A method and system for intelligently monitoring the inventory of a grain reserve

Abstract

The application relates to the technical field of grain reserve management, and discloses a kind of intelligent supervision method and system of grain reserve warehouse, method includes: obtaining the historical data of each sensor from storage unit, constructs grain quality prediction model, trains grain quality prediction model based on historical data, collects the data of each sensor in granary as first data, inputs first data into grain quality prediction model, obtains first prediction result, based on first prediction result, environmental regulation unit adjusts the environment of granary and / or logistics scheduling unit schedules the in and out of grain.Based on first preset period, the data of each sensor is collected again as second data, the difference between second prediction result and first prediction result is compared, the environment of granary is readjusted by environmental regulation unit and / or the in and out of grain is rescheduled by logistics scheduling unit, the application realizes the intelligent management of grain reserve warehouse, and improves the security of grain storage.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of grain reserve management, and in particular to a grain reserve warehouse inventory intelligent supervision method and system. BACKGROUND

[0002] With the expansion of the size of the grain warehouse, mold and insect problems are prone to occur inside the grain warehouse. Since grain is a special and complex living body, the temperature field, humidity field and gas concentration in the grain pile are abnormally complex. Therefore, accurately grasping the temperature and humidity distribution and gas concentration in the grain warehouse has become one of the important methods for predicting the quality of stored grain.

[0003] A similar prior art is Chinese patent application No. CN118536090A, which relates to a method for generating a grain temperature field map based on warehouse temperature and humidity monitoring, comprising: drawing a spatial temperature and humidity distribution map of the interior of the grain warehouse in a plurality of monitoring periods; drawing a spatial grain warehouse temperature and humidity history distribution map in a plurality of historical monitoring periods; obtaining a training sample set according to the spatial temperature and humidity distribution map and the temperature and humidity history distribution map; training a grain temperature field map temperature state prediction model according to the training sample set; and predicting the temperature state of the corresponding grain warehouse in the monitoring period according to the grain temperature field map temperature state prediction model.

[0004] A similar prior art is Chinese patent application No. CN118469310A, which discloses a grain warehouse inventory management method and system, relating to the technical field of grain warehouse management. In this method, the grain warehouse is divided into regions to obtain a plurality of inventory management regions; the type of grain stored in the inventory management region is determined; the initial inventory quantity and the starting storage time corresponding to the grain type are obtained; the detection image corresponding to the grain type is collected by the camera device arranged in the inventory management region, and a plurality of initial environment data collected by the sensor arranged in the inventory management region are obtained; based on the initial inventory quantity, the starting storage time, the detection image and the plurality of initial environment data, the grain inventory situation corresponding to the grain type is obtained; the grain inventory situation includes a first grain inventory and a second grain inventory, the first grain inventory is a grain inventory that supports continued storage, and the second grain inventory is a grain inventory that is deteriorating or near expiration.

[0005] However, the above two applications are not efficient enough in grain inventory management, and the safety of grain storage needs to be further improved in the intelligent management of grain reserve warehouses. SUMMARY

[0006] To solve the above technical problems, the present application provides a grain reserve warehouse inventory intelligent supervision method and system to improve the safety and management efficiency of grain storage.

[0007] To achieve the above object, in a first aspect, the application provides a method for intelligently monitoring the inventory of a grain reserve warehouse, wherein a temperature sensor, a humidity sensor and a camera device are respectively installed in each grain storehouse, a gas concentration sensor is installed at a preset height from the bottom of the grain storehouse, and a weight sensor is installed on the bottom of the grain storehouse or a support structure, and the method for intelligently monitoring the inventory of the grain reserve warehouse is implemented by performing the following steps:

[0008] Step S1: historical data of each sensor is obtained from a storage unit, and the historical data is preprocessed to obtain preprocessed historical data, wherein the historical data includes temperature data, humidity data, gas concentration data, grain weight data and quality grade of the grain;

[0009] Step S2: a grain quality prediction model is constructed, the grain quality prediction model is trained based on the historical data, and the trained grain quality prediction model is obtained, data of each sensor in the grain storehouse is collected as first data, the first data is preprocessed, and the first data is input into the grain quality prediction model to obtain a first prediction result, and based on the first prediction result, an environment adjustment unit adjusts the environment of the grain storehouse and / or a logistics scheduling unit schedules the entry and exit of the grain;

[0010] Step S3: data of each sensor is collected again as second data based on a first preset period, a second prediction result predicted based on the second data is obtained based on the grain quality prediction model, a difference between the second prediction result and the first prediction result is compared, and based on the difference, the environment adjustment unit re-adjusts the environment of the grain storehouse and / or the logistics scheduling unit re-schedules the entry and exit of the grain.

