Grain storage management method and device and related equipment

Abstract

The invention provides a grain storage management method and device and related equipment, and the method comprises the steps: constructing a grain condition database containing multi-dimensional storage information, including the static archive information of a warehouse area and a warehouse, the grain inventory information, and the grain condition environment data collected in real time; according to the static archive information and the grain inventory information, a differentiated multi-parameter early warning rule is configured; based on the grain condition environment data, performing real-time monitoring analysis by using a multi-parameter early warning rule, when a monitoring index triggers an early warning threshold value, generating an early warning signal, locking an abnormal area for early warning, obtaining meteorological data of an abnormal reservoir area, and combining the grain condition environment data and the characteristics of the abnormal area to obtain the abnormal reservoir area. And acquiring a target disposal scheme by using a preset disposal decision model to perform disposal management, such as ventilation disposal management. The problems of uneven ventilation, high energy consumption and grain quality reduction caused by extensive management in a traditional grain storage mode are effectively solved, and accurate monitoring and intelligent ventilation decision making of grain storage are achieved.

Description

Technical Field

[0001] This application relates to the fields of agricultural Internet of Things and smart warehousing technology, and in particular to a grain storage management method, device and related equipment. Background Technology

[0002] With global population growth and climate change, food security has become a global focus. Grain storage is crucial for ensuring food quality and safety; however, traditional grain storage management methods have many shortcomings. Current technology often relies on regular manual inspections in grain depots, leading to a lag in assessing grain conditions. There is a lack of sophisticated methods for controlling humidity and temperature, making grain susceptible to mold, pests, or quality deterioration. Furthermore, traditional ventilation operations are largely based on the subjective experience of management personnel, easily resulting in excessive, "flood-like" ventilation. This not only wastes energy but can also cause unnecessary moisture loss from the grain or condensation problems due to improper ventilation timing (such as ventilating when external humidity is too high).

[0003] Therefore, the industry urgently needs a grain storage management method that can integrate multi-parameter information, provide accurate early warnings, and offer scientific ventilation recommendations. Summary of the Invention

[0004] In view of this, embodiments of this application provide a grain storage management method, apparatus and related equipment to at least partially solve the above problems.

[0005] In a first aspect, embodiments of this application provide a grain storage management method, including: Construct a grain condition database containing multi-dimensional warehousing information, including static archive information of storage areas and warehouses, grain inventory information, and real-time collected grain condition environmental data; Based on the static archive information and grain inventory information, differentiated multi-parameter early warning rules are configured for different warehouses; Based on the grain condition and environmental data, the multi-parameter early warning rules are used for real-time monitoring and analysis. When the monitoring indicators trigger the early warning threshold, an early warning signal is generated and the abnormal area is locked. In response to the warning signal, the meteorological data of the current storage area is obtained, and combined with the grain condition and environmental data and the characteristics of the abnormal area, the data is input into a preset disposal decision model to obtain a target disposal plan. The disposal management shall be carried out in accordance with the target disposal plan, which shall include at least a ventilation disposal strategy.

[0006] Secondly, based on the grain storage management method described in the first aspect of this application, embodiments of this application also provide a grain storage management device, comprising: The construction module is used to build a grain condition database containing multi-dimensional warehousing information, including static archive information of storage areas and warehouses, grain inventory information, and real-time collected grain condition environment data. The configuration module is used to configure differentiated multi-parameter early warning rules for different warehouses based on the static archive information and grain inventory information. The monitoring module is used to perform real-time monitoring and analysis based on the grain condition and environmental data and the multi-parameter early warning rules. When the monitoring indicators trigger the early warning threshold, an early warning signal is generated and the abnormal area is locked. The analysis module is used to respond to the early warning signal, acquire the current meteorological data of the storage area, and combine it with the grain condition and environmental data and the characteristics of the abnormal area, and input it into the preset disposal decision model to obtain the target disposal plan; An execution module is used to manage the disposal according to the target disposal plan, which includes at least a ventilation disposal strategy.

[0007] Thirdly, embodiments of this application also provide a computer storage medium storing computer-executable instructions, which, when executed, perform any of the grain storage management methods described in the first aspect of embodiments of this application.

[0008] Fourthly, embodiments of this application also provide an electronic device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction that causes the processor to execute any of the grain storage management methods described in the first aspect of the embodiments of this application.

