Remote breeding order automation management system and method based on the Internet of Things

Abstract

The present application relates to an automated management system and method for remote breeding orders based on the Internet of Things. The method includes: semantically encoding the acquired order information to obtain a semantically encoded feature vector of the order information; extracting multi-dimensional temporal features from the acquired time series of feed inventory to obtain a temporal correlation feature vector of the absolute quantity of feed inventory and a temporal correlation feature vector of the consumption of feed inventory; dynamically interactively fusing the temporal correlation feature vector of the absolute quantity of feed inventory and the temporal correlation feature vector of the consumption of feed inventory to obtain a fusion temporal representation vector of the absolute quantity and relative quantity of feed inventory; and determining whether to generate a replenishment prompt based on the interactive response characteristics between the semantically encoded feature vector of the order information and the fusion temporal representation vector of the absolute quantity and relative quantity of feed inventory. In this way, seamless connection between order processing and inventory management can be achieved, thereby significantly improving the operational efficiency and management accuracy of the breeding industry.

Description

Technical Field

[0001] The present application relates to the technical field of Internet of Things, and in particular, to an automated management system and method for remote breeding orders based on the Internet of Things. Background Art

[0002] With the rapid development of modern agriculture, the breeding industry has become an important force in promoting agricultural economic growth. However, the traditional breeding management model has many shortcomings in responding to changes in market demand, feed inventory management, and order processing. Especially in the large-scale and group farm model, excess feed inventory will increase storage costs, while inventory shortages may lead to production interruptions and delayed order delivery. Therefore, dynamic monitoring of feed inventory and efficient management of breeding orders are particularly important.

[0003] In the traditional breeding model, the monitoring of feed inventory often relies on manual inspection and record keeping, which is not only inefficient but also prone to errors, resulting in untimely updates of inventory data and difficulty in accurately reflecting the actual situation. Due to the untimely update of inventory information, farms often find it difficult to quickly assess inventory status when receiving breeding orders, which may lead to delays in order processing, affecting customer satisfaction and market competitiveness.

[0004] With the rapid development of Internet of Things technology, the breeding industry is gradually transforming towards intelligence and automation. Therefore, we are looking forward to an automated management system and method for remote breeding orders based on the Internet of Things. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present application provides an automated management system for remote farming orders based on the Internet of Things, the system comprising:

[0007] A user order information acquisition module, used to acquire user orders and extract order information from the user orders;

[0008] An order information semantic encoding module, used for semantically encoding the order information to obtain a semantic encoding feature vector of the order information;

[0009] A feed inventory information acquisition module is used to acquire the time series of feed inventory collected by the sensor;

[0010] A multi-dimensional time series feature extraction module for feed inventory is used to extract multi-dimensional time series features from the time series of the feed inventory to obtain a time series correlation feature vector of the absolute amount of feed inventory and a time series correlation feature vector of the consumption of feed inventory;

[0011] A multi-dimensional time series feature fusion module, used for dynamically and interactively fusing the time series associated feature vector of the feed inventory absolute quantity and the time series associated feature vector of the feed inventory consumption quantity to obtain a feed inventory absolute quantity-relative quantity fused time series representation vector;

[0012] The replenishment demand analysis module is used to determine whether to generate a replenishment prompt based on the interactive response characteristics between the semantic encoding feature vector of the order information and the feed inventory absolute quantity-relative quantity fusion time series representation vector.

[0013] Optionally, the multi-dimensional time series feature extraction module for feed inventory includes: a feed consumption calculation unit, used to calculate the difference between the feed inventory at every two adjacent times in the time series of the feed inventory to obtain the time series of feed consumption; and a time series encoding unit, used to perform time series encoding on the time series of the feed inventory and the time series of the feed consumption, respectively, to obtain the time series association feature vector of the absolute quantity of the feed inventory and the time series association feature vector of the feed inventory consumption.

[0014] Optionally, the time series encoding unit is used to: input the time series of the feed inventory and the time series of the feed consumption into a sequence encoder based on a Bi-LSTM model respectively to obtain the time series correlation feature vector of the absolute quantity of the feed inventory and the time series correlation feature vector of the feed inventory consumption.

[0015] Optionally, the multi-dimensional time series feature fusion module is used to: input the feed inventory absolute quantity time series associated feature vector and the feed inventory consumption time series associated feature vector into a feature vector dynamic interaction fusion module based on gated response to obtain the feed inventory absolute quantity-relative quantity fused time series representation vector.

