrix, a trained opening prediction model can be obtained; the first temperature matrix, the second temperature matrix and the opening matrix are predicted based on the trained opening prediction model to obtain an opening prediction result; and the opening control parameters of each intelligent valve are determined based on the opening prediction result.

[0080] In the embodiment of the present invention, the above-mentioned trained opening prediction model can be a prediction model of machine learning or deep learning, and can be a recurrent neural network (RNN), a long short-term memory network (LSTM), etc. The above-mentioned trained opening prediction model is a trained opening prediction model obtained by training the prediction model through a training data set.

[0081] The first temperature matrix, the second temperature matrix and the opening matrix are predicted and processed by the trained opening prediction model to obtain the opening prediction result; and the opening control parameters of each intelligent valve are determined according to the opening prediction result.

[0082] The above-mentioned opening control parameters are the opening sizes of the valves of each diversion port that need to be adjusted by the intelligent valve during the control process, and are used to control the flow and pressure of the coolant to ensure that the equipment operates at an appropriate temperature.

[0083] Optionally, the trained opening prediction model includes a first feature extraction module, a second feature extraction module, a third feature extraction module, a first fusion module, a second fusion module and an output module. The first feature extraction module, the second feature extraction module and the third feature extraction module are arranged in parallel. The input end of the first fusion module is connected to the output end of the first feature extraction module and the second feature extraction module at the same time, the input end of the second fusion module is connected to the output end of the first fusion module and the third feature extraction module at the same time, the input end of the output module is connected to the output end of the second fusion module, the output module includes M×N output units, each output unit corresponds to an output of a predicted opening value, and in the step of performing prediction processing on the first temperature matrix, the second temperature matrix and the opening matrix based on the trained opening prediction model to obtain the opening prediction result, it can be The first feature extraction module performs a first feature extraction process on the first temperature matrix to obtain the first spatiotemporal feature of the first temperature matrix; and the second feature extraction module performs a second feature extraction process on the second temperature matrix to obtain the second spatiotemporal feature of the second temperature matrix; the first spatiotemporal feature and the second spatiotemporal feature are fused in the time dimension by the first fusion module to obtain the first fused spatial feature, and the fused spatial feature includes the first eigenvalue in the spatial dimension; the third feature extraction module performs a third feature extraction process on the opening matrix to obtain the spatial feature of the opening matrix, and the spatial feature includes the second eigenvalue in the spatial dimension; the second fusion module performs feature fusion on the fused spatial feature and the spatial feature in the spatial dimension to obtain the fused feature; the output module performs classification regression on the fused feature to obtain the opening prediction result.

[0084] In the embodiment of the present invention, the feature extraction module is used to extract useful features from the input data. The fusion module is used to fuse the features from different feature extraction modules to better predict the opening value. The output module includes M×N output units, each output unit corresponds to an output of a predicted opening value, and the output module is used to output M×N different opening values.

[0085] The above-mentioned spatiotemporal characteristics can be understood as characteristics in two dimensions of time and space. The spatiotemporal characteristics reflect the changing patterns of the temperature matrix over time and at different spatial positions.

[0086] The above feature fusion can be understood as the process of combining multiple feature vectors in a certain way to form a more powerful feature vector.

[0087] The above-mentioned first fusion spatial feature can be understood as the change of the first spatiotemporal feature and the second spatiotemporal feature in the spatial dimension.

[0088] The above classification and regression processing can be understood as a process of predicting the value or category of the output by analyzing the features of the input.

[0089] Optionally, in the step of obtaining a trained opening prediction model, a model to be trained and a training data set can be obtained. The model to be trained is a neural network model. The training data set includes multiple four-tuple sample data. Each four-tuple sample data includes a first sample temperature matrix, a second sample temperature matrix, a sample opening matrix and a label matrix. The first sample temperature matrix, the second sample temperature matrix and the sample opening matrix are collected from the same sample vacuum pump cooling device and the sample vacuum pump under different working conditions or time periods. The label matrix is ​​an ideal opening matrix annotated by experts. The model to be trained is supervisedly trained using the training data set to obtain a trained opening prediction model.