[0009] This application provides a grain storage management method, apparatus, and related equipment. It constructs a grain condition database containing multi-dimensional storage information, including static archives of the storage area and warehouses, grain inventory information, and real-time collected grain condition environmental data. Based on the static archives and grain inventory information, differentiated multi-parameter early warning rules are configured for different warehouses. Based on the grain condition environmental data, real-time monitoring and analysis are performed using the multi-parameter early warning rules. When a monitoring indicator triggers an early warning threshold, an early warning signal is generated, and an abnormal area is identified. In response to the early warning signal, meteorological data for the current storage area is acquired and, combined with the grain condition environmental data and the characteristics of the abnormal area, input into a pre-set disposal decision model to obtain a target disposal plan. Disposal management is performed according to the target disposal plan, which includes at least a ventilation disposal strategy. This disposal management, with the disposal management plan including at least a ventilation disposal strategy, determines the final disposal plan through multi-dimensional information coupling, effectively solving the problems of uneven ventilation, high energy consumption, and decreased grain quality caused by extensive management in traditional grain storage methods. It achieves precise monitoring and intelligent ventilation decision-making for grain storage. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0011] Figure 1 A schematic diagram illustrating the workflow of a grain storage management method provided in this application embodiment; Figure 2 A schematic diagram of the structure of a grain storage management device provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0012] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art should fall within the protection scope of the embodiments of this application.

[0013] It should be understood that the steps described in the method embodiments of this application may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this application is not limited in this respect.

[0014] Example 1 This application provides a method for grain storage management, such as... Figure 1 As shown, Figure 1 This application provides a schematic diagram illustrating the workflow of a grain storage management method according to an embodiment, including: Step S101: Construct a grain condition database containing multi-dimensional storage information. This multi-dimensional storage information includes static archive information of the storage area and warehouses, grain inventory information, and real-time collected grain condition and environmental data. In this embodiment, the static archive information refers to various attributes that do not change in the short term, such as the warehouse structure, design capacity, grain varieties, and storage time. This step in this embodiment, by constructing a multi-dimensional grain condition database, standardizes the data used for storage management, serving as the data foundation for ventilation and other management functions. This facilitates a shift from post-event management to pre-event prediction, better ensuring the safety of stored grain while improving management response efficiency and reducing management costs.

[0015] Specifically, in one optional implementation of this application embodiment, the real-time collected grain condition environmental data includes: temperature distribution data, humidity distribution data, grain moisture content, and coordinates of abnormal point distribution areas inside the grain pile; and one or more data from the external environment of the storage area, including temperature, humidity, wind speed, wind direction, rainfall, air pressure, and light parameters. In the actual application scenario of this application embodiment, this data can be obtained in various ways, such as manually entering information such as the capacity of a warehouse, the number of storage rooms, and location. Grain condition data, including grain temperature, humidity, and moisture content at different depths, can be collected in real time using sensors embedded in the grain pile. Meteorological data such as wind speed and rainfall in the external environment of the storage area can be obtained in real time through external weather stations or preset sensors. This allows for a more accurate determination of which multiple factors comprehensively affect the quality of stored grain, thereby achieving more precise warehouse management.

[0016] Step S102: Based on the static archive information and grain inventory information, configure differentiated multi-parameter early warning rules for different warehouses. In this embodiment, this step is limited to using grain inventory information in addition to static archive information, eliminating the signal-to-noise ratio collapse caused by the "overgeneralization" of static rules when setting early warning rules solely based on static archives. For example, if a uniform rule is used across the entire warehouse area, such as triggering an alarm when the grain temperature exceeds 25℃, 25℃ might be the normal heating temperature for stored grain in a new standard warehouse with good insulation, while in an older standard warehouse with poor insulation, condensation might occur at 23℃ in both chambers, leading to missed or false alarms. Therefore, this embodiment, in addition to using static archive information as the basis for setting early warning rules, also introduces differentiated grain inventory information, making the early warning rules for different storage warehouses more closely aligned with their actual conditions, thereby improving the accuracy of early warnings based on the configured differentiated multi-parameter early warning rules. This enables the construction of an agricultural Internet of Things (IoT) digital nerve ending with "one policy per warehouse," giving the system intelligent management functions such as precise perception, accurate decision-making, and precise execution.