[0016] Optionally, the multi-dimensional time series feature fusion module includes: a feature joint representation unit, used to perform feature cascading on the time series association feature vector of the absolute quantity of feed inventory and the time series association feature vector of the feed inventory consumption to obtain a joint representation vector of the absolute quantity and relative quantity time series features of the feed inventory; a gated response calculation unit, used to input the joint representation vector of the absolute quantity and relative quantity time series features of the feed inventory into a gated response function to obtain a response gate for information fusion; a feature fusion unit, used to calculate a difference with the response gate of the information fusion, and use the response gate of the information fusion and the difference as weights to calculate the position-weighted sum of the time series association feature vector of the absolute quantity of feed inventory and the time series association feature vector of the feed inventory consumption to obtain the absolute quantity and relative quantity fused time series representation vector of the feed inventory.

[0017] Optionally, the gated response calculation unit is used to: multiply the feed inventory absolute quantity-relative quantity time series feature joint representation vector using a predetermined weight vector to obtain an information interaction fusion correlation coefficient; add the information interaction fusion correlation coefficient and a predetermined bias parameter and then activate it through a sigmoid function to obtain a response gate for the information fusion.

[0018] Optionally, the replenishment demand analysis module includes: a feature interaction response analysis unit, used to input the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector into a feature interaction response module based on an adaptive distinguishable mechanism to obtain an order semantics-feed inventory interaction response representation vector; a replenishment management result generation unit, used to input the order semantics-feed inventory interaction response representation vector into a classifier-based intelligent management module to obtain a management result, and the management result is used to indicate whether to generate a replenishment prompt.

[0019] Optionally, the feature interaction response analysis unit includes: a position-by-position response subunit, which is used to calculate the position-by-position response between the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector to obtain an order semantics-feed inventory position-by-position response feature vector; a normalization subunit, which is used to use a Softmax function to normalize the order semantics-feed inventory position-by-position response feature vector to obtain a normalized order semantics-feed inventory position-by-position response feature vector; and a mask generation subunit, which is used to input the normalized order semantics-feed inventory position-by-position response feature vector into a learnable gating function. To obtain the order semantics-feed inventory response screening weight mask feature vector; a response weight masking subunit is used to calculate the position point multiplication between the order semantics-feed inventory response screening weight mask feature vector and the normalized order semantics-feed inventory position-by-position response feature vector to obtain the order semantics-feed inventory position-by-position response distinguishable weight mask feature vector; a response weight application subunit is used to calculate the position point multiplication between the order semantics-feed inventory position-by-position response distinguishable weight mask feature vector and the order semantics-feed inventory position-by-position response feature vector to obtain the order semantics-feed inventory interaction response representation vector.

[0020] In a second aspect, the present application provides a remote farming order automation management method based on the Internet of Things, the method comprising:

[0021] Obtaining a user order and extracting order information from the user order;

[0022] Performing semantic encoding on the order information to obtain a semantic encoding feature vector of the order information;

[0023] Get the time series of feed inventory collected by sensors;

[0024] Performing multi-dimensional time series feature extraction on the time series of the feed inventory to obtain a time series correlation feature vector of the absolute amount of the feed inventory and a time series correlation feature vector of the feed inventory consumption;

[0025] Dynamically interactively fusing the feed inventory absolute quantity time series correlation feature vector and the feed inventory consumption time series correlation feature vector to obtain a feed inventory absolute quantity-relative quantity fused time series representation vector;

[0026] Based on the interactive response characteristics between the semantic encoding feature vector of the order information and the absolute quantity-relative quantity fusion time series representation vector of the feed inventory, it is determined whether to generate a replenishment prompt.

[0027] Optionally, multi-dimensional time series feature extraction is performed on the time series of the feed inventory to obtain a time series correlation feature vector of the absolute quantity of feed inventory and a time series correlation feature vector of the feed inventory consumption, including: calculating the difference between the feed inventory at every two adjacent times in the time series of the feed inventory to obtain a time series of feed consumption; and performing time series encoding on the time series of the feed inventory and the time series of the feed consumption, respectively, to obtain a time series correlation feature vector of the absolute quantity of feed inventory and a time series correlation feature vector of the feed inventory consumption.

