Compound fertilizer granulation production management equipment and method based on Internet of Things

Through IoT technology and deep learning, the water vapor flow during the granulation process of composite fertilizers is intelligently controlled, solving the problem of unstable particle size in traditional methods and improving production efficiency and product quality.

CN118981190AInactive Publication Date: 2024-11-19SHANDONG HUAYI HEAVY IND TECH CO LTD

Patent Information

Application Number
CN202411024180.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional water vapor flow control methods rely on manual experience, resulting in unstable particle size of compound fertilizers, affecting product uniformity and fertilizer efficiency, and inefficient.

Method used

The composite fertilizer granulation production management equipment based on the Internet of Things is adopted, and data is collected using industrial cameras and flowmeters, particle size characteristics and water vapor flow characteristics are extracted through deep learning technology, and the water vapor valve opening is adaptively adjusted to achieve intelligent control.

Benefits of technology

The water vapor flow parameter control accuracy during the granulation process of compound fertilizer is improved, the granulation process is optimized, labor costs are reduced, and production efficiency is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118981190A_ABST
    Figure CN118981190A_ABST
Patent Text Reader

Abstract

The invention relates to the field of intelligent control, in particular to compound fertilizer granulation production management equipment and method based on the Internet of Things. A plurality of compound fertilizer images collected in the compound fertilizer granulation process are subjected to particle size analysis by using an artificial intelligence technology based on deep learning, a time sequence correlation change mode of the compound fertilizer particle size is mined, and the water vapor flow data monitored in real time is combined to determine the particle size of the compound fertilizer based on the dynamic influence relationship of the water vapor flow on the compound fertilizer particle size. The opening degree of the water vapor valve is adjusted in a self-adaptive mode, and the water vapor flow is controlled. Therefore, intelligent control over the water vapor flow parameters in the compound fertilizer granulation production process can be achieved, the granulation process is optimized, the labor cost is reduced, and the production efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and specifically to a compound fertilizer granulation production management device and method based on the Internet of Things. Background Art

[0002] When fertilizers contain more than two nutrients, they are called compound fertilizers. With the rapid development of modern agriculture, compound fertilizers are an important fertilizer product for increasing crop yields and improving soil quality, and their production efficiency and quality control are increasingly valued.

[0003] At present, the ammonia-based compound fertilizer granulation process has become an important development direction of compound fertilizer production technology due to its high output, low application cost and low energy consumption. The ammonia-based compound fertilizer granulation process mainly uses a rotary drum granulator to allow the material to fully contact with ammonia and steam during the rolling process to form granules. In this process, the reasonable control of water vapor flow is one of the key factors to ensure the stability of compound fertilizer particle size and product quality.

[0004] Traditional steam flow control methods usually rely on the operator's manual experience. This method is not only inefficient, but also has large errors, which can easily lead to instability in the particle size of compound fertilizers during production, thereby affecting the uniformity and fertilizer efficiency of the product. With the development of Internet of Things technology, more reasonable production management can be achieved through real-time monitoring and intelligent analysis of key parameters in the production process, improving product quality and production efficiency. Therefore, a compound fertilizer granulation production management device and method based on the Internet of Things is expected. Summary of the invention

[0005] This application is made in consideration of the above problems. One object of this application is to provide a compound fertilizer granulation production management device and method based on the Internet of Things.

[0006] The embodiment of the present application provides a compound fertilizer granulation production management device based on the Internet of Things, which includes:

[0007] The compound fertilizer granulation production monitoring module is used to obtain the time series of compound fertilizer images collected by the industrial camera and the time series of water vapor flow collected by the flow meter;

[0008] A compound fertilizer granularity feature extraction module is used to extract granularity features of each compound fertilizer image in the time series of the compound fertilizer image to obtain a sequence of compound fertilizer granularity image representation vectors;

[0009] A compound fertilizer particle size change time series analysis module is used to perform information time series propagation aggregation on the sequence of compound fertilizer particle size image representation vectors to obtain a compound fertilizer particle size time series associated implicit feature vector;

[0010] A water vapor flow time series encoding module, used for extracting water vapor flow time series correlation features in the time series of the water vapor flow to obtain a water vapor flow value time series correlation implicit feature vector;

[0011] The water vapor valve control module is used to determine the control instruction of the water vapor valve based on the time series interactive response characteristics between the implicit feature vector associated with the compound fertilizer particle size time series and the implicit feature vector associated with the water vapor flow value time series.

[0012] For example, according to the compound fertilizer granulation production management device based on the Internet of Things according to an embodiment of the present application, the compound fertilizer particle size feature extraction module is used to:

[0013] Each compound fertilizer image in the time series of the compound fertilizer image is respectively input into a compound fertilizer particle size feature extractor based on AlexNet to obtain a sequence of compound fertilizer particle size image representation vectors.

[0014] For example, according to the compound fertilizer granulation production management device based on the Internet of Things according to an embodiment of the present application, the compound fertilizer particle size change timing analysis module is used to:

[0015] The sequence of the compound fertilizer granularity image representation vectors is input into a node feature propagation network based on a node energy decay mechanism to obtain the compound fertilizer granularity temporal association implicit feature vector.

