Tobacco pile temperature prediction method, device and equipment and storage medium

Abstract

The embodiment of the invention discloses a tobacco pile temperature prediction method and device, equipment and a storage medium, and the method comprises the steps: obtaining the historical temperature data of a plurality of target tobacco piles, and the spatial relation data between the target tobacco piles; inputting the historical temperature data and the spatial relationship data into a target tobacco pile temperature prediction model to obtain tobacco pile temperature prediction data corresponding to the historical temperature data; wherein the tobacco pile temperature prediction data is the temperature data of the next time period corresponding to the historical temperature data. According to the technical scheme provided by the embodiment of the invention, the problem that the prediction precision is insufficient due to the fact that the future temperature of the tobacco pile is predicted by counting single temperature change data when the tobacco pile temperature analysis is carried out based on a statistical model in the prior art is solved; the historical temperature data and the spatial data of the target cigarette pile can be subjected to fusion analysis based on the target cigarette pile temperature prediction model, and the accuracy of cigarette pile temperature prediction is improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of tobacco leaf environment monitoring, and in particular, to a method, device, equipment and storage medium for predicting the temperature of tobacco stacks. Background Art

[0002] In the field of predicting the temperature of outdoor tobacco stacks, traditional methods mainly rely on statistical models, such as ARIMA, to analyze and predict temperature changes. Although these methods can provide predictions to a certain extent, they often ignore the complex spatial interactions and heat conduction effects between tobacco stacks, which limits the accuracy and reliability of the predictions. Summary of the Invention

[0003] The embodiments of the present invention provide a method, device, equipment and storage medium for predicting the temperature of tobacco stacks. The technical solutions of the embodiments of the present invention solve the problem that when analyzing the temperature of tobacco stacks based on statistical models in the prior art, predicting the future temperature of tobacco stacks by statistically analyzing single temperature change data has insufficient prediction accuracy. It is possible to perform fusion analysis on the historical temperature data and spatial data of the target tobacco stack based on the target tobacco stack temperature prediction model, and improve the accuracy of predicting the temperature of the tobacco stack.

[0004] In a first aspect, the embodiments of the present invention provide a method for predicting the temperature of a tobacco stack, and the method includes:

[0005] Obtaining historical temperature data of a plurality of target tobacco stacks, and spatial relationship data between the target tobacco stacks;

[0006] Inputting the historical temperature data and the spatial relationship data into a target tobacco stack temperature prediction model to obtain tobacco stack temperature prediction data corresponding to the historical temperature data;

[0007] Wherein, the tobacco stack temperature prediction data is temperature data for the next time period corresponding to the historical temperature data.

[0008] In a second aspect, the embodiments of the present invention provide a device for predicting the temperature of a tobacco stack, and the device includes:

[0009] A data acquisition module, configured to obtain historical temperature data of a plurality of target tobacco stacks, and spatial relationship data between the target tobacco stacks;

[0010] A tobacco stack temperature prediction module, configured to input the historical temperature data and the spatial relationship data into a target tobacco stack temperature prediction model to obtain tobacco stack temperature prediction data corresponding to the historical temperature data;

[0011] Wherein, the tobacco stack temperature prediction data is temperature data for the next time period corresponding to the historical temperature data.

[0012] In a third aspect, an embodiment of the present invention provides a computer device, which includes:

[0013] One or more processors;

[0014] A memory for storing one or more programs;

[0015] When the one or more programs are executed by the one or more processors, the one or more processors implement the stack temperature prediction method described in any embodiment.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the stack temperature prediction method described in any embodiment.

[0017] The technical solution provided by the embodiment of the present invention obtains historical temperature data of a plurality of target stacks and spatial relationship data between the target stacks; inputs the historical temperature data and the spatial relationship data into a target stack temperature prediction model to obtain stack temperature prediction data corresponding to the historical temperature data; wherein, the stack temperature prediction data is temperature data for the next time period corresponding to the historical temperature data. The technical solution of the embodiment of the present invention solves the problem of insufficient prediction accuracy in the prior art when analyzing stack temperature based on a statistical model, where the future temperature of the stack is predicted by statistically analyzing single temperature change data. It can perform fusion analysis on the historical temperature data and spatial data of the target stack based on the target stack temperature prediction model to improve the accuracy of stack temperature prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flowchart of a stack temperature prediction method provided by an embodiment of the present invention;