[0028] The above technical solution is adopted to obtain the semantic coding feature vector of the order information by semantically coding the acquired order information; extract the multi-dimensional time series features of the acquired feed inventory time series to obtain the feed inventory absolute quantity time series correlation feature vector and the feed inventory consumption time series correlation feature vector; dynamically interactively fuse the feed inventory absolute quantity time series correlation feature vector and the feed inventory consumption time series correlation feature vector to obtain the feed inventory absolute quantity-relative quantity fusion time series representation vector; based on the interactive response characteristics between the order information semantic coding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector, determine whether to generate a replenishment prompt. In this way, the seamless connection between order processing and inventory management can be achieved, thereby significantly improving the operational efficiency and management accuracy of the breeding industry.

[0029] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0031] Figure 1 It is a block diagram of an automated management system for remote farming orders based on the Internet of Things according to an exemplary embodiment.

[0032] Figure 2 The present invention is a flowchart of an automated management method for remote breeding orders based on the Internet of Things according to an exemplary embodiment.

[0033] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment.

[0034] Figure 4 This is an application scenario diagram of an automated management system for remote farming orders based on the Internet of Things, according to an exemplary embodiment. DETAILED DESCRIPTION

[0035] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0036] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0037] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0038] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0039] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0040] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0041] In order to solve the above problems, the present application provides an automated management system and method for remote breeding orders based on the Internet of Things, which performs semantic encoding on the acquired order information to obtain a semantic encoding feature vector of the order information; performs multi-dimensional temporal feature extraction on the acquired time series of feed inventory to obtain a temporal correlation feature vector of the absolute quantity of feed inventory and a temporal correlation feature vector of the consumption of feed inventory; performs dynamic interactive fusion on the temporal correlation feature vector of the absolute quantity of feed inventory and the temporal correlation feature vector of the consumption of feed inventory to obtain a temporal representation vector of the absolute quantity of feed inventory-relative quantity fusion; determines whether to generate a replenishment prompt based on the interactive response characteristics between the semantic encoding feature vector of order information and the temporal representation vector of the absolute quantity of feed inventory-relative quantity fusion. In this way, the seamless connection between order processing and inventory management can be achieved, thereby significantly improving the operational efficiency and management accuracy of the breeding industry.

[0042] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0043] Figure 1 FIG. 1 is a block diagram of an automated management system for remote farming orders based on the Internet of Things according to an exemplary embodiment. Figure 1 As shown, the system 100 includes:

[0044] A user order information acquisition module 101 is used to acquire user orders and extract order information from the user orders;

[0045] An order information semantic encoding module 102, configured to perform semantic encoding on the order information to obtain a semantic encoding feature vector of the order information;

[0046] A feed inventory information acquisition module 103 is used to acquire a time series of feed inventory collected by a sensor;

[0047] A feed inventory multi-dimensional time series feature extraction module 104 is used to perform multi-dimensional time series feature extraction on the time series of the feed inventory to obtain a feed inventory absolute quantity time series correlation feature vector and a feed inventory consumption quantity time series correlation feature vector;

[0048] A multi-dimensional time series feature fusion module 105 is used to dynamically and interactively fuse the time series associated feature vector of the feed inventory absolute quantity and the time series associated feature vector of the feed inventory consumption to obtain a feed inventory absolute quantity-relative quantity fused time series representation vector;

[0049] The replenishment demand analysis module 106 is used to determine whether to generate a replenishment prompt based on the interactive response characteristics between the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector.

[0050] In response to the above technical problems, the technical concept of this application is to use artificial intelligence technology based on deep learning to dynamically monitor and analyze the feed inventory of the breeding farm, dig out the time series change information of the feed inventory and consumption, and realize the automatic replenishment reminder of the breeding order based on the dynamic matching between order demand and feed inventory. In this way, the seamless connection between order processing and inventory management can be achieved, thereby significantly improving the operational efficiency and management accuracy of the breeding industry.

[0051] Based on this, in the technical solution of this application, first, the user order is obtained, and the order information is extracted from the user order. It should be understood that the order information covers key elements such as breeding types, demand, and delivery time, which is the basis for subsequent inventory management and replenishment decisions.

[0052] Next, in order to effectively extract the semantic features of the order information and convert it into a numerical feature vector that can be understood by a computer, the order information is further semantically encoded to obtain a semantically encoded feature vector of the order information. In an embodiment of the present application, a pre-trained deep language model such as BERT, RoBERTa, etc. can be used to semantically understand the order information, so as to utilize its powerful context understanding ability to accurately capture the key information in the order and generate a highly representative semantically encoded feature vector of the order information.

[0053] At the same time, sensors are used to continuously monitor the weight changes of feed inventory and obtain the time series of feed inventory to grasp the latest status of feed inventory and ensure the continuity and efficiency of breeding business.