[0017] Specifically, in an optional implementation of this application embodiment, differentiated multi-parameter early warning rules are configured for different warehouses based on the static archive information and grain inventory information. This includes: determining the variety characteristics, origin, and current storage month of the grain in different warehouses based on the static archive information and grain inventory information; setting a threshold set corresponding to the different warehouse levels based on the variety characteristics, origin, and current storage month; wherein the threshold set includes at least: a maximum temperature threshold, an average temperature threshold, a maximum humidity threshold, an average humidity threshold, a maximum moisture content threshold, an average moisture content threshold, and an abnormal area percentage threshold; when any indicator or combination of indicators in the real-time collected grain condition environmental data exceeds the corresponding threshold set, an alarm is triggered at different severity levels. In the actual application scenario of this application embodiment, the temperature, humidity, and moisture tolerance for safe storage of different varieties, origins, and storage ages of grain varies greatly (e.g., the mold risk of Northeast japonica rice and Southern indica rice, Northern wheat and Southern wheat, and new grain and old grain are completely different). Traditional, uniform thresholds (such as 14.5% moisture, 25℃, and 70% RH nationwide) easily lead to situations where high-risk grain piles fail to trigger timely alarms, while low-risk grain piles frequently trigger false alarms. This solution, by determining the variety characteristics, origin, and current storage month of the grain in different warehouses, achieves "tailored to grain conditions and dynamic thresholds," truly enabling precise prevention and control. Alarms are concentrated on genuinely high-risk grain piles, significantly reducing the probability of major grain storage accidents. Furthermore, traditional fixed thresholds have a relatively high false alarm rate, causing warehouse managers to be overwhelmed by false alarms while truly serious problems are overlooked. This solution sets thresholds through a multi-dimensional combination of variety, origin, and age, further reducing the false alarm rate. Managers can then focus only on alarms that truly require attention, further improving response efficiency. The limited threshold set includes multiple levels of indicators such as maximum value, average value, and abnormal area percentage, which can be combined for judgment (e.g., average temperature is not exceeded, but hot spot temperature exceeds and abnormal area percentage >5%, still considered serious), enabling true tiered early warning and early intervention for problems. The system can also automatically tighten or loosen moisture and temperature thresholds based on the current storage month (e.g., for rice stored for more than 18 months, the moisture threshold is automatically lowered by 0.3-0.5 percentage points), facilitating timely rotation by managers and preventing severe quality degradation caused by over-storage. When a specific combination of thresholds is reached, the system activates the ventilation system for precise ventilation, achieving closed-loop management under unattended conditions and effectively reducing the intensity of manual operations. This enhances the system's intelligence.

[0018] Step S103: Based on the grain condition data, real-time monitoring and analysis are performed using the multi-parameter early warning rules. When the monitored indicators trigger the early warning threshold, an early warning signal is generated and the abnormal area is locked. For example, for the condition "northern region, stored corn, November," an early warning rule is set that the highest grain temperature threshold is no higher than 20℃ and the average moisture threshold is no higher than 14%. When the system determines, based on the real-time acquired grain condition data, that the temperature measured at the grain temperature measuring point in warehouse A of that region reaches 22℃ (exceeding the 20℃ threshold), the system determines it as abnormal, locks warehouse A in that region, and triggers an alarm push. This allows managers to quickly identify abnormal storage areas and improves the automation and intelligence of warehouse management.

[0019] Step S104: In response to the warning signal, acquire the current meteorological data of the storage area, and combine it with the grain condition environmental data and the characteristics of the abnormal area, inputting it into a preset disposal decision model to obtain a target disposal plan. In this embodiment, this step does not directly determine the target disposal plan based on the warning signal, but rather requires combining real-time meteorological data of the current storage area with anomalies, grain environmental data, and characteristic information of the abnormal storage area as the final basis for obtaining the target disposal plan. This essentially upgrades the handling of stored grain from "passive alarm + manual experience-based decision-making" to a crucial step of "intelligent closed-loop precise disposal." It solves the two most critical technical pain points of traditional early warning grain management systems: "knowing something is going wrong, but not knowing how to handle it optimally" and "the optimal handling plan for the same warning signal is completely different under different weather conditions and different abnormal modes." In the practical application scenarios of this application embodiment, the preset strategy determined in steps S102-S103 can only achieve a high probability of correctness, belonging to a coarse-grained, experience-based, one-size-fits-all approach. However, by further introducing real-time weather forecasts, spatial characteristics of abnormal areas (such as hotspot locations, areas, and heating rates), and the current state of the grain pile, and inputting thresholds to create a disposal decision model for in-depth decision-making, a more refined, dynamic, and scenario-based optimal solution can be achieved. This further evolves the system implementing the method described in this embodiment from the traditional "alarm-like" to "diagnostic," achieving precise management of stored grain that is "most cost-effective, least damaging to the grain, and has the highest success rate." It eliminates the reliance on human experience with varying management levels and significantly improves the efficiency of determining the target disposal plan.