[0019] Figure 2 is another flowchart of a stack temperature prediction method provided by an embodiment of the present invention;

[0020] Figure 3 is a flowchart of the work for predicting stack temperature based on a target stack temperature prediction model provided by an embodiment of the present invention;

[0021] Figure 4 is a schematic structural diagram of a stack temperature prediction device provided by an embodiment of the present invention;

[0022] Figure 5 is a schematic structural diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] Figure 1 It is a flowchart of a method for predicting the temperature of a stack of tobacco leaves provided by an embodiment of the present invention. The embodiments of the present invention are applicable to scenarios where future temperature data of a stack of tobacco leaves needs to be predicted. This method can be executed by a device for predicting the temperature of a stack of tobacco leaves, and this device can be implemented in a software and / or hardware manner.

[0025] As Figure 1 shown, the method for predicting the temperature of a stack of tobacco leaves includes the following steps:

[0026] S110. Obtain historical temperature data of multiple target stacks of tobacco leaves, and spatial relationship data between the target stacks of tobacco leaves.

[0027] Among them, the target stack of tobacco leaves can be a stack of tobacco leaves for which temperature data needs to be predicted. Specifically, the target stack of tobacco leaves can be selected autonomously. In the technical solution of the embodiments of the present invention, multiple target stacks of tobacco leaves can be selected, and subsequently, the future temperature data of each target stack of tobacco leaves can be predicted separately. The historical temperature data can be the temperature data of the target stack of tobacco leaves in the past historical period. Exemplarily, the temperature of each target stack of tobacco leaves can be collected respectively based on a preset collection device, and then the historical temperature data of each target stack of tobacco leaves can be obtained. The historical temperature data of all target stacks of tobacco leaves can be obtained as sample data for subsequent temperature prediction.

[0028] Furthermore, the spatial relationship data can be data representing the spatial relative relationship between multiple target stacks of tobacco leaves. Exemplarily, the spatial distances between multiple target stacks of tobacco leaves can be measured, and then the spatial relationship data can be determined based on the spatial distances between multiple target stacks of tobacco leaves.

[0029] S120. Input the historical temperature data and the spatial relationship data into a target stack temperature prediction model to obtain stack temperature prediction data corresponding to the historical temperature data.

[0030] Among them, the target stack temperature prediction model can be a model used for predicting the stack temperature. The target stack temperature prediction model can be obtained through pre-training. The specific model training method is set according to needs and is not limited here. Further, the stack temperature prediction data is the temperature data for the next time period corresponding to the historical temperature data. For example, if the historical temperature data is the temperature data of the target stack in the previous n time periods, the stack temperature prediction data can be the predicted temperature data in the (n + 1)-th time period. Specifically, the historical temperature data and the spatial relationship data can be input into the target stack temperature prediction model, so that the model performs fusion analysis on the input historical temperature data and the spatial relationship data, and outputs the stack temperature prediction data corresponding to the historical temperature data.

[0031] The technical solution provided by the embodiments of the present invention obtains the historical temperature data of multiple target stacks and the spatial relationship data between the target stacks; inputs the historical temperature data and the spatial relationship data into the target stack temperature prediction model to obtain the stack temperature prediction data corresponding to the historical temperature data; among them, the stack temperature prediction data is the temperature data for the next time period corresponding to the historical temperature data. The technical solution of the embodiments of the present invention solves the problem of insufficient prediction accuracy in the prior art when analyzing the stack temperature based on a statistical model by predicting the future temperature of the stack through statistically single temperature change data. It can perform fusion analysis on the historical temperature data and spatial data of the target stack based on the target stack temperature prediction model, improving the accuracy of stack temperature prediction.

[0032] Figure 2 It is a flowchart of another stack temperature prediction method provided by the embodiments of the present invention. The embodiments of the present invention are applicable to scenarios where future temperature data of stacks is predicted. On the basis of the above embodiments, this embodiment further illustrates how to obtain the historical temperature data of multiple target stacks and the spatial relationship data between the target stacks; and how the stack temperature prediction model performs fusion analysis on the historical temperature data and spatial relationship data. This device can be implemented in a software and / or hardware manner and integrated into a computer device with application development functions.

[0033] As Figure 2 shown, the stack temperature prediction method includes the following steps:

[0034] S210. Obtain the historical temperature data of multiple target stacks and the spatial relationship data between the target stacks.