[0054] Next, in order to deeply explore the dynamic change information of the feed inventory contained in the time series of the feed inventory and realize a comprehensive and multi-angle understanding of the feed inventory consumption status, in the technical solution of the present application, the time series of the feed inventory is further subjected to a multi-dimensional time series analysis. First, in order to more intuitively reveal the dynamic consumption of feed, in the technical solution of the present application, by calculating the difference between the feed inventory at every two adjacent times in the time series of the feed inventory, a time series of feed consumption is generated to more intuitively reflect the daily consumption rate of feed.

[0055] In one embodiment of the present application, the multi-dimensional time series feature extraction module for feed inventory includes: a feed consumption calculation unit, used to calculate the difference between the feed inventory at every two adjacent times in the time series of the feed inventory to obtain the time series of feed consumption; a time series encoding unit, used to perform time series encoding on the time series of the feed inventory and the time series of the feed consumption, respectively, to obtain the time series association feature vector of the absolute amount of feed inventory and the time series association feature vector of the feed inventory consumption.

[0056] Then, in order to fully capture the temporal changes in the absolute amount of feed inventory and consumption, this application uses a sequence encoder based on the Bi-LSTM model to perform time series modeling analysis on the time series of the feed inventory and the time series of the feed consumption, respectively, so as to utilize the bidirectional processing capability and long short-term memory characteristics of the Bi-LSTM model to capture the fore-and-aft dependencies and changing trends of feed inventory and consumption in the time dimension, fully understand the inventory fluctuations and consumption speed of the feed, and thus obtain the time series correlation feature vector of the absolute amount of feed inventory and the time series correlation feature vector of feed inventory consumption, providing important information basis for subsequent inventory management.

[0057] In one embodiment of the present application, the timing encoding unit is used to: input the time series of the feed inventory and the time series of the feed consumption into a sequence encoder based on a Bi-LSTM model respectively to obtain the time series correlation feature vector of the absolute quantity of the feed inventory and the time series correlation feature vector of the feed inventory consumption.

[0058] Secondly, the time-series correlation feature vector of the absolute quantity of feed inventory and the time-series correlation feature vector of the feed inventory consumption are further integrated to comprehensively represent the inventory status information and consumption dynamic information of the feed, and to more comprehensively understand the real-time status and future trends of the feed inventory. At the same time, considering that the absolute quantity of feed inventory determines the production capacity and short-term supply capacity of the farm, and the consumption of feed inventory reflects the operating efficiency and market demand response speed of the farm, the two may have different relative importance in the process of information integration. To this end, the present application introduces a feature vector dynamic interactive fusion module based on gated response to fuse the time-series correlation feature vector of the absolute quantity of feed inventory and the time-series correlation feature vector of the feed inventory consumption. Specifically, the feature vector dynamic interactive fusion module dynamically adjusts the weights of the two in the fusion process based on the gating mechanism to achieve flexible integration of feed inventory status information and consumption dynamic information, thereby obtaining a feed inventory absolute quantity-relative quantity fusion time series representation vector, providing more accurate data support for subsequent replenishment decisions.

[0059] In one embodiment of the present application, the multi-dimensional time series feature fusion module is used to: input the feed inventory absolute quantity time series associated feature vector and the feed inventory consumption time series associated feature vector into a feature vector dynamic interaction fusion module based on gated response to obtain the feed inventory absolute quantity-relative quantity fused time series representation vector.

[0060] Among them, the multi-dimensional time series feature fusion module includes: a feature joint representation unit, which is used to perform feature cascade on the time series association feature vector of the absolute quantity of feed inventory and the time series association feature vector of the feed inventory consumption to obtain a joint representation vector of the absolute quantity and relative quantity of feed inventory time series features; a gated response calculation unit, which is used to input the joint representation vector of the absolute quantity and relative quantity of feed inventory time series features into a gated response function to obtain a response gate for information fusion; a feature fusion unit, which is used to calculate a difference with the response gate of the information fusion, and use the response gate of the information fusion and the difference as weights to calculate the position-weighted sum of the time series association feature vector of the absolute quantity of feed inventory and the time series association feature vector of the feed inventory consumption to obtain the absolute quantity and relative quantity fused time series representation vector of feed inventory.

[0061] Furthermore, in one embodiment of the present application, the gated response calculation unit is used to: multiply the feed inventory absolute quantity-relative quantity time series feature joint representation vector using a predetermined weight vector to obtain an information interaction fusion correlation coefficient; add the information interaction fusion correlation coefficient and a predetermined bias parameter and then activate it through a sigmoid function to obtain a response gate for the information fusion.