[0016] For example, according to the compound fertilizer granulation production management device based on the Internet of Things according to an embodiment of the present application, the compound fertilizer particle size change time series analysis module includes:

[0017] A node energy calculation unit, for calculating the node energy statistical paradigm value of each compound fertilizer granularity image representation vector in the sequence of compound fertilizer granularity image representation vectors based on the maximum value, average value and variance of each compound fertilizer granularity image representation vector to obtain a sequence of node energy statistical paradigm values, wherein the node energy statistical paradigm value corresponding to the current compound fertilizer granularity image representation vector in the sequence of node energy statistical paradigm values ​​is used as the current node energy statistical paradigm value, and other node energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of current node energy statistical paradigm values ​​and historical node energy statistical paradigm values;

[0018] A node propagation space span statistics unit is used to count the node propagation space span values ​​between each other compound fertilizer granularity image representation vector in the sequence of compound fertilizer granularity image representation vectors and the current compound fertilizer granularity image representation vector to obtain a sequence of node propagation space span values;

[0019] An energy propagation attenuation coefficient calculation unit is used to determine the node energy propagation attenuation coefficient values ​​of other compound fertilizer granularity image representation vectors in the sequence of compound fertilizer granularity image representation vectors based on the sequence of node propagation space span values ​​and the sequence of historical node energy statistical paradigm values ​​to obtain a sequence of node energy propagation attenuation coefficient values, wherein the node energy propagation attenuation coefficient value is inversely correlated with the historical node energy statistical paradigm value;

[0020] A historical node information aggregation unit, used to calculate the weighted sum of all other compound fertilizer granularity image representation vectors in the sequence of compound fertilizer granularity image representation vectors using the sequence of node energy propagation attenuation coefficient values ​​as a weight sequence to obtain a historical node energy attenuation time series aggregation feature vector;

[0021] The global feature propagation aggregation unit is used to fuse the current node energy statistical paradigm value to calculate the weighted sum of the historical node energy decay time series aggregation feature vector and the current compound fertilizer granularity image representation vector to obtain the compound fertilizer granularity time series associated implicit feature vector.

[0022] For example, according to the compound fertilizer granulation production management device based on the Internet of Things according to an embodiment of the present application, the node energy calculation unit is used to:

[0023] Calculating a weighted sum between the variance of the compound fertilizer granularity image representation vector and a bias term to obtain a first intermediate term, and calculating a weighted sum between the square of the difference between the maximum value and the mean value of the compound fertilizer granularity image representation vector, the variance of the compound fertilizer granularity image representation vector and the bias term to obtain a second intermediate term;

[0024] The first intermediate term is divided by the second intermediate term to obtain the node energy statistical normal form value.

[0025] For example, according to the compound fertilizer granulation production management device based on the Internet of Things according to an embodiment of the present application, the energy propagation attenuation coefficient calculation unit is used to:

[0026] Using a preset attenuation constant as a weight, calculating the weighted sum of each node propagation space span value in the sequence of node propagation space span values ​​and its own exponential function value with a natural constant as the base to obtain a sequence of space span attenuation factors;

[0027] The corresponding division results of each historical node energy statistical normal value in the sequence of the historical node energy statistical normal value and each spatial span attenuation factor in the sequence of the spatial span attenuation factor are calculated to obtain the sequence of the node energy propagation attenuation coefficient values.

[0028] For example, according to the compound fertilizer granulation production management device based on the Internet of Things according to an embodiment of the present application, the water vapor flow timing coding module is used to:

[0029] The time series of the water vapor flow is input into a sequence encoder based on a bidirectional gated recurrent unit to obtain a temporal association implicit feature vector of the water vapor flow value.

[0030] For example, according to the compound fertilizer granulation production management equipment based on the Internet of Things according to the embodiment of the present application, the water vapor valve control module includes:

[0031] A time series interaction response analysis unit, used for inputting the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector into a feature interaction adaptive response module to obtain a compound fertilizer particle size-water vapor flow time series interaction response feature vector;

[0032] The control instruction generating unit is used to input the compound fertilizer particle size-water vapor flow time series interactive response feature vector into the classifier-based valve controller to obtain the control instruction, and the control instruction is used to indicate whether the valve opening should increase, decrease or remain unchanged.

[0033] For example, according to the compound fertilizer granulation production management device based on the Internet of Things according to an embodiment of the present application, the timing interaction response analysis unit is used to:

[0034] Calculating the position-by-position response between the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector to obtain the compound fertilizer particle size-water vapor flow rate position-by-position response feature vector;

[0035] Using a Softmax function to normalize the compound fertilizer particle size-water vapor flow rate position-by-position response feature vector to obtain a normalized compound fertilizer particle size-water vapor flow rate position-by-position response feature vector;

[0036] Inputting the normalized compound fertilizer particle size-water vapor flow rate position-by-position response feature vector into a learnable gating function to obtain a compound fertilizer particle size-water vapor flow rate response screening weight mask vector;

[0037] Calculate the position point multiplication between the compound fertilizer particle size-water vapor flow rate response screening weight mask vector and the normalized compound fertilizer particle size-water vapor flow rate position-by-position response feature vector to obtain a compound fertilizer particle size-water vapor flow rate position-by-position response distinguishable weight mask vector;

[0038] The compound fertilizer particle size-water vapor flow rate position-by-position response distinguishable weight mask vector is calculated and multiplied by the position point between the compound fertilizer particle size-water vapor flow rate position-by-position response feature vector to obtain the compound fertilizer particle size-water vapor flow rate time series interactive response feature vector.