[0090] In the embodiment of the present invention, the neural network model is a complex network system formed by a large number of simple processing units (called neurons) widely connected to each other, reflecting many basic characteristics of human brain functions and is a highly complex nonlinear dynamic learning system. The neural network model can be a recurrent neural network (RNN), a long short-term memory network (LSTM), a convolutional neural network (CNN), etc.

[0091] The above supervised training can be understood as guiding the learning process of the model by using labeled training data. In supervised learning, each training sample has a known label or output value, and the goal of the model is to build a prediction model by learning the relationship between the input variables and the output variables.

[0092] The first sample temperature matrix can be understood as the temperature matrix of the outer surface of the sample pump body; the second sample temperature matrix can be understood as the temperature matrix of the sample coolant pipeline; the sample opening matrix can be understood as the valve opening matrix of the cooling injection pipeline. The first sample temperature matrix, the second sample temperature matrix and the sample opening matrix are collected from the same sample vacuum pump cooling device and the sample vacuum pump under different working conditions or time periods, and the label matrix is ​​the ideal opening matrix marked by experts.

[0093] The training model is supervisedly trained by the training data set, and after the training is completed, a trained opening prediction model can be obtained. The first temperature matrix, the second temperature matrix and the opening matrix can be predicted by the trained opening prediction model to obtain an opening prediction result.

[0094] like Figure 2As shown, an embodiment of the present invention provides a vacuum pump cooling device control device, the vacuum pump cooling device includes a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each branch port in the coolant pipeline, the intelligent valve is connected to the host computer for communication, the intelligent valve is used to collect the coolant temperature and control the valve opening, and the vacuum pump cooling device control device is applied to the host computer. The vacuum pump cooling device control device includes:

[0095] The acquisition module 201 is used to acquire, during the operation of the vacuum pump, a first temperature distribution of the outer surface of the pump body through a temperature sensor, wherein the temperature sensor array is arranged on the outer surface of the pump body; and a second temperature distribution of the coolant pipeline through the smart valve and a valve opening distribution of the cooling injection pipeline, wherein the first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, each of the first sampling points corresponds to one temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, each of the second sampling points corresponds to one smart valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point;

[0096] A determination module 202, configured to determine an opening control parameter of each of the intelligent valves based on the first temperature distribution, the second temperature distribution, and the valve opening distribution;

[0097] The adjustment module 203 is used to send the opening control parameter to the corresponding smart valve, so that the smart valve adjusts its own opening according to the opening control parameter.

[0098] Optionally, the steps of the determination module 202 include:

[0099] A construction submodule, for constructing a first temperature matrix, a second temperature matrix and an opening matrix based on the first temperature distribution, the second temperature distribution, the valve opening distribution and the unfolded plane of the outer surface of the pump body, wherein the first temperature matrix and the second temperature matrix have the same spatiotemporal resolution, and the opening matrix and the second temperature matrix have the same spatial resolution;

[0100] A determination submodule is used to determine the opening control parameters of each of the intelligent valves based on the first temperature matrix, the second temperature matrix and the opening matrix.

[0101] Optionally, the construction submodule includes:

[0102] A plane unfolding unit, used for plane unfolding the outer surface of the pump body to obtain the unfolded plane of the outer surface of the pump body;

[0103] a mapping unit, configured to map a first sampling point of the first temperature distribution to the unfolded plane according to a position of a corresponding temperature sensor on the outer surface of the pump body, and map a second sampling point of the second temperature distribution to the unfolded plane according to a projection position of a corresponding smart valve on the outer surface of the pump body, to obtain a first mapping plane, wherein the first mapping plane includes the first sampling point and the second sampling point;

[0104] a subdivision unit, configured to perform Delaunay triangulation on the first mapping plane according to the sampling points in the first mapping plane to obtain a first mapping plane, wherein the second mapping plane includes a plurality of Delaunay triangles, and the circumscribed circle of each of the Delaunay triangles does not contain any other sampling points in the face region where the Delaunay triangle is located;

[0105] A construction unit is used to construct a first temperature matrix, a second temperature matrix and an opening matrix based on the second mapping plane.