[0035] Among them, the target smoke stack can be the smoke stack for which temperature data prediction is required. Specifically, the target smoke stack can be selected autonomously. In the technical solution of the embodiment of the present invention, multiple target smoke stacks can be selected, and subsequently, the future temperature data of each target smoke stack can be predicted separately. The historical temperature data can be the temperature data of the target smoke stack in the past historical period. Exemplarily, the temperature of each target smoke stack can be collected based on a preset collection device, and then the historical temperature data of each target smoke stack can be obtained. The historical temperature data of all target smoke stacks can be obtained as the sample data for subsequent temperature prediction.

[0036] Optionally, obtaining the historical temperature data of multiple target smoke stacks includes: for each target smoke stack, collecting the temperature data of the target smoke stack at preset time intervals to obtain a temperature time series; preprocessing the temperature time series and using the preprocessed temperature time series as the historical temperature data; where the preprocessing includes: outlier deletion, adjacent time interval interpolation, and normalization processing.

[0037] Among them, the temperature time series can be a data series of the temperature of the target smoke stack changing with time. Specifically, for each target smoke stack, the collected temperature data of the target smoke stack can be sorted in chronological order to obtain the temperature time series. Further, outliers in the temperature time series (such as extreme values generated by temperature sensor failures) can be removed, and the outliers can be corrected using the adjacent time interval interpolation method to ensure data smoothness. Since data may be missing due to equipment failures or network interruptions, the missing values can be filled using the average method of adjacent time intervals. The minimum-maximum normalization method can also be used to map the temperature values to the range [0,1]. The formula is as follows: Where T max and T min are the maximum and minimum temperature values in the data respectively, and the historical temperature data series T i，t of each smoke stack is obtained, where i represents the smoke stack number and t represents the time step.

[0038] Optionally, when obtaining the spatial relationship data between target smoke stacks, the geographical location information of each target smoke stack can be obtained separately, and the spatial distance between adjacent target smoke stacks can be determined according to the geographical location information; a weighted adjacency matrix can be constructed based on the spatial distance between adjacent target smoke stacks, the weighted adjacency matrix can be symmetrically normalized, and the normalized weighted adjacency matrix can be used as the spatial relationship data.

[0039] Among them, the geographical location information can be information representing the spatial location of the target smoke stack. Exemplarily, the spatial coordinate information of the smoke stack can be used as the geographical location information. Further, the distance between smoke stacks is calculated using a distance function. The distance function dist(i,j) is used to represent the spatial distance between adjacent target smoke stacks. Then, a weighted adjacency matrix of adjacent smoke stacks is defined using a Gaussian kernel function:

[0040]

[0041] Among them, σ is the distance scale parameter, and ∈ is the adjacency threshold, ensuring that only adjacent smoke stacks within a certain range are considered. Further, to avoid numerical instability during the graph convolution process, the adjacency matrix A can be symmetrically normalized to obtain: Among them, D is a diagonal matrix, and the elements are

[0042] S220. Input the historical temperature data into the temperature analysis sub-model to obtain temperature features.

[0043] The target smoke stack temperature prediction model includes: a temperature analysis sub-model, a spatial analysis sub-model, and a fusion analysis sub-model. Among them, the temperature analysis sub-model is used to extract features from the temperature data of the smoke stack, the spatial analysis sub-model is used to extract features from the spatial data of the smoke stack, and the fusion analysis sub-model is used to perform fusion analysis on the extracted multiple features. The temperature features can be feature data regarding the historical temperature data of the smoke stack. Specifically, the historical temperature data can be input into the temperature analysis sub-model so that the temperature analysis sub-model extracts features from the input historical temperature data to obtain temperature features.

[0044] Optionally, the temperature analysis sub-model includes at least one LSTM network and a Transformer module. Inputting the historical temperature data into the temperature analysis sub-model to obtain temperature features includes: inputting the historical temperature data into the LSTM network to obtain short-term temperature features; inputting the short-term temperature features into the Transformer module to obtain temperature features; among them, the temperature features include: long-short-term dependence features and temperature change features; the long-short-term dependence features are used to represent the influence relationship between the short-term temperature and long-term temperature of the smoke stack on the future temperature.