[0039] The embodiment of the present application also provides a compound fertilizer granulation production management method based on the Internet of Things, which includes:

[0040] Obtain the time series of compound fertilizer images collected by an industrial camera and the time series of water vapor flow collected by a flow meter;

[0041] Extracting granularity features of each compound fertilizer image in the time series of the compound fertilizer images to obtain a sequence of compound fertilizer granularity image representation vectors;

[0042] Performing information time series propagation aggregation on the sequence of compound fertilizer granularity image representation vectors to obtain a compound fertilizer granularity time series associated implicit feature vector;

[0043] Extracting the water vapor flow time series correlation feature in the time series of the water vapor flow to obtain the water vapor flow value time series correlation implicit feature vector;

[0044] Based on the time series interactive response characteristics between the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector, a control instruction of the water vapor valve is determined.

[0045] According to the compound fertilizer granulation production management equipment and method based on the Internet of Things of the embodiment of the present application, it uses artificial intelligence technology based on deep learning to perform particle size analysis on multiple compound fertilizer images collected during the compound fertilizer granulation process, dig out the temporal correlation change pattern of the compound fertilizer particle size, and combine the real-time monitored water vapor flow data, based on the dynamic influence of water vapor flow on compound fertilizer particle size, adaptively adjust the opening of the water vapor valve, and control the size of the water vapor flow. In this way, it is possible to achieve intelligent control of water vapor flow parameters in the compound fertilizer granulation production process, optimize the granulation process, reduce labor costs, and improve production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0047] Figure 1 A schematic diagram of the application architecture of a compound fertilizer granulation production management device based on the Internet of Things in an embodiment of the present application is shown;

[0048] Figure 2The schematic diagram of the structure of the compound fertilizer granulation production management equipment based on the Internet of Things in the embodiment of the present application is shown;

[0049] Figure 3 A schematic diagram showing the structure of a compound fertilizer particle size change timing analysis module of a compound fertilizer granulation production management device based on the Internet of Things in an embodiment of the present application is shown; and

[0050] Figure 4 A flow chart of a compound fertilizer granulation production management method based on the Internet of Things in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0052] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0053] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0054] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0055] Figure 1 A schematic diagram of the application architecture of a compound fertilizer granulation production management device based on the Internet of Things in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0056] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0057] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0058] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the corresponding method of the compound fertilizer granulation production management device based on the Internet of Things in the embodiment of the present application.

[0059] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the compound fertilizer granulation production management methods based on the Internet of Things in the embodiments of the present application according to the obtained program instructions.

[0060] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0061] The following is a non-restrictive description of the Internet of Things-based compound fertilizer granulation production management equipment provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also belong to the scope of protection of this application.

[0062] In view of the above technical problems, the technical concept of this application is to use artificial intelligence technology based on deep learning to perform particle size analysis on multiple compound fertilizer images collected during the compound fertilizer granulation process, dig out the temporal correlation change pattern of the compound fertilizer particle size, and combine the real-time monitored water vapor flow data, based on the dynamic influence of water vapor flow on compound fertilizer particle size, adaptively adjust the opening of the water vapor valve, and control the size of the water vapor flow. In this way, it is possible to achieve intelligent control of water vapor flow parameters in the compound fertilizer granulation production process, optimize the granulation process, reduce labor costs, and improve production efficiency.

[0063] Based on this, Figure 2A schematic structural diagram of a compound fertilizer granulation production management device 800 based on the Internet of Things in an embodiment of the present application is shown. The compound fertilizer granulation production management device 800 based on the Internet of Things includes: a compound fertilizer granulation production monitoring module 810, which is used to obtain the time series of compound fertilizer images collected by the industrial camera and the time series of water vapor flow collected by the flow meter; a compound fertilizer particle size feature extraction module 820, which is used to extract the particle size features of each compound fertilizer image in the time series of the compound fertilizer image to obtain a sequence of compound fertilizer particle size image representation vectors; a compound fertilizer particle size change timing analysis module 830, which is used to perform information timing propagation aggregation on the sequence of compound fertilizer particle size image representation vectors to obtain a compound fertilizer particle size timing-related implicit feature vector; a water vapor flow timing encoding module 840, which is used to extract the water vapor flow timing-related features in the water vapor flow time series to obtain a water vapor flow value timing-related implicit feature vector; a water vapor valve control module 850, which is used to determine the control instruction of the water vapor valve based on the timing interaction response characteristics between the compound fertilizer particle size timing-related implicit feature vector and the water vapor flow value timing-related implicit feature vector.

[0064] Specifically, in the technical solution of the present application, first, the time series of compound fertilizer images collected by the industrial camera and the time series of water vapor flow collected by the flowmeter are obtained. It should be understood that in the production process of compound fertilizer, water vapor not only serves as a heat source, but also affects the formation and quality characteristics of particles. Specifically, the size of the water vapor flow rate will affect the humidity of the material during the granulation process. Suitable humidity contributes to the formation of particles, but too high humidity may cause particles to agglomerate, and too low humidity may make the particles too loose, affecting their dissolution rate and crop absorption efficiency. Therefore, in the technical solution of the present application, it is expected to continuously collect compound fertilizer images and water vapor flow, and use deep learning technology to analyze the dynamic influence of water vapor flow on the particle size state of compound fertilizer, so as to achieve adaptive control of water vapor flow. It is worth mentioning that the above-mentioned detection data can be further transmitted to the WEB cloud platform.