[0011] In combination with the first aspect, in a first implementation manner of the first aspect of the application, the grain quality prediction model is trained by:

[0012] dividing the historical data into a training data set, a validation data set and a test data set, training the grain quality prediction model using the training data set, adjusting parameters of the grain quality prediction model using the validation data set, evaluating the performance of the grain quality prediction model using the test data set, and obtaining the grain quality prediction model when the accuracy rate of the output result of the grain quality prediction model is greater than or equal to a preset threshold.

[0013] In combination with the first aspect, in a second implementation manner of the first aspect of the application, adjusting the environment of the grain storehouse includes:

[0014] When the first prediction result is lower than the latest quality grade of the grain, the temperature data, humidity data, gas concentration data and grain weight data are compared respectively to obtain a factor causing the first prediction result to decrease, when the decreasing factor is the temperature data, a refrigeration device close to the granary is turned on, when the decreasing factor is the humidity data, a dehumidification device and a ventilation device close to the granary are turned on; when the decreasing factor is the gas concentration data, a controlled atmosphere storage technology is used, and when the decreasing factor is the grain weight data, the grain in the granary is dispatched to be taken out of the granary or part of the grain in the granary is transferred to other granaries.

[0015] With reference to the first aspect, in a third implementation manner of the first aspect of the present application, adjusting the environment of the granary further includes:

[0016] When there are multiple factors combined to affect, based on the combination of the factors, the corresponding device is adjusted, and data of each sensor is collected based on a second preset period, a new prediction result is obtained based on the grain quality prediction model, and the environment of the granary is continuously adjusted based on the new prediction result, wherein the second preset period is less than the first prediction period.

[0017] With reference to the first aspect, in a fourth implementation manner of the first aspect of the present application, dispatching the grain to enter or exit includes:

[0018] When the first prediction result is higher than or equal to the latest quality grade of the grain, the grain of the same type as the grain in the granary is unloaded into the granary, otherwise the grain is unloaded into other granaries or newly-built granaries.

[0019] With reference to the first aspect, in a fifth implementation manner of the first aspect of the present application, dispatching the grain to enter or exit further includes:

[0020] After the environment of the granary is adjusted for a preset time, when the quality grade of the grain in the granary does not change, a granary that affects the change of the external environment of the granary is obtained as a target granary, and the grain in the target granary is dispatched to be taken out of the target granary or part of the grain in the target granary is transferred to other granaries.

[0021] With reference to the first aspect, in a sixth implementation manner of the first aspect of the present application, obtaining the quality grade of the grain includes:

[0022] Obtaining a grain image captured by the camera from the storage unit, performing denoising and enhancement operations on the grain image, and obtaining a preprocessed grain image, extracting a grain sample from the grain image based on a feature of the grain, the feature of the grain including color, texture, and shape of the grain, identifying a feature parameter related to the grain quality from the grain sample, and obtaining the quality grade of the grain based on the feature parameter related to the grain quality, the feature parameter related to the grain quality including area, perimeter, and aspect ratio of the grain particle.

[0023] In a seventh implementation form of the first aspect, the quality grade of the grain is obtained by:

[0024] The first quality threshold and the second quality threshold are preset, the mass of each grain particle is calculated, the first number of the grain particles in the grain sample whose mass is greater than or equal to the first quality threshold is counted, the second number of the grain particles in the grain sample whose mass is less than the first quality threshold and greater than or equal to the second quality threshold is counted, the third number of the grain particles in the grain sample whose mass is less than the second quality threshold is counted, the first proportion of the first number in the total number of grain particles in the grain sample is calculated, the second proportion of the second number in the total number of grain particles in the grain sample is calculated, the third proportion of the third number in the total number of grain particles in the grain sample is calculated, and the quality grade of the grain corresponding to the grain sample is calculated based on the first proportion, the second proportion, and the third proportion.