[0045] Preferably, the temperature analysis sub-model includes two layers of LSTM networks and a Transformer module. Input the historical temperature data into the LSTM network, and extract time series features through two layers of LSTM units to capture the temperature change features and their short-term dependence relationships of each target smoke stack. Preferably, each layer of LSTM contains 128 neurons, and Dropout is set to 0.2 to reduce overfitting. Obtain the time series features Hi of each smoke stack at different time steps ,tAs the output of the hidden layer of the LSTM, H i,t = LSTM(T i,t ), and use the time series feature Hi ,t as the input of the Transformer network. The Transformer network will capture the long-term dependencies and temperature change characteristics of the smoke stack temperature, and obtain the enhanced output feature Z i,t , Z i,t = Transformer(H i,t ).

[0046] Furthermore, the above long short-term dependence features may include short-term temperature features and long-term temperature features. The short-term temperature features can be used to represent the influence relationship of the short-term temperature of the smoke stack on the future temperature. The long-term temperature features can be used to represent the influence relationship of the long-term temperature of the smoke stack on the future temperature. The temperature change features can be used to represent the features of the change law of the smoke stack temperature over time.

[0047] S230. Input the spatial relationship data and the temperature features into the spatial analysis sub-model to obtain spatial features.

[0048] Among them, the spatial features can be feature data regarding the historical spatial data of the smoke stack. Specifically, the spatial relationship data and the temperature features can be input into the spatial analysis sub-model, so that the spatial analysis sub-model extracts features from the input data to obtain spatial features.

[0049] Optionally, the spatial analysis sub-model includes: at least one GCN network. By inputting the spatial relationship data and the temperature features into the spatial analysis sub-model, inputting the spatial relationship data and the temperature features into the spatial analysis sub-model to obtain spatial features includes: inputting the spatial relationship data and the temperature features into the GCN network to obtain spatial features.

[0050] Preferably, the spatial analysis sub-model can include two GCN networks. The two-layer graph convolutional network (GCN) can process the time features and adjacency matrix of each smoke stack to capture the spatial correlation between the smoke stacks. The propagation formula of the GCN: where G (l) is the input feature matrix of the l-th layer, is the symmetric normalized adjacency matrix, W (l) is the feature matrix of the l-th layer, and σ(·) is the ReLU activation function. After two-layer graph convolution, the spatial feature G i,t of each smoke stack is obtained, which contains the temperature relationship information with its adjacent smoke stacks.

[0051] S240. Input the temperature features and the spatial features into the fusion analysis sub-model to obtain the predicted smoke stack temperature data.

[0052] Among them, the predicted data of the temperature of the stack can be the temperature data of the next time period corresponding to the historical temperature data. For example, if the historical temperature data is the temperature data of the target stack in the previous n time periods, the predicted data of the temperature of the stack can be the predicted temperature data in the (n + 1)-th time period. Specifically, the temperature feature and the spatial feature can be input into the fusion analysis sub-model, so that the fusion analysis sub-model performs fusion analysis on the input features to obtain the predicted data of the temperature of the stack.

[0053] Optionally, the fusion analysis sub-model includes: a feature fusion layer and an MLP layer. Inputting the temperature feature and the spatial feature into the fusion analysis sub-model to obtain the predicted data of the temperature of the stack includes: inputting the temperature feature and the spatial feature into the feature fusion layer to obtain a fused feature; inputting the fused feature into the MLP layer to output the predicted data of the temperature of the stack.

[0054] Among them, the fused feature can be the feature obtained after feature fusion. Specifically, the feature fusion layer can use weighted fusion for feature fusion to obtain the joint feature F i,t , the weighted fusion formula: F i,t = α·H i,t +(1 - α)·G i,t , where α ∈ [0, 1] is the weight of the time feature, and (1 - α) is the weight of the spatial feature. Since the change of the stack's own temperature is more important, α = 0.7 and 1 - α = 0.3 can be preferably selected, indicating that the time feature of the stack is more important.

[0055] To better understand the technical solution provided by the present invention, the following introduces specific embodiments. Exemplarily, Figure 3 is a flowchart of the operation of predicting the temperature of a stack based on the target stack temperature prediction model provided by an embodiment of the present invention. As Figure 3 shown, the operation process of the target stack temperature prediction model for predicting the temperature of the stack includes the following steps:

[0056] Step 1: Input the preprocessed temperature time series data into the improved LSTM module. Extract the short-term time series features between the temperatures of the stack through two layers of LSTM units, and then pass through the Transformer module to capture the long-term dependence relationship and temperature change features in each stack temperature change sequence, and finally capture the historical temperature change features and their long-term and short-term dependence relationships of each stack to obtain the time feature.