[0062] Specifically, the feed inventory absolute quantity time series correlation feature vector and the feed inventory consumption quantity time series correlation feature vector are fused using the following interactive fusion formula to obtain the feed inventory absolute quantity-relative quantity fused time series representation vector, wherein the interactive fusion formula is:

[0063]

[0064] Wherein, v1 is the time series correlation feature vector of the absolute quantity of feed inventory, v2 is the time series correlation feature vector of the feed inventory consumption, σ is the sigmoid function, W is the predetermined weight vector, b is the predetermined bias parameter, is the response gate of information fusion, v a It is the time series representation vector of the absolute and relative quantity fusion of the feed inventory.

[0065] In one embodiment of the present application, the replenishment demand analysis module includes: a feature interaction response analysis unit, which is used to input the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector into a feature interaction response module based on an adaptive distinguishable mechanism to obtain an order semantics-feed inventory interaction response representation vector; a replenishment management result generation unit, which is used to input the order semantics-feed inventory interaction response representation vector into a classifier-based intelligent management module to obtain a management result, and the management result is used to indicate whether to generate a replenishment prompt.

[0066] Then, by further matching and analyzing the current order demand and the feed inventory status, the fit between the two and the potential contradiction between supply and demand are evaluated to facilitate making appropriate replenishment decisions. In the technical solution of the present application, a feature interaction response module based on an adaptive distinguishable mechanism is introduced to respond to the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector. Specifically, the feature interaction response module first constructs the correlation between the order demand and the feed inventory status by calculating the position-by-position response features between the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector. Then, the position-by-position response features of the two are normalized to generate response weights, and a gating mechanism is introduced to adaptively learn the feature importance and output the weight mask. The weight distribution is further refined through the mask operation to improve the distinguishability of feature selection. Then, the original position-by-position response features are weighted by position using the response weights after mask screening to obtain the order semantics-feed inventory interaction response representation vector. This position-by-position weight refinement can effectively improve the information screening capability in the feature association response analysis process, thereby more accurately understanding the dynamic matching relationship between order demand and feed inventory.

[0067] Furthermore, in one embodiment of the present application, the feature interaction response analysis unit includes: a position-by-position response subunit, which is used to calculate the position-by-position response between the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector to obtain an order semantics-feed inventory position-by-position response feature vector; a normalization subunit, which is used to use a Softmax function to normalize the order semantics-feed inventory position-by-position response feature vector to obtain a normalized order semantics-feed inventory position-by-position response feature vector; and a mask generation subunit, which is used to input the normalized order semantics-feed inventory position-by-position response feature vector into a learnable The gate function of Xi is used to obtain the order semantics-feed inventory response screening weight mask feature vector; the response weight masking subunit is used to calculate the position point multiplication between the order semantics-feed inventory response screening weight mask feature vector and the normalized order semantics-feed inventory position-by-position response feature vector to obtain the order semantics-feed inventory position-by-position response distinguishable weight mask feature vector; the response weight applying subunit is used to calculate the position point multiplication between the order semantics-feed inventory position-by-position response distinguishable weight mask feature vector and the order semantics-feed inventory position-by-position response feature vector to obtain the order semantics-feed inventory interactive response representation vector.

[0068] Specifically, the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector are processed by the following feature interaction response formula to obtain the order semantics-feed inventory interaction response representation vector, wherein the feature interaction response formula is:

[0069] v r =v b / v a

[0070] v n =softmax(v r )

[0071]

[0072] Among them, v a represents the absolute quantity and relative quantity fusion time series representation vector of the feed inventory, v b represents the semantic encoding feature vector of the order information, v r represents the order semantics-feed inventory position-by-position response feature vector, softmax is the normalized exponential function, v n represents the normalized order semantics-feed inventory position-by-position response feature vector, v e represents the order semantics-feed inventory response screening weight mask feature vector, exp(·) represents the exponential function operation with e as the base, ⊙ represents the point multiplication by position, The order semantics-feed inventory interaction response representation vector is represented.

[0073] Next, the order semantics-feed inventory interaction response representation vector is input into a classifier-based intelligent management module to obtain a management result, and the management result is used to indicate whether to generate a replenishment prompt. That is, a classification algorithm is used to perform intelligent analysis and decision-making on the order semantics-feed inventory interaction response representation vector to identify whether the current order demand matches the feed inventory status and whether replenishment is needed. In the technical solution of the present application, the classifier-based intelligent management module has undergone a large amount of training and can effectively learn the characteristic patterns of the order semantics-feed inventory interaction response representation vector, accurately identify potential supply and demand imbalances, and generate replenishment prompts accordingly, so as to improve the accuracy and timeliness of replenishment decisions, reduce human judgment errors, and optimize inventory turnover.