[0025] In an eighth implementation form of the first aspect, the mass of each grain particle is calculated by:

[0026] The area, the perimeter, and the aspect ratio of each grain particle are calculated based on the grain sample, and the mass of the grain particle is calculated by the following formula:

[0027] M=S×α1+L×α2+K×α3

[0028] wherein M represents the mass of the grain particle, S represents the area of the grain particle, L represents the perimeter of the grain particle, K represents the aspect ratio of the grain particle, α1 represents a first coefficient, α2 represents a second coefficient, and α3 represents a third coefficient, the first coefficient, the second coefficient, and the third coefficient add up to 1, the first coefficient is greater than the third coefficient, and the third coefficient is greater than the second coefficient.

[0029] In a second aspect, the application provides a grain reserve warehouse inventory intelligent supervision system, which is characterized in that a temperature sensor, a humidity sensor and a camera are respectively installed in each grain warehouse, a gas concentration sensor is installed at a preset height from the bottom of the grain warehouse, and a weight sensor is installed on the bottom of the grain warehouse or a support structure.

[0030] An acquisition unit is configured to acquire historical data of each sensor from the storage unit, pre-process the historical data, and acquire pre-processed historical data, wherein the historical data includes temperature data, humidity data, gas concentration data, grain weight data and grain quality grade;

[0031] A prediction unit is configured to construct a grain quality prediction model, train the grain quality prediction model based on the historical data, acquire the trained grain quality prediction model, collect data of each sensor in the grain warehouse as first data, pre-process the first data, input the first data into the grain quality prediction model, acquire a first prediction result, and based on the first prediction result, an environment adjustment unit adjusts the environment of the grain warehouse and / or a logistics scheduling unit schedules the entry and exit of grain.

[0032] An adjustment unit is configured to collect data of each sensor again as second data based on a first preset period, acquire a second prediction result predicted based on the second data based on the grain quality prediction model, compare the difference between the second prediction result and the first prediction result, and based on the difference, the environment adjustment unit readjusts the environment of the grain warehouse and / or the logistics scheduling unit reschedules the entry and exit of grain.

[0033] Compared with the prior art, the application has at least the following advantages:

[0034] The technical solution provided in this application obtains historical data from each sensor from a storage unit and preprocesses the historical data to obtain the preprocessed historical data. This allows for more accurate capture of changing trends and potential risks in the grain storage environment, providing high-quality input data for subsequent prediction models. A grain quality prediction model is constructed, trained based on historical data, and the trained model is obtained. Data from each sensor in the granary is collected as first data, preprocessed, and input into the grain quality prediction model to obtain a first prediction result. Based on the first prediction result, an environmental adjustment unit adjusts the granary environment and / or a logistics scheduling unit schedules grain inflow and outflow. This achieves real-time monitoring and prediction of grain quality. Furthermore, by automatically adjusting environmental conditions and logistics scheduling, grain spoilage and loss are effectively prevented, thereby improving the security of grain reserves. Data from each sensor is collected again as second data based on a first preset period. A second prediction result based on the second data is obtained based on the grain quality prediction model. The difference between the second prediction result and the first prediction result is compared. Based on the difference, the environmental adjustment unit readjusts the granary environment and / or the logistics scheduling unit readjusts grain inflow and outflow. It is possible to continuously monitor changes in grain quality, adjust environmental conditions in a timely manner, and reduce grain losses. Through the coordination of the above steps, this application realizes intelligent inventory management of grain reserves and improves the safety of grain storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 This is a schematic diagram of an embodiment of an intelligent inventory supervision method for a grain reserve warehouse in an embodiment of the present application;

[0037] Figure 2 This is a schematic diagram of an embodiment of an intelligent inventory monitoring system for a grain reserve warehouse in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The embodiments of the present application provide a grain reserve warehouse inventory intelligent supervision method and system. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0039] It can be understood that the execution subject of the present application can be a grain reserve warehouse inventory intelligent supervision system, and can also be a terminal or a server, which is not limited here. The embodiments of the present application take the server as the execution subject for example.

[0040] For the convenience of understanding, the specific process of the embodiments of the present application is described below. Please refer to Figure 1 One embodiment of the grain reserve warehouse inventory intelligent supervision method in the embodiments of the present application includes the following steps.