[0057] Step 2: Input the symmetrically normalized adjacency matrix and the time feature into the GCN module. Use two layers of graph convolutional networks (GCN) to process the time feature and the adjacency matrix of each stack to capture the spatial correlation between the stacks and obtain the spatial feature.

[0058] Step 3: The feature fusion layer performs weighted fusion on the temporal feature and the spatial feature to obtain a joint feature, where the weight of the temporal feature is greater than the weight of the spatial feature to reflect the importance of the temperature change of the smoke stack itself.

[0059] Step 4: Use a multi-layer perception mechanism (MLP) as the output layer of the model, input the joint feature into the MLP layer, and predict the future temperature of the smoke stack, such as the temperature of the smoke stack in the next few hours.

[0060] The technical solution provided by the embodiment of the present invention obtains historical temperature data of multiple target smoke stacks and spatial relationship data between the target smoke stacks; inputs the historical temperature data into a temperature analysis sub-model to obtain temperature features; inputs the spatial relationship data and the temperature features into a spatial analysis sub-model to obtain spatial features; inputs the temperature features and the spatial features into a fusion analysis sub-model to obtain smoke stack temperature prediction data. The technical solution of the embodiment of the present invention solves the problem of insufficient prediction accuracy in the prior art when analyzing the temperature of a smoke stack based on a statistical model by predicting the future temperature of the smoke stack through statistical single temperature change data. It can perform fusion analysis on the historical temperature data and spatial data of the target smoke stack based on the target smoke stack temperature prediction model, and improve the accuracy of smoke stack temperature prediction.

[0061] Figure 4 It is a schematic structural diagram of a smoke stack temperature prediction device provided by an embodiment of the present invention. The embodiment of the present invention is applicable to a scenario of predicting future temperature data of a smoke stack. The device can be implemented in a software and / or hardware manner and integrated into a computer device with application development functions.

[0062] As Figure 4 shown, the smoke stack temperature prediction device includes: a data acquisition module 310 and a smoke stack temperature prediction module 320.

[0063] Among them, the data acquisition module 310 is used to acquire historical temperature data of multiple target smoke stacks and spatial relationship data between the target smoke stacks; the smoke stack temperature prediction module 320 is used to input the historical temperature data and the spatial relationship data into a target smoke stack temperature prediction model to obtain smoke stack temperature prediction data corresponding to the historical temperature data; wherein, the smoke stack temperature prediction data is temperature data for the next time period corresponding to the historical temperature data.

[0064] The technical solution provided by the embodiments of the present invention obtains the historical temperature data of multiple target stacks of tobacco, as well as the spatial relationship data between the target stacks of tobacco; inputs the historical temperature data and the spatial relationship data into the target stack temperature prediction model to obtain the stack temperature prediction data corresponding to the historical temperature data; wherein, the stack temperature prediction data is the temperature data for the next time period corresponding to the historical temperature data. The technical solution of the embodiments of the present invention solves the problem in the prior art that when analyzing the stack temperature based on a statistical model, predicting the future temperature of the stack by statistically analyzing a single temperature change data has insufficient prediction accuracy, and can perform a fusion analysis on the historical temperature data and spatial data of the target stack based on the target stack temperature prediction model, improving the accuracy of stack temperature prediction.

[0065] In an alternative embodiment, the target stack temperature prediction model includes: a temperature analysis sub-model, a spatial analysis sub-model, and a fusion analysis sub-model. The stack temperature prediction module 320 is specifically configured to: input the historical temperature data into the temperature analysis sub-model to obtain temperature features; input the spatial relationship data and the temperature features into the spatial analysis sub-model to obtain spatial features; input the temperature features and the spatial features into the fusion analysis sub-model to obtain the stack temperature prediction data.

[0066] In an alternative embodiment, the temperature analysis sub-model includes at least one LSTM network and a Transformer module. The stack temperature prediction module 320 includes: a temperature feature analysis unit, configured to: input the historical temperature data into the LSTM network to obtain short-term temperature features; input the short-term temperature features into the Transformer module to obtain the temperature features; wherein, the temperature features include: long-term and short-term dependence features and temperature change features; the long-term and short-term dependence features are used to represent the influence relationship of the short-term temperature and long-term temperature of the stack on the future temperature.