[0020] Optionally, in one implementation of this application embodiment, when obtaining the target treatment plan, the method further includes: introducing a time series prediction model (such as LSTM) to predict the grain temperature trend in a future preset time period, such as the next 24 or 48 hours, based on a preset sequence of grain temperature and warehouse temperature changes over a historical time period. Correspondingly, the step of combining the grain condition data and the characteristics of the abnormal area and inputting them into a preset treatment decision model to obtain the target treatment plan further includes: using the grain temperature trend information as one of the characteristics of the abnormal area and inputting it into the preset treatment decision model to obtain a target treatment plan that is more closely aligned with the future preset time period. This allows for more accurate early warning. For example, even if the current temperature is not exceeded, if the model predicts that condensation risk will occur after a preset time based on the current warming rate, preventative ventilation suggestions can be generated in advance before the risk occurs, truly achieving prevention rather than just post-event remediation. In addition, since the acquired meteorological data also has a certain degree of predictability for the future, the model can also support making decisions in advance based on meteorological data for a period of time in the future. For example, based on the weather forecast for the next 48 hours, it can determine the handling strategy of "ventilating for 2 hours today is enough, and there is no need to turn it on tomorrow because there is cold air", so as to avoid ineffective ventilation and reduce the operating energy consumption of ventilation equipment fans.

[0021] Optionally, in one implementation of this application embodiment, the target disposal plan includes at least one type of preventative disposal plan, emergency disposal plan, and industry-specific disposal plan; each type of plan includes at least one or more of the following information: ventilation fan start-up time, runtime, ventilation intensity, and expected energy consumption assessment. This step in this application embodiment defines the type of target disposal plan generated to address different actual storage anomalies. The preventative disposal plan refers to situations where current storage indicators such as grain temperature are normal, but hot and humid air or cold and dry air is expected to arrive in the next 3-7 days, requiring either complete lack of ventilation (missing the golden opportunity for ventilation) or a hasty decision to open ventilation. Specifically, under normal circumstances, missing 3-5 optimal opportunities for natural cooling / precipitation per year leads to a cumulative increase in grain temperature each month, resulting in significant energy consumption and losses in grain quality. Configuring such a preventative disposal plan to effectively address this situation can effectively reduce storage costs and extend the shelf life of stored grain. Emergency response plans refer to measures taken when grain piles in storage exhibit severe abnormalities such as high temperature or high humidity. These plans aim to prevent managers from hastily activating all ventilation equipment, which could lead to hot / humid air backflow, excessive moisture loss, cracking, or warehouse bursts. Industry-wide response plans, on the other hand, address policy-driven management practices such as regional rotation, moisture balancing before fumigation, and "final balancing before sealing warehouses" during the summer heat. These policies often result in a uniform policy across the region: "Open all windows for ventilation on a specific day," regardless of the specific grain conditions. For example, some warehouses already have high grain temperatures, and ventilation exacerbates the problem; others have low moisture content, and ventilation dries them out. These plans aim to reduce the occurrence of widespread grain spoilage incidents. This application's embodiments, by generating one or more of these three target response plans, enable more precise grain storage management and further improve storage intelligence. These three types of response plans, combined with parameterized configuration, transform traditional management practices, such as ventilation management, from tedious manual labor into truly intelligent and executable solutions. The correct customs clearance management strategy has been transformed from relying on human experience into a system-executable and deterministic action, preventing situations where desperate attempts to save grain lead to reckless actions that could damage the grain.

[0022] Optionally, in one embodiment of this application, the step of acquiring the current meteorological data of the storage area, combining it with the grain condition environmental data and the characteristics of the abnormal area, and inputting it into a pre-set disposal decision model to obtain a target disposal plan includes: vectorizing the real-time meteorological data and the grain condition environmental data of the abnormal area to construct an input feature vector; inputting the input feature vector into a pre-trained embedded disposal decision model; this model is deployed on a local edge computing unit, and its multi-layer network structure mines the nonlinear mapping relationship between input features and historical control cases through end-to-end learning, and abstracts high-level semantic features layer by layer during forward propagation; and decodes and generates the target disposal plan based on the prediction results of the output layer of the disposal decision model. This step in this embodiment of the application constructs a lightweight neural network model deployed on an edge device (such as a local grain storage controller), and establishes an intelligent mapping mechanism of "environmental input → disposal plan output" by learning a large number of historical control cases. The entire process requires no manual rule intervention and has adaptive, high-precision, and low-latency characteristics.

[0023] In the practical application scenarios of this application embodiment, end-to-end learning refers to configuring the model to directly use raw or pre-processed input features (such as vectorized data like temperature, humidity, CO2 concentration, and wind speed) as input, and the final ventilation control parameters (such as fan start / stop, airflow level, and duration) as output. It does not rely on manually designed rules, intermediate variables, or staged judgment logic; all optimization is automatically completed by the internal parameters of the neural network. Historical grain warehouse operation data is collected, including data corresponding to each record: Input side: real-time meteorological data (such as outdoor temperature and humidity, air pressure), and internal environmental data of the grain pile (such as temperature, humidity, and pest indicators of each layer). Output side: The ventilation type (such as cooling ventilation, precipitation ventilation, recirculation fumigation, etc.) and specific control parameters used at the time.