[0106] Optionally, the construction unit includes:

[0107] A first determining subunit, configured to determine, in the second mapping plane, the Delaunay triangle having the smallest sum of distances from its vertices to its centroid as a target Delaunay triangle;

[0108] A second determining subunit, configured to determine a grid size of the second mapping plane based on the target Delaunay triangle;

[0109] a dividing subunit, configured to divide the second mapping plane based on the grid size to obtain a third mapping plane, wherein the third mapping plane includes M×N grid units;

[0110] A first mapping subunit is used to map the first sampling point of the first temperature distribution and the corresponding first temperature value sequence to the corresponding grid unit in the third mapping plane to obtain a first temperature matrix; and

[0111] A second mapping subunit is used to map the second sampling points of the second temperature distribution and the corresponding second temperature value sequence to corresponding grid cells in the third mapping plane to obtain a second temperature matrix; and

[0112] The third mapping subunit is used to map the third sampling point of the opening matrix and the corresponding valve opening value to the corresponding grid unit in the third mapping plane to obtain the opening matrix.

[0113] Optionally, the determining submodule includes:

[0114] An acquisition unit, used for acquiring a trained opening prediction model;

[0115] A processing unit, configured to perform prediction processing on the first temperature matrix, the second temperature matrix and the opening matrix based on the trained opening prediction model to obtain an opening prediction result;

[0116] A determination unit is used to determine the opening control parameters of each of the intelligent valves based on the opening prediction result.

[0117] Optionally, the processing unit includes:

[0118] a first processing subunit, configured to perform a first feature extraction process on the first temperature matrix through the first feature extraction module to obtain a first spatiotemporal feature of the first temperature matrix; and

[0119] A second processing subunit, configured to perform a second feature extraction process on the second temperature matrix through the second feature extraction module to obtain a second spatiotemporal feature of the second temperature matrix;

[0120] A first feature fusion subunit is used to perform feature fusion on the first spatiotemporal feature and the second spatiotemporal feature in the time dimension through the first fusion module to obtain a first fused spatial feature, wherein the fused spatial feature includes a first eigenvalue in the spatial dimension;

[0121] A third processing subunit is used to perform a third feature extraction process on the aperture matrix through the third feature extraction module to obtain a spatial feature of the aperture matrix, wherein the spatial feature includes a second eigenvalue in a spatial dimension;

[0122] A second feature fusion subunit is used to perform feature fusion on the fused spatial feature and the spatial feature in the spatial dimension through the second fusion module to obtain a fused feature;

[0123] The fourth processing subunit performs classification regression processing on the fusion features through the output module to obtain an opening prediction result.

[0124] Optionally, the acquiring unit includes:

[0125] An acquisition subunit is used to acquire a model to be trained and a training data set, wherein the model to be trained is a neural network model, and the training data set includes a plurality of four-tuple sample data, each of which includes a first sample temperature matrix, a second sample temperature matrix, a sample opening matrix, and a label matrix, wherein the first sample temperature matrix, the second sample temperature matrix, and the sample opening matrix are collected from the same sample vacuum pump cooling device and the sample vacuum pump under different working conditions or time periods, and the label matrix is ​​an ideal opening matrix annotated by an expert;

[0126] The training subunit is used to perform supervised training on the model to be trained through the training data set to obtain a trained opening prediction model.

[0127] It should be noted that the vacuum pump cooling device control device provided in the embodiment of the present invention can be applied to a host computer, a single-chip microcomputer, a smart phone, a computer, a server and other equipment that can perform a vacuum pump cooling device control method.