[0065] Specifically, firstly, the compound fertilizer particle size feature extractor based on AlexNet is used to perform feature extraction on each compound fertilizer image in the time series of the compound fertilizer image, so as to capture the complex features in the compound fertilizer image through a deep architecture, and to mine key information such as the particle size, shape, and distribution of the compound fertilizer, and convert the original image into a more compact and representative vector form to obtain a sequence of compound fertilizer particle size image representation vectors, which effectively reflects the particle size characteristics of the compound fertilizer.

[0066] Accordingly, the compound fertilizer granularity feature extraction module 820 is used to: input each compound fertilizer image in the time series of the compound fertilizer image into the compound fertilizer granularity feature extractor based on AlexNet to obtain a sequence of compound fertilizer granularity image representation vectors.

[0067] Next, in order to capture the dynamic evolution law of the particle size characteristics of compound fertilizer in the time dimension, it is necessary to further integrate the features of the sequence of the compound fertilizer particle size image representation vector. At the same time, considering that in the time series, the correlation of the particle size changes between each image will gradually weaken with the increase of the time span, for this reason, in the technical solution of the present application, a node feature propagation network based on the node energy decay mechanism is further introduced to perform feature aggregation on the sequence of the compound fertilizer particle size image representation vector to obtain the compound fertilizer particle size time series associated implicit feature vector. Specifically, the node feature propagation network uses the energy intensity of each compound fertilizer particle size image representation vector as the basis for feature propagation aggregation, and introduces the node energy decay mechanism in the feature propagation aggregation process. By calculating the spatial span between each compound fertilizer particle size image representation vector and the current (i.e., the end of the time series) compound fertilizer particle size image representation vector, the corresponding energy decay coefficient is generated, so that when the feature is aggregated, the feature influence of the early time step is gradually weakened, and the feature of the recent time step is enhanced, thereby more accurately capturing and reflecting the dynamic change trend of the compound fertilizer particle size feature over time.

[0068] Accordingly, the compound fertilizer granularity variation time series analysis module 830 is used to: input the sequence of compound fertilizer granularity image representation vectors into a node feature propagation network based on a node energy decay mechanism to obtain the compound fertilizer granularity time series associated implicit feature vector.

[0069] Specifically, Figure 3As shown, the compound fertilizer granularity change timing analysis module 830 includes: a node energy calculation unit 831, which is used to calculate the node energy statistical paradigm value of each compound fertilizer granularity image representation vector based on the maximum value, average value and variance of each compound fertilizer granularity image representation vector in the sequence of compound fertilizer granularity image representation vectors to obtain a sequence of node energy statistical paradigm values, wherein the node energy statistical paradigm value corresponding to the current compound fertilizer granularity image representation vector in the sequence of node energy statistical paradigm values ​​is used as the current node energy statistical paradigm value, and other node energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of current node energy statistical paradigm values ​​and historical node energy statistical paradigm values; a node propagation space span statistics unit 832, which is used to count the node propagation space span values ​​between each other compound fertilizer granularity image representation vector in the sequence of compound fertilizer granularity image representation vector and the current compound fertilizer granularity image representation vector to obtain a sequence of node propagation space span values; The propagation attenuation coefficient calculation unit 833 is used to determine the node energy propagation attenuation coefficient values ​​of other compound fertilizer granularity image representation vectors in the sequence of compound fertilizer granularity image representation vectors based on the sequence of node propagation space span values ​​and the sequence of historical node energy statistical paradigm values ​​to obtain a sequence of node energy propagation attenuation coefficient values, wherein the node energy propagation attenuation coefficient values ​​are inversely correlated with the historical node energy statistical paradigm values; the historical node information aggregation unit 834 is used to use the sequence of node energy propagation attenuation coefficient values ​​as a weight sequence to calculate the weighted sum of all other compound fertilizer granularity image representation vectors in the sequence of compound fertilizer granularity image representation vectors to obtain a historical node energy attenuation time series aggregation feature vector; the global feature propagation aggregation unit 835 is used to fuse the current node energy statistical paradigm value to calculate the weighted sum of the historical node energy attenuation time series aggregation feature vector and the current compound fertilizer granularity image representation vector to obtain the compound fertilizer granularity time series associated implicit feature vector.

[0070] Among them, the node energy calculation unit 831 is used to: calculate the weighted sum between the variance and the bias term of the compound fertilizer granularity image representation vector to obtain a first intermediate term, and calculate the square of the difference between the maximum value and the mean of the compound fertilizer granularity image representation vector, the variance of the compound fertilizer granularity image representation vector and the weighted sum between the bias term to obtain a second intermediate term; calculate the first intermediate term divided by the second intermediate term to obtain the node energy statistical paradigm value.

[0071] Among them, the energy propagation attenuation coefficient calculation unit 833 is used to: use a preset attenuation constant as a weight to calculate the weighted sum of each node propagation space span value in the sequence of node propagation space span values ​​and its own exponential function value with the natural constant as the base to obtain a sequence of spatial span attenuation factors; calculate the corresponding division results of each historical node energy statistical norm value in the sequence of historical node energy statistical norm values ​​and each spatial span attenuation factor in the sequence of spatial span attenuation factors to obtain the sequence of node energy propagation attenuation coefficient values.