[0041] A temperature sensor, a humidity sensor and a camera device are respectively installed inside each grain warehouse, a gas concentration sensor is installed at a preset height from the bottom of the grain warehouse, and a weight sensor is installed on the bottom of the grain warehouse or the support structure.

[0042] Step S1: Obtain historical data of each sensor from the storage unit, and pre-process the historical data to obtain pre-processed historical data, the historical data including temperature data, humidity data, gas concentration data, grain weight data and grain quality grade;

[0043] Specifically, by installing a temperature sensor, a humidity sensor, a camera device and a gas concentration sensor inside each grain warehouse, and installing a weight sensor on the bottom of the grain warehouse or the support structure, the historical data of these sensors can be collected from the storage unit regularly, and the collected data can be pre-processed, including data cleaning (removing outliers and missing values), normalization processing and feature extraction, etc., so as to facilitate model training and prediction. The data pre-processing belongs to the prior art, which will not be described here. By obtaining the historical data, the change trend and potential risks of the grain storage environment can be more accurately captured, and high-quality input data can be provided for the subsequent prediction model.

[0044] Step S2: Constructing a grain quality prediction model, training the grain quality prediction model based on historical data, and obtaining the trained grain quality prediction model. Collect the data of each sensor in the grain depot as the first data, preprocess the first data, and input the first data into the grain quality prediction model to obtain the first prediction result. Based on the first prediction result, the environmental conditioning unit adjusts the environment of the grain depot and / or the logistics scheduling unit schedules the entry and exit of the grain.

[0045] Specifically, a machine learning algorithm such as random forest, support vector machine or neural network is used to construct the grain quality prediction model. The model is trained using preprocessed historical data until the model reaches the expected accuracy. Real-time data collected from sensors in the grain depot are used as the first data, which are preprocessed in the same way and input into the trained grain quality prediction model to obtain the first prediction result. According to the first prediction result, the environmental conditioning unit automatically adjusts the environmental conditions of the grain depot (such as temperature, humidity and gas concentration, etc.), and the logistics scheduling unit adjusts the entry and exit of the grain according to the prediction result to maintain the quality of the grain. Through this step, real-time monitoring and prediction of grain quality are achieved, and through automatic adjustment of environmental conditions and logistics scheduling, deterioration and loss of grain are effectively prevented, and the safety of grain reserves is improved.

[0046] Step S3: Collecting data of each sensor again as second data based on a first preset period, and obtaining second prediction results based on second data based on the grain quality prediction model, comparing the difference between the second prediction results and the first prediction results, and based on the difference, the environmental conditioning unit re-adjusts the environment of the grain depot and / or the logistics scheduling unit re-schedules the entry and exit of the grain.

[0047] Specifically, a preset period (such as every few hours, every day) is set, and at the end of each period, the data of all sensors are collected again as second data. The grain quality prediction model is used to obtain the second prediction result based on the second data. Compare the difference between the second prediction result and the first prediction result to analyze the trend of the change in grain quality. If the difference exceeds a preset threshold, the environmental conditioning unit adjusts the environmental conditions of the grain depot according to the new prediction result and / or the logistics scheduling unit adjusts the entry and exit of the grain to more effectively maintain the quality of the grain. Through periodic data collection and prediction, the change in grain quality can be continuously monitored, and the environmental conditions can be adjusted in a timely manner to reduce grain loss.

[0048] Through the cooperation between the above steps, the application realizes intelligent management of the grain reserves, and improves the safety of grain storage.

[0049] Further, the grain quality prediction model is trained as follows:

[0050] The historical data is divided into a training data set, a validation data set, and a test data set. The training data set is used to train the grain quality prediction model, and the validation data set is used to adjust the parameters of the grain quality prediction model. The performance of the grain quality prediction model is evaluated using the test data set. When the accuracy of the output results of the grain quality prediction model is greater than or equal to the preset threshold, the grain quality prediction model is obtained.