[0128] The vacuum pump cooling device control device provided in the embodiment of the present invention can implement each process implemented by the vacuum pump cooling device control method in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0129] See also Figure 3 , Figure 3 A vacuum pump cooling device is provided in an embodiment of the present invention. The vacuum pump cooling device includes a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each branch port in the coolant pipeline. The intelligent valve is communicatively connected to a host computer. The intelligent valve is used to collect the coolant temperature and control the valve opening. The vacuum pump cooling device is used to implement the steps in the vacuum pump cooling device control method provided in an embodiment of the present invention.

[0130] See also Figure 4 , Figure 4 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, it includes: a memory 402, a processor 401, and a computer program of a vacuum pump cooling device control method stored in the memory 402 and executable on the processor 401, wherein:

[0131] The processor 401 is used to call the computer program stored in the memory 402 to perform the following steps:

[0132] During the operation of the vacuum pump, a first temperature distribution of the outer surface of the pump body is collected by a temperature sensor, and the temperature sensor array is arranged on the outer surface of the pump body; and a second temperature distribution of the coolant pipeline and a valve opening distribution of the cooling injection pipeline are collected by the smart valve, wherein the first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, and each of the first sampling points corresponds to a temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, and each of the second sampling points corresponds to a smart valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point;

[0133] Determining an opening control parameter of each of the smart valves based on the first temperature distribution, the second temperature distribution, and the valve opening distribution;

[0134] The opening control parameter is sent to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameter.

[0135] Optionally, the step of determining the opening control parameters of each of the smart valves based on the first temperature distribution, the second temperature distribution, and the valve opening distribution, performed by the processor 401, includes:

[0136] Based on the first temperature distribution, the second temperature distribution, the valve opening distribution and the unfolded plane of the outer surface of the pump body, a first temperature matrix, a second temperature matrix and an opening matrix are constructed, wherein the first temperature matrix and the second temperature matrix have the same spatiotemporal resolution, and the opening matrix and the second temperature matrix have the same spatial resolution;

[0137] Based on the first temperature matrix, the second temperature matrix and the opening matrix, the opening control parameters of each of the smart valves are determined.

[0138] Optionally, the step of constructing a first temperature matrix, a second temperature matrix, and an opening matrix based on the first temperature distribution, the second temperature distribution, the valve opening distribution, and the unfolded plane of the outer surface of the pump body performed by the processor 401 includes:

[0139] Performing a planar unfolding on the outer surface of the pump body to obtain an unfolded plane of the outer surface of the pump body;

[0140] Mapping the first sampling point of the first temperature distribution to the unfolded plane according to the position of the corresponding temperature sensor on the outer surface of the pump body, and mapping the second sampling point of the second temperature distribution to the unfolded plane according to the projection position of the corresponding smart valve on the outer surface of the pump body, to obtain a first mapping plane, wherein the first mapping plane includes the first sampling point and the second sampling point;

[0141] Performing Delaunay triangulation on the first mapping plane according to the sampling points in the first mapping plane to obtain a first mapping plane, wherein the second mapping plane includes a plurality of Delaunay triangles, and the circumscribed circle of each of the Delaunay triangles does not include any other sampling points in the face region where the Delaunay triangle is located;

[0142] Based on the second mapping plane, a first temperature matrix, a second temperature matrix and an opening matrix are constructed.

[0143] Optionally, the step of constructing a first temperature matrix, a second temperature matrix, and an opening matrix based on the second mapping plane performed by the processor 401 includes:

[0144] Determine the Delaunay triangle with the smallest sum of distances from its vertices to its centroid in the second mapping plane as the target Delaunay triangle;

[0145] determining a grid size of the second mapping plane based on the target Delaunay triangle;

[0146] Dividing the second mapping plane based on the grid size to obtain a third mapping plane, wherein the third mapping plane includes M×N grid units;

[0147] Mapping the first sampling point of the first temperature distribution and the corresponding first temperature value sequence into corresponding grid cells in the third mapping plane to obtain a first temperature matrix; and

[0148] Mapping the second sampling points of the second temperature distribution and the corresponding second temperature value sequence into corresponding grid cells in the third mapping plane to obtain a second temperature matrix; and

[0149] The third sampling point of the opening matrix and the corresponding valve opening value are mapped to the corresponding grid unit in the third mapping plane to obtain the opening matrix.