[0072] In a specific example, the compound fertilizer granularity change time series analysis module 830 is used to: process the sequence of the compound fertilizer granularity image representation vectors using the following time series propagation fusion formula to obtain the compound fertilizer granularity time series associated implicit feature vector, wherein the time series propagation fusion formula is:

[0073] X={x1,x2,...x k-1 ,x k}

[0074]

[0075] Wherein, X represents the sequence of the compound fertilizer particle size image representation vector, the value of k is the number of feature vectors in the sequence of the compound fertilizer particle size image representation vector, and x1, x2, x i 、x k-1 and x k respectively represent the first, second, ith, k-1th and current compound fertilizer granularity image representation vectors in the sequence of the compound fertilizer granularity image representation vector, represents the mth eigenvalue of the i-th compound fertilizer particle size image representation vector, the value of L is the length of the i-th compound fertilizer particle size image representation vector, μ i is the mean of the eigenvalues ​​of the i-th compound fertilizer particle size image representation vector, σ 2 i is the variance of each eigenvalue of the i-th compound fertilizer particle size image representation vector, max(x i ) represents the maximum value of each eigenvalue of the i-th compound fertilizer particle size image representation vector, ∈ is the bias term, is the node energy statistical paradigm value of the i-th compound fertilizer particle size image representation vector, Count(x k →x i ) represents the node propagation space span value between the i-th compound fertilizer granularity image representation vector and the current compound fertilizer granularity image representation vector, γ and δ are different preset attenuation constants, a and b are different weight hyperparameters, is the statistical paradigm value of the current node energy, and v1 represents the implicit feature vector associated with the compound fertilizer granularity time series.

[0076] At the same time, in order to capture the temporal variation law of water vapor flow, a sequence encoder based on a bidirectional gated recurrent unit (Bi-GRU) is further used to perform temporal encoding on the time series of the water vapor flow. By utilizing the bidirectional information transmission capability and gating mechanism of Bi-GRU, the long-term dependency of water vapor flow data can be mined, and the changing trend of water vapor flow in the time dimension and the potential periodic pattern can be effectively extracted, so as to obtain the implicit feature vector of the time series association of water vapor flow values.

[0077] Correspondingly, the water vapor flow time series encoding module 840 is used to: input the time series of the water vapor flow into a sequence encoder based on a bidirectional gated recurrent unit to obtain the water vapor flow value time series associated implicit feature vector.

[0078] Furthermore, in order to capture the potential correlation between water vapor flow and compound fertilizer particle size, the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector are further input into the feature interaction adaptive response module for feature interaction response analysis to obtain the compound fertilizer particle size-water vapor flow time series interaction response feature vector. Specifically, the feature interaction adaptive response module first generates preliminary response features by calculating the position-by-position responses of the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector, and then normalizes it using the Softmax function to balance the scale of the feature response and make the feature expression more balanced. Next, by introducing a gated screening mechanism, the response weight mask of each position is adaptively generated based on the normalized response features, and the weight distribution is further refined through the mask operation to improve the distinguishability of feature selection. Finally, through the dot multiplication operation, the masked response weights are applied to the original position-by-position response features to achieve position-by-position weighted processing to enhance the pertinence and effectiveness of feature interaction, thereby more accurately revealing the dynamic response characteristics of the influence of water vapor flow rate changes on the particle size of compound fertilizer.

[0079] Next, the compound fertilizer particle size-water vapor flow rate time series interactive response feature vector is input into a classifier-based valve controller to obtain a control instruction, and the control instruction is used to indicate whether the valve opening should be increased, decreased, or kept unchanged. It should be understood that the classifier-based valve controller performs feature pattern learning on the compound fertilizer particle size-water vapor flow rate time series interactive response feature vector, and combines the decision rules learned during the training process to predict the direction in which the valve opening should be adjusted under the current water vapor flow conditions. Specifically, the classifier adopts a multi-classification structure, which can output multiple possible control instruction category probabilities including increase, decrease, and remain unchanged, thereby determining the final valve opening adjustment decision based on the control instruction category with the highest probability, so as to achieve continuous adjustment decisions on the valve opening.

[0080] Correspondingly, the water vapor valve control module 850 includes: a timing interaction response analysis unit, which is used to input the compound fertilizer particle size timing-related implicit feature vector and the water vapor flow value timing-related implicit feature vector into a feature interaction adaptive response module to obtain a compound fertilizer particle size-water vapor flow timing interaction response feature vector; a control instruction generation unit, which is used to input the compound fertilizer particle size-water vapor flow timing interaction response feature vector into a classifier-based valve controller to obtain the control instruction, and the control instruction is used to indicate that the valve opening should increase, decrease or remain unchanged.