[0074] Preferably, inputting the order semantics-feed inventory interaction response representation vector into a classifier-based intelligent management module to obtain a management result comprises:

[0075] Calculate the first norm and second norm of the mean vector of the semantic encoding feature vector of the order information and the fusion time series representation vector of the absolute quantity and relative quantity of the feed inventory;

[0076] 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-to-point sum vector of the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector to obtain a first order semantics-feed inventory interaction response syndrome vector;

[0077] The dot product vector of the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector is dotted with the square root of the length of the order semantics-feed inventory interactive response representation vector to obtain a second order semantics-feed inventory interactive response syndrome vector;

[0078] Calculating a weighted sum of the first order semantics-feed inventory interaction response syndrome vector and the second order semantics-feed inventory interaction response syndrome vector to obtain an order semantics-feed inventory interaction response correction vector;

[0079] Performing a dot multiplication of the order semantics-feed inventory interaction response correction vector and the order semantics-feed inventory interaction response representation vector to obtain a corrected order semantics-feed inventory interaction response representation vector; and

[0080] The corrected order semantics-feed inventory interaction response representation vector is input into a classifier-based intelligent management module to obtain a management result.

[0081] The order semantics-feed inventory interaction response correction vector is expressed as:

[0082]

[0083] V1 and V2 are the semantic encoding feature vector of the order information and the fusion time series representation vector of the absolute and relative quantity of the feed inventory, respectively. μ is the mean vector of the semantic encoding feature vector of the order information and the fusion time series representation vector of the absolute quantity and relative quantity of the feed inventory, ||V μ ||1 and ||V μ ||2 respectively represent the first norm and second norm of the mean vector, L is the length of the order semantics-feed inventory interaction response representation vector, and α and β are weighted sums as hyperparameters, represents point addition by location, ⊙ represents point multiplication by location, and V' represents the order semantics-feed inventory interaction response correction vector.

[0084] Here, considering that the order information semantic encoding feature vector and the feed inventory absolute-relative quantity fused time series representation vector respectively represent the encoding semantic features of the order information and the bidirectional short-range-long-range time series gating response dynamic interactive fusion features of the feed inventory and feed consumption, when it performs the feature interactive response based on the adaptive distinguishable mechanism, it is expected to further compensate for the category deviation caused by the imbalance of the correspondence ratio under the fine-grained distribution of the sequence, so as to improve the accuracy of the classification results.

[0085] Therefore, based on the mean vector norm between the sequences of the semantic encoding feature vector of the order information and the absolute-relative quantity fusion time series representation vector of the feed inventory, the constrained representation of the structured foreground and background distinction of the supermanifold of the semantic encoding feature vector of the order information and the absolute-relative quantity fusion time series representation vector of the feed inventory is used to model the feature-level key correspondence between the feature vectors of the semantic encoding feature vector of the order information and the absolute-relative quantity fusion time series representation vector of the feed inventory, and adjust the global correlation relationship between the corresponding features, so as to make positive fine-grained corresponding suggestions through the unbalanced ratio control between the corresponding feature values ​​of the semantic encoding feature vector of the order information and the absolute-relative quantity fusion time series representation vector of the feed inventory, so as to avoid the category imbalance between the feature vectors of the semantic encoding feature vector of the order information and the absolute-relative quantity fusion time series representation vector of the feed inventory by focusing on the focus.

[0086] In this way, by correcting the order semantics-feed inventory interaction response representation vector through the order semantics-feed inventory interaction response correction vector, the accuracy of the management result obtained by inputting the order semantics-feed inventory interaction response representation vector into the classifier-based intelligent management module can be improved.

[0087] In summary, the above solution uses deep learning-based artificial intelligence technology to dynamically monitor and analyze the feed inventory of the breeding farm, dig out the time series change information of the feed inventory and consumption, and realize the automatic replenishment reminder of the breeding order based on the dynamic matching between order demand and feed inventory. In this way, the seamless connection between order processing and inventory management can be achieved, thereby significantly improving the operational efficiency and management accuracy of the breeding industry.