[0150] Optionally, the step of determining the opening control parameters of each of the smart valves based on the first temperature matrix, the second temperature matrix and the opening matrix performed by the processor 401 includes:

[0151] Obtain the trained opening prediction model;

[0152] Based on the trained opening prediction model, prediction processing is performed on the first temperature matrix, the second temperature matrix and the opening matrix to obtain an opening prediction result;

[0153] The opening control parameters of each of the intelligent valves are determined based on the opening prediction result.

[0154] Optionally, the trained opening prediction model includes a first feature extraction module, a second feature extraction module, a third feature extraction module, a first fusion module, a second fusion module and an output module. The first feature extraction module, the second feature extraction module and the third feature extraction module are arranged in parallel. The input end of the first fusion module is connected to the output end of the first feature extraction module and the second feature extraction module at the same time. The input end of the second fusion module is connected to the output end of the first fusion module and the third feature extraction module at the same time. The input end of the output module is connected to the output end of the second fusion module. The output module includes M×N grid units, each output unit corresponds to an output of a predicted opening value. The processor 401 performs prediction processing on the first temperature matrix, the second temperature matrix and the opening matrix based on the trained opening prediction model to obtain the opening prediction result, which includes:

[0155] Performing a first feature extraction process on the first temperature matrix by the first feature extraction module to obtain a first spatiotemporal feature of the first temperature matrix; and

[0156] Performing a second feature extraction process on the second temperature matrix by the second feature extraction module to obtain a second spatiotemporal feature of the second temperature matrix;

[0157] The first fusion module fuses the first spatiotemporal feature and the second spatiotemporal feature in the time dimension to obtain a first fused spatial feature, where the fused spatial feature includes a first eigenvalue in the spatial dimension;

[0158] Performing a third feature extraction process on the aperture matrix by the third feature extraction module to obtain a spatial feature of the aperture matrix, wherein the spatial feature includes a second eigenvalue in a spatial dimension;

[0159] The fused spatial feature and the spatial feature are fused in the spatial dimension by the second fusion module to obtain a fused feature;

[0160] The output module performs classification and regression processing on the fusion features to obtain an opening prediction result.

[0161] Optionally, the step of acquiring a trained opening prediction model performed by the processor 401 includes:

[0162] Acquire a model to be trained and a training data set, wherein the model to be trained is a neural network model, and the training data set includes a plurality of four-tuple sample data, each of which includes a first sample temperature matrix, a second sample temperature matrix, a sample opening matrix, and a label matrix, wherein the first sample temperature matrix, the second sample temperature matrix, and the sample opening matrix are collected from the same sample vacuum pump cooling device and the sample vacuum pump under different working conditions or time periods, and the label matrix is ​​an ideal opening matrix annotated by an expert;

[0163] The model to be trained is subjected to supervised training using the training data set to obtain a trained opening prediction model.

[0164] It should be noted that the electronic device provided in the embodiment of the present invention can be applied to smart phones, computers, servers and other devices that can perform the vacuum pump cooling device control method.