[0067] In an alternative embodiment, the spatial analysis sub-model includes: at least one GCN network. The stack temperature prediction module 320 includes: a spatial feature analysis unit, configured to: input the spatial relationship data and the temperature features into the GCN network to obtain the spatial features.

[0068] In an alternative embodiment, the fusion analysis sub-model includes: a feature fusion layer and an MLP layer. The stack temperature prediction module 320 includes: a feature fusion analysis unit, configured to: input the temperature features and the spatial features into the feature fusion layer to obtain fusion features; input the fusion features into the MLP layer to output the stack temperature prediction data.

[0069] In an alternative embodiment, the data acquisition module 310 includes: a spatial data acquisition unit, configured to: acquire the geographical location information of each target smoke stack respectively, determine the spatial distance between adjacent target smoke stacks according to the geographical location information; construct a weighted adjacency matrix based on the spatial distance between the adjacent target smoke stacks, perform symmetric normalization processing on the weighted adjacency matrix, and use the normalized weighted adjacency matrix as the spatial relationship data.

[0070] In an alternative embodiment, the data acquisition module 310 further includes: a temperature data acquisition unit, configured to: for each target smoke stack, collect the temperature data of the target smoke stack at preset time intervals to obtain a temperature time series; preprocess the temperature time series, and use the preprocessed temperature time series as the historical temperature data; wherein, the preprocessing includes: outlier deletion, adjacent time interval interpolation, and normalization processing.

[0071] The smoke stack temperature prediction device provided by the embodiments of the present invention can execute the smoke stack temperature prediction method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0072] Figure 5 It is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Figure 5 It shows a block diagram of an exemplary computer device 12 suitable for implementing the embodiments of the present invention. Figure 5 The shown computer device 12 is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capabilities and can be configured in a smoke stack temperature prediction device.

[0073] As Figure 5 shown, the computer device 12 is presented in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0074] The bus 18 can be one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any bus structure in a variety of bus structures. For example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0075] Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0076] System memory 28 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. Computer device 12 can further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 can be used for reading and writing on non-removable, non-volatile magnetic media ( Figure 5 not shown, commonly referred to as a "hard disk drive"). Although Figure 5 not shown in the figure, a disk drive for reading and writing on removable non-volatile disks (such as "floppy disks") and an optical disk drive for reading and writing on removable non-volatile optical disks (such as CD-ROM, DVD-ROM or other optical media) can be provided. In these cases, each drive can be connected to bus 18 through one or more data media interfaces. System memory 28 can include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0077] A program / utility 40 having a set (at least one) of program modules 42 can be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment. Program modules 42 generally perform the functions and / or methods in the embodiments described in the present invention.

[0078] Computer device 12 can also communicate with one or more external devices 14 (such as a keyboard, a pointing device, a display 24, etc.), and can also communicate with one or more devices that enable a user to interact with the computer device 12, and / or communicate with any device that enables the computer device 12 to communicate with one or more other computing devices (such as a network card, a modem, etc.). Such communication can be carried out through an input / output (I / O) interface 22. And, computer device 12 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and / or a public network, such as the Internet) through a network adapter 20. As Figure 5 shown, network adapter 20 communicates with other modules of computer device 12 through bus 18. It should be understood that although Figure 5Not shown in the figure, other hardware and / or software modules may be used in combination with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0079] The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28. For example, it implements the stack temperature prediction method provided by the embodiments of the present invention. The method includes:

[0080] Obtain the historical temperature data of multiple target stacks and the spatial relationship data between the target stacks; input the historical temperature data and the spatial relationship data into the target stack temperature prediction model to obtain the stack temperature prediction data corresponding to the historical temperature data; wherein, the stack temperature prediction data is the temperature data of the next time period corresponding to the historical temperature data.

[0081] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the stack temperature prediction method provided by any embodiment of the present invention, including:

[0082] Obtain the historical temperature data of multiple target stacks and the spatial relationship data between the target stacks; input the historical temperature data and the spatial relationship data into the target stack temperature prediction model to obtain the stack temperature prediction data corresponding to the historical temperature data; wherein, the stack temperature prediction data is the temperature data of the next time period corresponding to the historical temperature data.

[0083] The computer storage medium of the embodiments of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program, which can be used by or in combination with an instruction execution system, apparatus, or device.