[0024] The above data was organized into supervised learning sample pairs (xi, yi), where xi is the input data feature vector and yi is the multi-dimensional output data label. The dataset containing these supervised learning sample pairs was used to train the embedded disposal decision model (neural network model) end-to-end. The loss function during training comprehensively considered the ventilation type classification accuracy and the regression error of control parameters (e.g., cross-entropy + MAE combination). The trained model effectively avoids the probability of "rule fragmentation" and difficulty in covering boundary conditions in traditional expert systems, enabling the model to capture complex conditions and make better decision patterns, improving the model's generalization ability. Simultaneously, the embedded neural network model simplifies the system structure for implementing this method, facilitating manual parameter adjustment and optimization, and improving the model's output accuracy. Furthermore, researchers found that the relationship between the grain environment and disposal plans is not a simple linear one (e.g., "turn on the fan when the temperature is high"), but is influenced by multiple coupled factors (e.g., grain type, storage type, season, dew point risk, etc.). In this embodiment, the neural network, by introducing nonlinear activation functions (such as ReLU and Swish) and a multi-layer structure, can better fit arbitrarily complex nonlinear functions, thereby accurately characterizing this high-dimensional, nonlinear input-output dependency. For example, the embedded decision-making model is configured with at least three fully connected layers (or lightweight convolutional / attention modules), each followed by a nonlinear activation. During model training, the weights are continuously adjusted through backpropagation, so that the model output closely approximates the desired treatment plan. This improves the robustness of the final target treatment plan and avoids the probability of misoperation. In this embodiment, the layer-by-layer abstraction of high-level semantic features during forward propagation refers to the process where, during the forward propagation from the input layer to the output layer, each hidden layer transforms the representation of the previous layer, gradually extracting more discriminative and semantically meaningful features. For example, shallow networks capture local patterns in raw sensor data (such as sudden temperature changes or humidity gradients in a certain layer of grain); mid-layer networks combine multiple shallow features to form composite semantics (such as "dry and hot in the upper layer, wet and cold in the lower layer"); deep networks can extract high-level semantics directly related to ventilation decisions (such as "there is a risk of condensation, intermittent low-volume ventilation is required"). This achieves a semantic leap from raw data to "executable strategies" and gives the model contextual understanding capabilities similar to expert experience. It provides a foundation for subsequent interpretability analysis (such as visualization of feature importance), making this solution more valuable for application.

[0025] Step S105: Perform disposal management according to the target disposal plan, wherein the disposal management plan includes at least a ventilation disposal strategy.

[0026] Furthermore, in an optional implementation of this application embodiment, the specific process of ventilation management according to the target treatment plan can be as follows: the control parameters in the target treatment plan are converted into equipment control commands through the Internet of Things (IoT) gateway of the system, and remotely sent to the corresponding warehouse controller, so as to control the warehouse controller to adjust the opening and ventilation speed of the ventilation equipment according to the equipment control commands, thereby realizing unattended automated operation and improving the automation level of the method described in this embodiment.

[0027] Optionally, in one possible implementation of this application embodiment, the method further includes: after the execution of instructions for handling management according to the target handling plan, wherein the handling management plan includes at least a ventilation handling strategy, collecting data on changes in the grain condition environment after execution; determining the improvement effect of the ventilation management operation on the temperature and humidity of the grain pile based on the data on changes in the grain condition environment, and generating an effect evaluation report; storing the initial conditions, operating parameters, and effect evaluation report of this ventilation into the safe handling experience database to correct the weight parameters of the handling decision model, thereby optimizing the parameters of the handling decision model; and using the optimized ventilation decision model to support subsequent ventilation decisions. This step in this application embodiment transforms discrete human experience (ventilation experience management) into reusable algorithm model data, enabling the system implementing this solution to possess up-to-date expert attributes, avoiding fluctuations in management level due to personnel changes, and ensuring the reliability of subsequent handling plans.

[0028] Furthermore, in an optional implementation of this application embodiment, when constructing a multi-dimensional grain condition database, the grain inventory information also includes quality fingerprint data for each batch of grain. The quality fingerprint data includes: bulk density, percentage of imperfect grains, fatty acid value, and mold rate. Correspondingly, when determining the target disposal plan, the disposal decision model can also use the quality fingerprint data as a correction coefficient to adjust the output target disposal plan. For example, when the mold rate or fatty acid value shows an upward trend, the temperature threshold for triggering ventilation recommendations is lowered.

[0029] Furthermore, in an optional implementation of this application embodiment, the method further includes: periodically performing cluster analysis on historical early warning records to statistically analyze the early warning trigger frequency of different warehouses and different areas; if the early warning frequency of a specific area is identified as significantly higher than the average level, then generating warehouse structure inspection suggestions or airtightness testing suggestions for that specific area. This step in this application embodiment limits the solution to not only generating ventilation suggestions but also providing assistance for the maintenance and management of warehouse facilities, thereby further improving the applicability of the method described in this embodiment.