[0165] The electronic device provided in the embodiment of the present invention can implement each process implemented by the vacuum pump cooling device control method in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0166] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A vacuum pump cooling device control method, characterized in that: The vacuum pump cooling device comprises a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each diversion port in the coolant pipeline, wherein the intelligent valve is connected to the host computer for communication, and the intelligent valve is used to collect the coolant temperature and control the valve opening. The vacuum pump cooling device control method is applied to the host computer, and the method comprises the following steps: During the operation of the vacuum pump, a first temperature distribution of the outer surface of the pump body is collected by a temperature sensor, and the temperature sensor array is arranged on the outer surface of the pump body; and a second temperature distribution of the coolant pipeline and a valve opening distribution of the coolant pipeline are collected by the intelligent valve, wherein the first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, each of the first sampling points corresponds to a temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, each of the second sampling points corresponds to a intelligent valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point; Performing a planar unfolding on the outer surface of the pump body to obtain an unfolded plane of the outer surface of the pump body; Mapping the first sampling point of the first temperature distribution to the unfolded plane according to the position of the corresponding temperature sensor on the outer surface of the pump body, and mapping the second sampling point of the second temperature distribution to the unfolded plane according to the projection position of the corresponding smart valve on the outer surface of the pump body, to obtain a first mapping plane, wherein the first mapping plane includes the first sampling point and the second sampling point; Performing Delaunay triangulation on the first mapping plane according to the sampling points in the first mapping plane to obtain a second mapping plane, wherein the second mapping plane includes a plurality of Delaunay triangles, and the circumscribed circle of each of the Delaunay triangles does not include any other sampling points in the face region where the Delaunay triangle is located; Based on the second mapping plane, a first temperature matrix, a second temperature matrix and an aperture matrix are constructed, wherein the first temperature matrix and the second temperature matrix have the same temporal and spatial resolution, and the aperture matrix and the second temperature matrix have the same spatial resolution; Determining an opening control parameter of each of the smart valves based on the first temperature matrix, the second temperature matrix, and the opening matrix; The opening control parameter is sent to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameter.

2. The vacuum pump cooling device control method according to claim 1, characterized in that: The step of constructing a first temperature matrix, a second temperature matrix and an opening matrix based on the second mapping plane includes: Determine the Delaunay triangle with the smallest sum of distances from its vertices to its centroid in the second mapping plane as the target Delaunay triangle; determining a grid size of the second mapping plane based on the target Delaunay triangle; Dividing the second mapping plane based on the grid size to obtain a third mapping plane, wherein the third mapping plane includes M×N grid units; Mapping the first sampling point of the first temperature distribution and the corresponding first temperature sequence into corresponding grid cells in the third mapping plane to obtain a first temperature matrix; and Mapping the second sampling points of the second temperature distribution and the corresponding second temperature sequence into corresponding grid cells in the third mapping plane to obtain a second temperature matrix; and The third sampling point of the opening matrix and the corresponding valve opening value are mapped to the corresponding grid unit in the third mapping plane to obtain the opening matrix.

3. The vacuum pump cooling device control method according to claim 1 or 2, characterized in that: The step of determining the opening control parameters of each of the intelligent valves based on the first temperature matrix, the second temperature matrix and the opening matrix comprises: Obtain the trained opening prediction model; Based on the trained opening prediction model, prediction processing is performed on the first temperature matrix, the second temperature matrix and the opening matrix to obtain an opening prediction result; The opening control parameters of each of the intelligent valves are determined based on the opening prediction result.

4. The vacuum pump cooling device control method according to claim 3, characterized in that: The trained opening prediction model includes a first feature extraction module, a second feature extraction module, a third feature extraction module, a first fusion module, a second fusion module and an output module. The first feature extraction module, the second feature extraction module and the third feature extraction module are arranged in parallel. The input end of the first fusion module is connected to the output ends of the first feature extraction module and the second feature extraction module at the same time. The input end of the second fusion module is connected to the output ends of the first fusion module and the third feature extraction module at the same time. The input end of the output module is connected to the output end of the second fusion module. The output module includes M×N output units, each output unit corresponds to an output of a predicted opening value. The step of performing prediction processing on the first temperature matrix, the second temperature matrix and the opening matrix based on the trained opening prediction model to obtain the opening prediction result includes: Performing a first feature extraction process on the first temperature matrix by the first feature extraction module to obtain a first spatiotemporal feature of the first temperature matrix; and Performing a second feature extraction process on the second temperature matrix by the second feature extraction module to obtain a second spatiotemporal feature of the second temperature matrix; The first fusion module fuses the first spatiotemporal feature and the second spatiotemporal feature in the time dimension to obtain a first fused spatial feature, where the fused spatial feature includes a first eigenvalue in the spatial dimension; Performing a third feature extraction process on the aperture matrix by the third feature extraction module to obtain a spatial feature of the aperture matrix, wherein the spatial feature includes a second eigenvalue in a spatial dimension; The fused spatial feature and the spatial feature are fused in the spatial dimension by the second fusion module to obtain a fused feature; The output module performs classification and regression processing on the fusion features to obtain an opening prediction result.