[0088] Figure 2 is a flow chart of an automated management method for remote farming orders based on the Internet of Things according to an exemplary embodiment. Figure 2 As shown, the method includes:

[0089] Step S201: Obtain a user order and extract order information from the user order;

[0090] Step S202: semantically encode the order information to obtain a semantically encoded feature vector of the order information;

[0091] Step S203, obtaining the time series of feed inventory collected by the sensor;

[0092] Step S204: extract multi-dimensional time series features from the time series of the feed inventory to obtain a time series correlation feature vector of the absolute amount of feed inventory and a time series correlation feature vector of the feed inventory consumption;

[0093] Step S205, dynamically interactively fuse the time series correlation feature vector of the feed inventory absolute quantity and the time series correlation feature vector of the feed inventory consumption to obtain a feed inventory absolute quantity-relative quantity fused time series representation vector;

[0094] Step S206: Determine whether to generate a replenishment prompt based on the interactive response characteristics between the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector.

[0095] In one embodiment of the present application, multi-dimensional time series feature extraction is performed on the time series of the feed inventory to obtain a time series correlation feature vector of the absolute quantity of feed inventory and a time series correlation feature vector of the feed inventory consumption, including: calculating the difference between the feed inventory at every two adjacent times in the time series of the feed inventory to obtain a time series of feed consumption; and performing time series encoding on the time series of the feed inventory and the time series of the feed consumption, respectively, to obtain a time series correlation feature vector of the absolute quantity of feed inventory and a time series correlation feature vector of the feed inventory consumption.

[0096] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0097] like Figure 3As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0098] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0099] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0100] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with 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 the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0101] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0102] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0103] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate 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., via the Internet using an Internet service provider).

[0104] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0105] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0106] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0107] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 foregoing.

[0108] Figure 4 FIG. 1 is an application scenario diagram of an automated management system for remote farming orders based on the Internet of Things according to an exemplary embodiment. Figure 4 As shown, in this application scenario, first, a user order is obtained, and order information is extracted from the user order (for example, Figure 4 C1 shown in FIG); obtaining a time series of feed inventory collected by a sensor (e.g., Figure 4 Then, the obtained order information and feed inventory time series are input into a server deployed with an IoT-based remote farming order automation management algorithm (e.g., Figure 4 In S) shown in , the server is capable of processing the time series of the order information and the feed inventory based on the remote farming order automation management algorithm of the Internet of Things to determine whether to generate a replenishment prompt.

[0109] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0110] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0111] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. An automated management system for remote breeding orders based on the Internet of Things, characterized in that: include: A user order information acquisition module, used to acquire user orders and extract order information from the user orders; An order information semantic encoding module, used for semantically encoding the order information to obtain a semantic encoding feature vector of the order information; A feed inventory information acquisition module is used to acquire the time series of feed inventory collected by the sensor; A multi-dimensional time series feature extraction module for feed inventory is used to extract multi-dimensional time series features from the time series of the feed inventory to obtain a time series correlation feature vector of the absolute amount of feed inventory and a time series correlation feature vector of the consumption of feed inventory; A multi-dimensional time series feature fusion module, used for dynamically and interactively fusing the time series associated feature vector of the feed inventory absolute quantity and the time series associated feature vector of the feed inventory consumption quantity to obtain a feed inventory absolute quantity-relative quantity fused time series representation vector; A replenishment demand analysis module, used to determine whether to generate a replenishment prompt based on the interactive response characteristics between the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector; Wherein, the replenishment demand analysis module includes: A feature interaction response analysis unit, configured to input the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector into a feature interaction response module based on an adaptive distinguishable mechanism to obtain an order semantics-feed inventory interaction response representation vector; A replenishment management result generating unit, used for inputting the order semantics-feed inventory interactive response representation vector into a classifier-based intelligent management module to obtain a management result, wherein the management result is used to indicate whether to generate a replenishment prompt; Wherein, the feature interaction response analysis unit includes: A position-by-position response subunit, used for calculating the position-by-position response between the order information semantic encoding feature vector and the feed inventory absolute quantity-relative quantity fusion time series representation vector to obtain an order semantics-feed inventory position-by-position response feature vector; A normalization subunit, used for normalizing the order semantics-feed inventory position-by-position response feature vector using a Softmax function to obtain a normalized order semantics-feed inventory position-by-position response feature vector; a mask generation subunit, configured to input the normalized order semantics-feed inventory position-by-position response feature vector into a learnable gating function to obtain an order semantics-feed inventory response screening weight mask feature vector; a response weight masking subunit, for calculating the position-wise point multiplication between the order semantics-feed inventory response screening weight mask feature vector and the normalized order semantics-feed inventory position-by-position response feature vector to obtain an order semantics-feed inventory position-by-position response distinguishable weight mask feature vector; A response weight applying subunit is used to calculate the location point multiplication between the order semantics-feed inventory location-by-location response distinguishable weight mask feature vector and the order semantics-feed inventory location-by-location response feature vector to obtain the order semantics-feed inventory interaction response representation vector.