[0030] Optionally, in one embodiment of this application, the method further includes: displaying the early warning status trend information between various storage cells in the storage area in the form of charts on a visual interface; and displaying generated ventilation suggestions based on the early warning status trends, wherein the ventilation suggestions include targeted treatment measures for abnormal areas. For example, real-time monitoring of the early warning status between various cells in the storage area and displaying these trend information in the form of charts allows for real-time changes in the storage environment, enabling managers to take intuitive and timely countermeasures to better ensure the safety of stored grain.

[0031] This embodiment improves the efficiency of human-computer interaction during the implementation of this solution by limiting the interaction method, allowing managers to intuitively understand complex data and quickly locate problematic grain conditions and make decisions.

[0032] Optionally, in one embodiment of this application, the method further includes: assigning differentiated management operation permissions according to the roles of system users, and recording user login logs and operation logs in real time; wherein, the differentiated management operation permissions include at least the permission to modify early warning rules, the permission to remotely operate warehouse environment control equipment, and the permission to view logs. This embodiment of the application uses this step to ensure the security and accountability of the system implementing this solution, reduce security incidents caused by misoperation, and better guarantee the smooth operation of the system.

[0033] This application provides a grain storage management method. It constructs a grain condition database containing multi-dimensional storage information, including static archives of the storage area and warehouses, grain inventory information, and real-time collected grain condition and environmental data. Based on the static archives and grain inventory information, differentiated multi-parameter early warning rules are configured for different warehouses. Based on the grain condition and environmental data, real-time monitoring and analysis are performed using the multi-parameter early warning rules. When a monitoring indicator triggers an early warning threshold, an early warning signal is generated, and an abnormal area is identified. In response to the early warning signal, meteorological data for the current storage area is acquired and, combined with the grain condition and environmental data and the characteristics of the abnormal area, input into a pre-set disposal decision model to obtain a target disposal plan. Disposal management is carried out according to the target disposal plan, which includes at least a ventilation disposal strategy. This ventilation management method, through multi-dimensional information coupling, determines the final disposal plan, effectively solving the problems of uneven ventilation, high energy consumption, and decreased grain quality caused by extensive management in traditional grain storage methods. It achieves precise monitoring and intelligent ventilation decision-making for grain storage.

[0034] Example 2 Based on the grain storage management method provided in Embodiment 1 of this application, this embodiment also provides a corresponding grain storage management device, such as... Figure 2 As shown, Figure 2This is a schematic diagram of the structure of a grain storage management device 20 provided in an embodiment of this application. The grain storage management device 20 includes: The construction module 201 is used to construct a grain condition database containing multi-dimensional storage information, including static archive information of storage areas and warehouses, grain inventory information, and real-time collected grain condition environment data. Configuration module 202 is used to configure differentiated multi-parameter early warning rules for different warehouses based on the static archive information and grain inventory information; The monitoring module 203 is used to perform real-time monitoring and analysis based on the grain condition and environmental data and the multi-parameter early warning rules. When the monitoring indicators trigger the early warning threshold, an early warning signal is generated and the abnormal area is locked. Analysis module 204 is used to respond to the early warning signal, acquire the current meteorological data of the storage area, and combine the grain condition and environmental data with the characteristics of the abnormal area, inputting the data into a preset disposal decision model to obtain a target disposal plan; The execution module 205 is used to perform disposal management according to the target disposal plan, wherein the disposal management plan includes at least a ventilation disposal strategy.

[0035] Optionally, in one embodiment of this application, the real-time collected grain environmental data includes: temperature distribution data, humidity distribution data, grain moisture content, and coordinates of abnormal point distribution areas inside the grain pile; as well as temperature, humidity, wind speed, wind direction, rainfall, air pressure, and light parameters of the external environment of the storage area.

[0036] Optionally, in one embodiment of this application, the configuration module 202 is further configured to: determine the variety characteristics, origin, and current storage month of grain in different storage locations based on the static archive information and grain inventory information; set a threshold set corresponding to the different storage locations based on the variety characteristics, origin, and current storage month; wherein the threshold set includes at least: a maximum temperature threshold, an average temperature threshold, a maximum humidity threshold, an average humidity threshold, a maximum moisture content threshold, an average moisture content threshold, and an abnormal area percentage threshold; when any indicator or combination of indicators in the real-time collected grain condition and environmental data exceeds the corresponding threshold set, an alarm is triggered at different severity levels.