5. The vacuum pump cooling device control method according to claim 4, characterized in that: The step of obtaining the trained opening prediction model comprises: Acquire a model to be trained and a training data set, wherein the model to be trained is a neural network model, and the training data set includes a plurality of four-tuple sample data, each of which includes a first sample temperature matrix, a second sample temperature matrix, a sample opening matrix, and a label matrix, wherein the first sample temperature matrix, the second sample temperature matrix, and the sample opening matrix are collected from the same sample vacuum pump cooling device and the sample vacuum pump under different working conditions or time periods, and the label matrix is ​​an ideal opening matrix annotated by an expert; The model to be trained is subjected to supervised training using the training data set to obtain a trained opening prediction model.

6. A vacuum pump cooling device control device, characterized in that: The vacuum pump cooling device comprises a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each diversion port in the coolant pipeline, the intelligent valve is connected to the host computer for communication, the intelligent valve is used to collect the coolant temperature and control the valve opening, the vacuum pump cooling device control device is applied to the host computer, and the vacuum pump cooling device control system comprises: A collection module, used for collecting a first temperature distribution of the outer surface of the pump body through a temperature sensor during the operation of the vacuum pump, wherein the temperature sensor array is arranged on the outer surface of the pump body; and collecting a second temperature distribution of the coolant pipeline and a valve opening distribution of the coolant pipeline through the intelligent valve, wherein the first temperature distribution includes a first sampling point and a first temperature sequence corresponding to the first sampling point, each of the first sampling points corresponds to one temperature sensor, the second temperature distribution includes a second sampling point and a second temperature sequence corresponding to the second sampling point, each of the second sampling points corresponds to one intelligent valve, and the valve opening distribution includes a third sampling point and a valve opening value corresponding to the third sampling point; A determination module is used to perform a planar unfolding of the outer surface of the pump body to obtain an unfolded plane of the outer surface of the pump body; map the first sampling point of the first temperature distribution to the unfolded plane according to the position of the corresponding temperature sensor on the outer surface of the pump body, and map the second sampling point of the second temperature distribution to the unfolded plane according to the projection position of the corresponding intelligent valve on the outer surface of the pump body to obtain a first mapping plane, the first mapping plane includes the first sampling point and the second sampling point; perform Delaunay triangulation on the first mapping plane according to the sampling points in the first mapping plane to obtain a second mapping plane, the second mapping plane includes a plurality of Delaunay triangles, and the circumscribed circle of each Delaunay triangle does not contain any other sampling point in the face area where the Delaunay triangle is located; based on the second mapping plane, construct a first temperature matrix, a second temperature matrix and an opening matrix, the first temperature matrix and the second temperature matrix have the same spatiotemporal resolution, and the opening matrix and the second temperature matrix have the same spatial resolution; based on the first temperature matrix, the second temperature matrix and the opening matrix, determine the opening control parameters of each of the intelligent valves; The adjustment module is used to send the opening control parameter to the corresponding intelligent valve, so that the intelligent valve adjusts its own opening according to the opening control parameter.

7. A vacuum pump cooling device, characterized in that: The vacuum pump cooling device includes a coolant pipeline arranged on the outer surface of the pump body of the vacuum pump and an intelligent valve arranged at each branch port in the coolant pipeline, the intelligent valve is communicatively connected to the host computer, the intelligent valve is used to collect the coolant temperature and control the valve opening, and the vacuum pump cooling device is used to implement the steps in the vacuum pump cooling device control method as described in any one of claims 1 to 5.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps in the vacuum pump cooling device control method according to any one of claims 1 to 5 are implemented.

Images