2. The remote breeding order automation management system based on the Internet of Things according to claim 1 is characterized in that: The feed inventory multi-dimensional time series feature extraction module includes: A feed consumption calculation unit, used for calculating the difference between the feed inventories at every two adjacent times in the time series of the feed inventory to obtain the time series of the feed consumption; A time series encoding unit is used to perform time series encoding on the time series of the feed inventory and the time series of the feed consumption respectively to obtain the time series correlation feature vector of the absolute amount of the feed inventory and the time series correlation feature vector of the feed inventory consumption.

3. The remote breeding order automation management system based on the Internet of Things according to claim 2 is characterized in that: The timing coding unit is used for: The time series of the feed inventory and the time series of the feed consumption are respectively input into a sequence encoder based on a Bi-LSTM model to obtain a time series correlation feature vector of the absolute quantity of the feed inventory and a time series correlation feature vector of the feed inventory consumption.

4. The remote breeding order automation management system based on the Internet of Things according to claim 3 is characterized in that: The multi-dimensional time series feature fusion module is used to: The feed inventory absolute quantity time series associated feature vector and the feed inventory consumption time series associated feature vector are input into a feature vector dynamic interactive fusion module based on gated response to obtain the feed inventory absolute quantity-relative quantity fusion time series representation vector.

5. The remote breeding order automation management system based on the Internet of Things according to claim 4 is characterized in that: The multi-dimensional time series feature fusion module includes: A feature joint representation unit, used for performing feature concatenation on the feed inventory absolute quantity time series correlation feature vector and the feed inventory consumption time series correlation feature vector to obtain a feed inventory absolute quantity-relative quantity time series feature joint representation vector; A gated response calculation unit, used for inputting the combined representation vector of the absolute quantity and relative quantity time series characteristics of the feed inventory into a gated response function to obtain a response gate for information fusion; A feature fusion unit is used to calculate a difference with a response gate of the information fusion, and use the response gate of the information fusion and the difference as weights to calculate the position-weighted sum of the time series associated feature vector of the absolute quantity of feed inventory and the time series associated feature vector of the feed inventory consumption to obtain the feed inventory absolute quantity-relative quantity fusion time series representation vector.

6. The remote breeding order automation management system based on the Internet of Things according to claim 5 is characterized in that: The gate response calculation unit is used to: Multiplying the feed inventory absolute quantity-relative quantity time series feature joint representation vector by a predetermined weight vector to obtain an information interaction fusion correlation coefficient; After the information interaction fusion correlation coefficient and the predetermined bias parameter are added, activation processing is performed through a sigmoid function to obtain a response gate for the information fusion.

7. A remote breeding order automation management method based on the Internet of Things, using the remote breeding order automation management system based on the Internet of Things according to any one of claims 1 to 6, characterized in that: include: Obtaining a user order and extracting order information from the user order; Performing semantic encoding on the order information to obtain a semantic encoding feature vector of the order information; Get the time series of feed inventory collected by sensors; Performing multi-dimensional time series feature extraction on the time series of the feed inventory to obtain a time series correlation feature vector of the absolute amount of the feed inventory and a time series correlation feature vector of the feed inventory consumption; Dynamically interactively fusing the feed inventory absolute quantity time series correlation feature vector and the feed inventory consumption time series correlation feature vector to obtain a feed inventory absolute quantity-relative quantity fused time series representation vector; Based on the interactive response characteristics between the semantic encoding feature vector of the order information and the absolute quantity-relative quantity fusion time series representation vector of the feed inventory, it is determined whether to generate a replenishment prompt.

8. The method for automated management of remote breeding orders based on the Internet of Things according to claim 7, characterized in that: Performing multi-dimensional time series feature extraction on the time series of the feed inventory to obtain a time series correlation feature vector of the absolute amount of feed inventory and a time series correlation feature vector of the feed inventory consumption, including: Calculating the difference between the feed inventories at every two adjacent times in the time series of the feed inventory to obtain a time series of feed consumption; The time series of the feed inventory and the time series of the feed consumption are respectively time-series encoded to obtain the time series correlation feature vector of the absolute amount of the feed inventory and the time series correlation feature vector of the feed inventory consumption.

Images