[0037] Optionally, in one embodiment of this application, the analysis module 204 is further configured to vectorize the real-time meteorological data and the grain condition and environmental data of the abnormal area to construct an input feature vector; input the input feature vector into a pre-trained embedded disposal decision model; the model is deployed on a local edge computing unit, and its multi-layer network structure mines the nonlinear mapping relationship between the input features and historical control cases through end-to-end learning, and abstracts high-level semantic features layer by layer during the forward propagation process; and decodes and generates the target disposal plan based on the prediction results of the output layer of the disposal decision model.

[0038] Optionally, in one embodiment of this application, the target treatment plan includes at least one of a preventive treatment plan, an emergency treatment plan, and an industry-specific treatment plan; each plan includes at least one or more of the following information: ventilation fan start-up time, runtime, ventilation intensity, and expected energy consumption assessment.

[0039] Optionally, in one embodiment of this application, the device 20 further includes an iteration module (not shown in the figures). This iteration module is used to collect data on changes in the grain condition environment after the execution of instructions for handling management according to the target handling plan, which includes at least a ventilation handling strategy; determine the improvement effect of the ventilation management operation on the temperature and humidity of the grain pile based on the data on changes in the grain condition environment, and generate an effect evaluation report; store the initial conditions, operating parameters, and effect evaluation report of this ventilation in the safe handling experience database to correct the weight parameters of the handling decision model, thereby optimizing the parameters of the handling decision model; and use the optimized ventilation decision model to support subsequent ventilation decisions.

[0040] Optionally, in one embodiment of this application, the device 20 further includes an interaction module (not shown in the figures), which is used to display the early warning status trend information between various storage warehouses in the storage area in the form of charts on a visual interface; based on the early warning status trend, it displays the generated ventilation suggestions, which include targeted treatment measures for abnormal areas.

[0041] Optionally, in one embodiment of this application, the device 20 further includes an allocation module (not shown in the figures), which is used to allocate differentiated management operation permissions according to the role of the system user, and record the user's login log and operation log in real time; wherein, the differentiated management operation permissions include at least the permission to modify the warning rules, the permission to remotely operate the warehouse environment control equipment, and the permission to view the logs.

[0042] This application provides a grain storage management device. A construction module builds a grain condition database containing multi-dimensional storage information, including static archives of the storage area and warehouses, grain inventory information, and real-time collected grain condition environmental data. A configuration module configures differentiated multi-parameter early warning rules for different warehouses based on the static archives and grain inventory information. A monitoring module performs real-time monitoring and analysis based on the grain condition environmental data and the multi-parameter early warning rules. When a monitoring indicator triggers an early warning threshold, an early warning signal is generated and an abnormal area is identified. An analysis module responds to the early warning signal by acquiring the current meteorological data of the storage area and, combined with the grain condition environmental data and the characteristics of the abnormal area, inputs it into a pre-set disposal decision model to obtain a target disposal plan. An execution module performs disposal management according to the target disposal plan, which includes at least a ventilation disposal strategy. This ventilation management device has a relatively simple structure. Through multi-dimensional information coupling, it determines the final disposal plan, effectively solving the problems of uneven ventilation, high energy consumption, and decreased grain quality caused by extensive management in traditional grain storage methods. It achieves precise monitoring and intelligent ventilation decision-making for grain storage.

[0043] Example 3 This application also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements any of the grain storage management methods described in the foregoing embodiment one of this application. Example 4 This application also provides an electronic device, such as... Figure 3 As shown, Figure 3 This application provides a schematic diagram of the structure of an electronic device 30, which includes: One or more processors 301, communication interface 302, memory 303 and communication bus 304, the processors 301, memory 303 and communication interface 302 communicate with each other through communication bus 304; Memory 303 is used to store one or more programs; When the one or more programs are executed by the one or more processors 301, the one or more processors 301 implement any of the grain storage management methods described in Embodiment 1 of this application.

[0044] This application has now described specific embodiments of the subject matter. In some cases, the actions described in the claims can be performed in a different order and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing can be advantageous.

[0045] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system layer onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0046] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0047] The system layers, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0048] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0049] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0050] Those skilled in the art will understand that embodiments of this application can be provided as methods, system-level, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0051] This application can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific transactions or implement specific abstract data types. This application can also be practiced in distributed computing environments where transactions are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0052] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system-level embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0053] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for managing a grain storage, characterized by, The method comprises the following steps: constructing a grain condition database containing multi-dimensional warehouse information, which includes static archive information of warehouse area and warehouse, grain inventory information, and real-time collected grain condition environment data; configuring differentiated multi-parameter early warning rules for different warehouses according to the static archive information and the grain inventory information; based on the grain condition environment data, using the multi-parameter early warning rules for real-time monitoring and analysis, when the monitoring index triggers the early warning threshold, generating an early warning signal and locking the abnormal area; in response to the early warning signal, acquiring the current weather data of the warehouse area, and combining the grain condition environment data and the characteristics of the abnormal area, inputting into the preset disposal decision model to obtain a target disposal scheme; disposing and managing according to the target disposal scheme, and the disposal and management scheme at least includes a ventilation disposal strategy.