[0081] The timing interaction response analysis unit is used to: calculate the position-by-position response between the compound fertilizer particle size timing-related implicit feature vector and the water vapor flow value timing-related implicit feature vector to obtain the compound fertilizer particle size-water vapor flow position-by-position response feature vector; use the Softmax function to normalize the compound fertilizer particle size-water vapor flow position-by-position response feature vector to obtain the normalized compound fertilizer particle size-water vapor flow position-by-position response feature vector; input the normalized compound fertilizer particle size-water vapor flow position-by-position response feature vector into the learnable gating function to obtain A compound fertilizer particle size-water vapor flow rate response screening weight mask vector; calculating the position point multiplication between the compound fertilizer particle size-water vapor flow rate response screening weight mask vector and the normalized compound fertilizer particle size-water vapor flow rate position response feature vector to obtain a compound fertilizer particle size-water vapor flow rate position response distinguishable weight mask vector; calculating the position point multiplication between the compound fertilizer particle size-water vapor flow rate position response distinguishable weight mask vector and the compound fertilizer particle size-water vapor flow rate position response feature vector to obtain the compound fertilizer particle size-water vapor flow rate time series interactive response feature vector.

[0082] Correspondingly, the timing interaction response analysis unit is used to: process the compound fertilizer particle size timing-related implicit feature vector and the water vapor flow value timing-related implicit feature vector using the following feature interaction response fusion formula to obtain the compound fertilizer particle size-water vapor flow timing interaction response feature vector, wherein the feature interaction response fusion formula is:

[0083] v r =v2 / v1

[0084] v n =softmax(v r )

[0085]

[0086] Wherein, v1 represents the implicit characteristic vector of the compound fertilizer particle size time series association, v2 represents the implicit characteristic vector of the water vapor flow value time series association, and v r represents the position-by-position response feature vector of the compound fertilizer particle size-water vapor flow rate, softmax is a normalized exponential function, v n represents the normalized compound fertilizer particle size-water vapor flow rate position-by-position response feature vector, v a represents the compound fertilizer particle size-water vapor flow rate response screening weight mask vector, exp(·) represents the exponential function operation with e as the base, ⊙ represents the multiplication by position point, It represents the compound fertilizer particle size-water vapor flow time series interactive response characteristic vector.

[0087] Preferably, the water vapor valve control module 850 is used to: calculate the first norm and second norm of the mean vector of the implicit characteristic vector associated with the compound fertilizer particle size time series and the implicit characteristic vector associated with the water vapor flow value time series; calculate the weighted sum of the inverse of the square root of the second norm and the first norm, and perform point multiplication with the point addition sum vector of the implicit characteristic vector associated with the compound fertilizer particle size time series and the implicit characteristic vector associated with the water vapor flow value time series to obtain a first compound fertilizer particle size-water vapor flow time series interactive response correction subvector; perform point multiplication of the dot product vector of the implicit characteristic vector associated with the compound fertilizer particle size time series and the implicit characteristic vector associated with the water vapor flow value time series with the square root of the length of the compound fertilizer particle size-water vapor flow time series interactive response characteristic vector. The method comprises the following steps: performing point multiplication to obtain a second compound fertilizer particle size-water vapor flow timing interaction response correction subvector; calculating a weighted sum of the first compound fertilizer particle size-water vapor flow timing interaction response correction subvector and the second compound fertilizer particle size-water vapor flow timing interaction response correction subvector to obtain a compound fertilizer particle size-water vapor flow timing interaction response correction vector; performing point multiplication on the compound fertilizer particle size-water vapor flow timing interaction response correction vector and the compound fertilizer particle size-water vapor flow timing interaction response characteristic vector to obtain a corrected compound fertilizer particle size-water vapor flow timing interaction response characteristic vector; and inputting the corrected compound fertilizer particle size-water vapor flow timing interaction response characteristic vector into the compound fertilizer particle size-water vapor flow timing interaction response characteristic vector to obtain a control instruction.

[0088] The calculation process of the compound fertilizer particle size-water vapor flow rate time series interactive response correction vector is expressed by the following correction formula:

[0089]

[0090] Among them, V1 is the implicit characteristic vector of the compound fertilizer particle size time series association, V2 is the implicit characteristic vector of the water vapor flow value time series association, V μ is the mean vector of the implicit feature vector associated with the compound fertilizer particle size time series and the implicit feature vector associated with the water vapor flow value time series, ‖·‖1 represents the first norm of the feature vector, ‖·‖2 represents the second norm of the feature vector, L is the length of the compound fertilizer particle size-water vapor flow time series interactive response feature vector, and α and β are weighted sums of weights as hyperparameters, represents addition at the position point, ⊙ represents multiplication at the position point, and V′ represents the compound fertilizer particle size-water vapor flow rate time series interactive response correction vector.

[0091] Here, considering that the compound fertilizer granularity time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector respectively represent the compound fertilizer image granularity time series associated implicit feature and the water vapor flow value time series associated implicit feature, when they perform feature interactive adaptive response, it is expected to further compensate for the extreme category deviation caused by the unbalanced correspondence ratio caused by the difference in feature distribution mode under the fine-grained distribution of the time series, so as to improve the accuracy of the classification results.

[0092] Therefore, based on the mean vector norm between the implicit feature vector associated with the compound fertilizer granularity time series and the implicit feature vector associated with the water vapor flow value time series, the constrained representation of the structured foreground and background distinction of the supermanifold of the implicit feature vector associated with the compound fertilizer granularity time series and the implicit feature vector associated with the water vapor flow value time series is used to model the feature-level key correspondence between the feature vectors of the implicit feature vector associated with the compound fertilizer granularity time series and the implicit feature vector associated with the water vapor flow value time series, and adjust the global correlation relationship between the corresponding features, so as to make positive fine-grained correspondence suggestions through the unbalanced proportion control between the corresponding feature values ​​of the implicit feature vector associated with the compound fertilizer granularity time series and the implicit feature vector associated with the water vapor flow value time series, so as to avoid the extreme imbalance of categories between the feature vectors of the implicit feature vector associated with the compound fertilizer granularity time series and the implicit feature vector associated with the water vapor flow value time series by focusing on the focus.