[0084] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0085] The program code contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0086] The computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0087] Those of ordinary skill in the art should understand that the various modules or steps of the present invention described above may be implemented using a general-purpose computing device. They may be centralized on a single computing device or distributed across a network composed of multiple computing devices. Optionally, they may be implemented using program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps of them may be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0088] Note that the above is only a preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments here, and various obvious changes, re-adjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for predicting smoke pile temperature, characterized in that: include: Acquire historical temperature data of a plurality of target cigarette piles and spatial relationship data between the target cigarette piles; Inputting the historical temperature data and the spatial relationship data into a target cigarette pile temperature prediction model to obtain cigarette pile temperature prediction data corresponding to the historical temperature data; The smoke pile temperature prediction data is the temperature data of the next time period corresponding to the historical temperature data.

2. The method according to claim 1, characterized in that The target cigarette pile temperature prediction model includes: a temperature analysis sub-model, a spatial analysis sub-model and a fusion analysis sub-model. The historical temperature data and the spatial relationship data are input into the target cigarette pile temperature prediction model to obtain the cigarette pile temperature prediction data corresponding to the historical temperature data, including: Inputting the historical temperature data into the temperature analysis sub-model to obtain temperature characteristics; Inputting the spatial relationship data and the temperature characteristics into the spatial analysis sub-model to obtain spatial characteristics; The temperature feature and the spatial feature are input into the fusion analysis sub-model to obtain the smoke pile temperature prediction data.

3. The method according to claim 2, characterized in that The temperature analysis sub-model includes at least one LSTM network and a Transformer module, and the historical temperature data is input into the temperature analysis sub-model to obtain the temperature characteristics, including: Inputting the historical temperature data into the LSTM network to obtain short-term temperature features; Inputting the short-term temperature feature into the Transformer module to obtain the temperature feature; The temperature characteristics include: long-term and short-term dependence characteristics and temperature change characteristics; the long-term and short-term dependence characteristics are used to represent the influence of the short-term temperature and long-term temperature of the smoke pile on the future temperature.

4. The method according to claim 2, characterized in that: The spatial analysis sub-model includes: at least one GCN network, and the spatial relationship data and the temperature feature are input into the spatial analysis sub-model to obtain the spatial feature, including: The spatial relationship data and the temperature feature are input into the GCN network to obtain the spatial feature.

5. The method according to claim 2, characterized in that: The fusion analysis sub-model includes: a feature fusion layer and an MLP layer. The temperature feature and the spatial feature are input into the fusion analysis sub-model to obtain the smoke pile temperature prediction data, including: Inputting the temperature feature and the spatial feature into the feature fusion layer to obtain a fusion feature; The fused features are input into the MLP layer, and the smoke pile temperature prediction data is output.

6. The method according to claim 1, characterized in that Acquiring spatial relationship data between the target cigarette piles, including: respectively obtaining geographical location information of each target cigarette pile, and determining the spatial distance between adjacent target cigarette piles according to the geographical location information; A weighted adjacency matrix is ​​constructed according to the spatial distances between the adjacent target cigarette piles, the weighted adjacency matrix is ​​symmetrically normalized, and the normalized weighted adjacency matrix is ​​used as the spatial relationship data.

7. The method according to claim 1, characterized in that The step of obtaining historical temperature data of a plurality of target cigarette piles includes: For each target cigarette pile, temperature data of the target cigarette pile is collected at preset time intervals to obtain a temperature time series; The temperature time series is preprocessed, and the preprocessed temperature time series is used as the historical temperature data; wherein the preprocessing includes: outlier deletion, adjacent time interval interpolation and normalization processing.

8. A smoke pile temperature prediction device, characterized in that: The device comprises: A data acquisition module, used to acquire historical temperature data of multiple target cigarette piles and spatial relationship data between the target cigarette piles; A cigarette pile temperature prediction module, used for inputting the historical temperature data and the spatial relationship data into a target cigarette pile temperature prediction model to obtain cigarette pile temperature prediction data corresponding to the historical temperature data; The smoke pile temperature prediction data is the temperature data of the next time period corresponding to the historical temperature data.

9. A computer device, characterized in that: The computer device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method for predicting the temperature of a cigarette pile as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for predicting the temperature of a cigarette pile as claimed in any one of claims 1 to 7 is implemented.

Images