2. The grain storage management method according to claim 1, characterized by, The real-time collected grain condition environment data includes: temperature distribution data, humidity distribution data, grain moisture content, and abnormal point distribution area coordinates inside the grain pile; and one or more data of temperature, humidity, wind speed, wind direction, rainfall, air pressure, and illumination parameters outside the warehouse area.

3. The grain storage management method according to claim 1, characterized by, The method of configuring differentiated multi-parameter early warning rules for different warehouses according to the static archive information and the grain inventory information comprises: determining the variety characteristics, origin and current storage month of the grain in different warehouses according to the static archive information and the grain inventory information; setting a hierarchical threshold set corresponding to different warehouses according to the variety characteristics, origin and current storage month; wherein the threshold set at least includes: maximum temperature threshold, average temperature threshold, maximum humidity threshold, average humidity threshold, maximum moisture threshold, average moisture threshold and abnormal area proportion threshold; when any index or combined index in the real-time collected grain condition environment data exceeds the corresponding threshold set, the alarm level of different severity is triggered to perform alarm.

4. The grain storage management method according to claim 1, characterized by, The method of acquiring the current weather data of the warehouse area, combining the grain condition environment data and the characteristics of the abnormal area, and inputting into the preset disposal decision model to obtain a target disposal scheme comprises: performing vectorization processing on the real-time weather data and the grain condition environment data of the abnormal area to construct an input feature vector; inputting the input feature vector into a pre-trained embedded disposal decision model; the model is deployed in a local edge computing unit, and a multi-layer network structure thereof mines the nonlinear mapping relationship between the input features and historical control cases through an end-to-end learning manner, and abstracts high-level semantic features layer by layer in a forward propagation process; decoding the target disposal scheme according to the prediction result of the output layer of the disposal decision model.

5. The grain storage management method according to claim 4, characterized by, The target disposal scheme includes at least one of a preventive disposal scheme, an emergency disposal scheme and an industry disposal scheme; each scheme at least contains one or more information of opening time information, running time information, ventilation intensity information and predicted energy consumption evaluation information of the ventilator.

6. The grain storage management method according to claim 1, characterized by, The method further comprises: after the instruction of disposing and managing according to the target disposal scheme is executed, collecting the grain condition environment change data after execution; According to the grain condition environment change data, the improvement effect of the treatment management operation on the temperature and humidity of the grain pile is determined, and an effect evaluation report is generated; The initial conditions, operation parameters and effect evaluation report of the current ventilation are stored in the safe treatment experience library to correct the weight parameters of the treatment decision model and optimize the parameters of the treatment decision model; The optimized ventilation decision model is used to support subsequent ventilation decision.

7. The grain storage management method according to claim 1, characterized by, The method further comprises: The trend information of the early warning state of each warehouse in the storage area is displayed in a chart form on the visualization interface; Based on the early warning state trend, the generated ventilation suggestion is associated and displayed, and the ventilation suggestion includes specific treatment measures for the abnormal area.

8. The grain storage management method according to claim 1, characterized by, The method further comprises: According to the role of the system user, differentiated management operation permissions are assigned, and the login log and operation log of the user are recorded in real time; The differentiated management operation permissions at least include the modification permission of the early warning rule, the remote operation permission of the warehouse environment control device and the log viewing permission.

9. A grain storage management device, characterized by, Comprise: The construction module is configured to construct a grain condition database containing multi-dimensional warehouse information, including static file information of the storage area and warehouse, grain inventory information, and real-time collected grain condition environment data; The configuration module is configured to configure differentiated multi-parameter early warning rules for different warehouses according to the static file information and grain inventory information; The monitoring module is configured to perform real-time monitoring analysis based on the grain condition environment data and using the multi-parameter early warning rules, and generate an early warning signal and lock an abnormal area when a monitoring index triggers an early warning threshold; The analysis module is configured to obtain meteorological data of the current storage area in response to the early warning signal, and input the meteorological data, the grain condition environment data and the characteristics of the abnormal area into a preset treatment decision model to obtain a target treatment scheme; The execution module is configured to perform treatment management according to the target treatment scheme, and the treatment management scheme at least includes a ventilation treatment strategy.

10. A computer storage medium, characterized in that, The computer storage medium stores computer executable instructions, and the computer executable instructions are executed to perform the grain storage management method of any one of claims 1-8. The computer storage medium stores computer executable instructions, and the computer executable instructions are executed to perform the grain storage management method of any one of claims 1-8.

Images