[0093] In this way, by correcting the compound fertilizer particle size-water vapor flow timing interaction response characteristic vector through the compound fertilizer particle size-water vapor flow timing interaction response correction vector, the accuracy of the control instructions obtained by inputting the compound fertilizer particle size-water vapor flow timing interaction response characteristic vector into the compound fertilizer particle size-water vapor flow timing interaction response characteristic vector can be improved.

[0094] Based on the above embodiments, see Figure 4 As shown in FIG. 1 , it is a flow chart of a compound fertilizer granulation production management method based on the Internet of Things in an embodiment of the present application. For example, the compound fertilizer granulation production management method based on the Internet of Things can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 4As shown, according to the embodiment of the present application, the compound fertilizer granulation production management method based on the Internet of Things includes the following steps: S510, acquiring the time series of compound fertilizer images collected by the industrial camera and the time series of water vapor flow collected by the flow meter; S520, performing granularity feature extraction on each compound fertilizer image in the time series of the compound fertilizer image to obtain a sequence of compound fertilizer granularity image representation vectors; S530, performing information time series propagation aggregation on the sequence of compound fertilizer granularity image representation vectors to obtain a compound fertilizer granularity time series associated implicit feature vector; S540, extracting the water vapor flow time series associated features in the water vapor flow time series to obtain a water vapor flow value time series associated implicit feature vector; S550, determining the control instruction of the water vapor valve based on the time series interactive response characteristics between the compound fertilizer granularity time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector.

[0095] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned compound fertilizer granulation production management method based on the Internet of Things have been referred to above. Figure 2 to Figure 3 The description of the compound fertilizer granulation production management equipment 800 based on the Internet of Things has been introduced in detail, and therefore, its repeated description will be omitted.

[0096] Based on the above embodiment, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the compound fertilizer granulation production management method based on the Internet of Things in the above embodiment when executing the program.

[0097] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0098] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the compound fertilizer granulation production management method based on the Internet of Things according to the embodiment of the present application described with reference to the above drawings can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0099] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer executable instructions from the computer-readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the compound fertilizer granulation production management method based on the Internet of Things according to the embodiment of the present application.

[0100] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0101] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0102] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0103] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0104] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A compound fertilizer granulation production management equipment based on the Internet of Things, It is characterized in that include: The compound fertilizer granulation production monitoring module is used to obtain the time series of compound fertilizer images collected by the industrial camera and the time series of water vapor flow collected by the flow meter; A compound fertilizer granularity feature extraction module is used to extract granularity features of each compound fertilizer image in the time series of the compound fertilizer image to obtain a sequence of compound fertilizer granularity image representation vectors; A compound fertilizer particle size change time series analysis module is used to perform information time series propagation aggregation on the sequence of compound fertilizer particle size image representation vectors to obtain a compound fertilizer particle size time series associated implicit feature vector; A water vapor flow time series encoding module, used for extracting water vapor flow time series correlation features in the time series of the water vapor flow to obtain a water vapor flow value time series correlation implicit feature vector; The water vapor valve control module is used to determine the control instruction of the water vapor valve based on the time series interactive response characteristics between the implicit feature vector associated with the compound fertilizer particle size time series and the implicit feature vector associated with the water vapor flow value time series.

2. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 1, It is characterized in that The compound fertilizer particle size feature extraction module is used for: Each compound fertilizer image in the time series of the compound fertilizer image is respectively input into a compound fertilizer particle size feature extractor based on AlexNet to obtain a sequence of compound fertilizer particle size image representation vectors.

3. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 2, It is characterized in that The compound fertilizer particle size change time series analysis module is used for: The sequence of the compound fertilizer granularity image representation vectors is input into a node feature propagation network based on a node energy decay mechanism to obtain the compound fertilizer granularity temporal association implicit feature vector.

4. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 3, It is characterized in that The compound fertilizer particle size change time series analysis module comprises: A node energy calculation unit, for calculating the node energy statistical paradigm value of each compound fertilizer granularity image representation vector in the sequence of compound fertilizer granularity image representation vectors based on the maximum value, average value and variance of each compound fertilizer granularity image representation vector to obtain a sequence of node energy statistical paradigm values, wherein the node energy statistical paradigm value corresponding to the current compound fertilizer granularity image representation vector in the sequence of node energy statistical paradigm values ​​is used as the current node energy statistical paradigm value, and other node energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of current node energy statistical paradigm values ​​and historical node energy statistical paradigm values; A node propagation space span statistics unit is used to count the node propagation space span values ​​between each other compound fertilizer granularity image representation vector in the sequence of compound fertilizer granularity image representation vectors and the current compound fertilizer granularity image representation vector to obtain a sequence of node propagation space span values; An energy propagation attenuation coefficient calculation unit is used to determine the node energy propagation attenuation coefficient values ​​of other compound fertilizer granularity image representation vectors in the sequence of compound fertilizer granularity image representation vectors based on the sequence of node propagation space span values ​​and the sequence of historical node energy statistical paradigm values ​​to obtain a sequence of node energy propagation attenuation coefficient values, wherein the node energy propagation attenuation coefficient value is inversely correlated with the historical node energy statistical paradigm value; A historical node information aggregation unit, used to calculate the weighted sum of all other compound fertilizer granularity image representation vectors in the sequence of compound fertilizer granularity image representation vectors using the sequence of node energy propagation attenuation coefficient values ​​as a weight sequence to obtain a historical node energy attenuation time series aggregation feature vector; The global feature propagation aggregation unit is used to fuse the current node energy statistical paradigm value to calculate the weighted sum of the historical node energy decay time series aggregation feature vector and the current compound fertilizer granularity image representation vector to obtain the compound fertilizer granularity time series associated implicit feature vector.

5. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 4, It is characterized in that The node energy calculation unit is used to: Calculating a weighted sum between the variance of the compound fertilizer granularity image representation vector and a bias term to obtain a first intermediate term, and calculating a weighted sum between the square of the difference between the maximum value and the mean value of the compound fertilizer granularity image representation vector, the variance of the compound fertilizer granularity image representation vector and the bias term to obtain a second intermediate term; The first intermediate term is divided by the second intermediate term to obtain the node energy statistical normal form value.

6. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 5, It is characterized in that The energy propagation attenuation coefficient calculation unit is used to: Using a preset attenuation constant as a weight, calculating the weighted sum of each node propagation space span value in the sequence of node propagation space span values ​​and its own exponential function value with a natural constant as the base to obtain a sequence of space span attenuation factors; The corresponding division results of each historical node energy statistical normal value in the sequence of the historical node energy statistical normal value and each spatial span attenuation factor in the sequence of the spatial span attenuation factor are calculated to obtain the sequence of the node energy propagation attenuation coefficient values.

7. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 6, It is characterized in that The water vapor flow timing coding module is used to: The time series of the water vapor flow is input into a sequence encoder based on a bidirectional gated recurrent unit to obtain a temporal association implicit feature vector of the water vapor flow value.

8. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 7, It is characterized in that The water vapor valve control module comprises: A time series interaction response analysis unit, used for inputting the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector into a feature interaction adaptive response module to obtain a compound fertilizer particle size-water vapor flow time series interaction response feature vector; The control instruction generating unit is used to input the compound fertilizer particle size-water vapor flow time series interactive response feature vector into the classifier-based valve controller to obtain the control instruction, and the control instruction is used to indicate whether the valve opening should increase, decrease or remain unchanged.

9. The compound fertilizer granulation production management equipment based on the Internet of Things according to claim 8, It is characterized in that The time sequence interaction response analysis unit is used to: Calculating the position-by-position response between the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector to obtain the compound fertilizer particle size-water vapor flow rate position-by-position response feature vector; Using a Softmax function to normalize the compound fertilizer particle size-water vapor flow rate position-by-position response feature vector to obtain a normalized compound fertilizer particle size-water vapor flow rate position-by-position response feature vector; Inputting the normalized compound fertilizer particle size-water vapor flow rate position-by-position response feature vector into a learnable gating function to obtain a compound fertilizer particle size-water vapor flow rate response screening weight mask vector; Calculate the position point multiplication between the compound fertilizer particle size-water vapor flow rate response screening weight mask vector and the normalized compound fertilizer particle size-water vapor flow rate position-by-position response feature vector to obtain a compound fertilizer particle size-water vapor flow rate position-by-position response distinguishable weight mask vector; The compound fertilizer particle size-water vapor flow rate position-by-position response distinguishable weight mask vector is calculated and multiplied by the position point between the compound fertilizer particle size-water vapor flow rate position-by-position response feature vector to obtain the compound fertilizer particle size-water vapor flow rate time series interactive response feature vector.

10. A compound fertilizer granulation production management method based on the Internet of Things, It is characterized in that include: Obtain the time series of compound fertilizer images collected by an industrial camera and the time series of water vapor flow collected by a flow meter; Extracting granularity features of each compound fertilizer image in the time series of the compound fertilizer images to obtain a sequence of compound fertilizer granularity image representation vectors; Performing information time series propagation aggregation on the sequence of compound fertilizer granularity image representation vectors to obtain a compound fertilizer granularity time series associated implicit feature vector; Extracting the water vapor flow time series correlation feature in the time series of the water vapor flow to obtain the water vapor flow value time series correlation implicit feature vector; Based on the time series interactive response characteristics between the compound fertilizer particle size time series associated implicit feature vector and the water vapor flow value time series associated implicit feature vector, a control instruction of the water vapor valve is determined.

Citation Information

Patent Citations

  • Event prediction system updating method and system

    CN113283589A

  • Recommendation method based on bidirectional sparse trust

    CN113486259A

  • Intelligent preparation process and system of tiamulin fumarate premix

    CN116898822A

  • Online monitoring method and system for mushroom bran fermented feed production

    CN117535452A

  • Automatic pressure swing adsorption control system and method thereof

    CN118079592A

Cited By

  • Camera image anomaly detection method and system

    CN120823563A

  • Fertilizer production self-adaptive control method and system based on strategy selection

    CN121635211A

  • Method and system for adaptive control of fertilizer production based on strategy selection

    CN121635211B