[0051] Specifically, the preprocessed historical data is randomly divided into three parts: training data set, validation data set and test data set. The common division ratio is 70% training data set, 15% validation data set and 15% test data set, but the specific ratio can be adjusted according to the actual situation. The training data set is used to train the model and learn the patterns and relationships in the data. The validation data set is used for model parameter adjustment, that is, during the training process, it is used to adjust the hyperparameters of the model to prevent overfitting or underfitting of the model. The test data set is used to evaluate the performance of the model, which provides the performance of the model on unseen data. By dividing the data set, it can be ensured that the model will not overfit the training data during the training process, and at the same time, the generalization ability of the model can be evaluated through independent validation and test data sets. Select the appropriate algorithm (such as random forest, support vector machine, neural network, etc.) and set the initial parameters, and use the training data set to train the grain quality prediction model. Apply optimization algorithms such as gradient descent to minimize prediction errors, and train the model through multiple iterations. The training process enables the model to learn the relationship between grain quality and sensor data, enabling it to predict grain quality. Adjust the parameters of the model using the validation data set, including adjusting the learning rate, the depth of the tree (for decision tree models), the regularization parameter, etc. Through techniques such as cross-validation, evaluate the performance of the model under different parameter settings and select the best parameter combination. Improve the accuracy and robustness of the model to ensure the predictive performance of the model on new data. Use the test data set to evaluate the performance of the model to ensure the reliability of the model's prediction results. Calculate the accuracy of the model's output results and other indicators. If the accuracy of the model is greater than or equal to the preset threshold, the model performance is considered to meet the requirements and can be used for actual grain quality prediction. Through the implementation of this step, the grain quality prediction model can adapt to new data and environmental changes while ensuring high accuracy, providing strong technical support for intelligent inventory management of grain reserves.

[0052] Further, adjusting the environment of the grain warehouse includes:

[0053] When the first prediction result decreases compared to the latest quality grade of the grain, compare the temperature data, humidity data, gas concentration data and grain weight data respectively to obtain the factors causing the decrease in the first prediction result. When the decrease factor is temperature data, open the refrigeration equipment close to the grain depot. When the decrease factor is humidity data, open the dehumidification equipment and ventilation equipment close to the grain depot. When the decrease factor is gas concentration data, use the controlled atmosphere storage technology to reduce the oxygen concentration in the grain depot. When the decrease factor is grain weight data, dispatch the grain in the grain depot for delivery or transfer part of the grain in the grain depot to other grain depots.

[0054] Specifically, when the output of the grain quality prediction model shows that the quality grade of the grain decreases compared to the latest time, an analysis process is automatically triggered. The latest temperature data, humidity data, gas concentration data and grain weight data are compared with historical baseline or preset threshold respectively to determine the potential factors causing the quality decrease. If the analysis result shows that the temperature data is the factor causing the quality decrease, the refrigeration equipment close to the grain depot is automatically opened to reduce the temperature in the grain depot. The opening of the refrigeration equipment will be carried out according to the preset temperature control strategy until the temperature returns to the range suitable for grain storage. If the humidity data is the factor causing the quality decrease, the dehumidification equipment and ventilation equipment are started to reduce the humidity in the grain depot. The operation of the dehumidification and ventilation equipment will be adjusted according to the real-time humidity data to ensure that the humidity is controlled within the optimal range. When the gas concentration data is identified as the factor causing the quality decrease, the controlled atmosphere storage technology is adopted to reduce the oxygen concentration in the grain depot by filling nitrogen or carbon dioxide and the like, so as to inhibit the grain respiration and microbial activity and delay the grain aging. The adjustment of the gas concentration will be carried out according to the real-time monitoring data to ensure that the predetermined gas ratio is achieved. If the change of the grain weight data causes the quality grade to decrease, it may be due to excessive or insufficient storage of the grain, and the grain delivery will be automatically dispatched or part of the grain will be transferred to other grain depots to keep the grain storage amount in each grain depot within the appropriate range. The grain dispatching will be carried out according to the inventory management strategy and the instruction of the logistics scheduling unit to ensure the efficiency and cost-effectiveness of the grain flow. Through the implementation of the adjustment measures, the monitoring and response of the grain depot environment are more accurate, which helps to discover and solve the problems that may cause the quality of the grain to decrease in time. At the same time, through the accurate control of the temperature, humidity and gas concentration, the best storage environment for the grain can be provided to prevent the grain from being mildewed, infested and deteriorated and to ensure the safety of the grain. Through the implementation of the present step, the inventory intelligent supervision method of the grain reserve depot not only improves the quality and efficiency of the grain storage, but also enhances the adaptability to environmental changes and the prevention ability to potential risks.

[0055] Further, adjusting the environment of the grain depot further includes:

[0056] When multiple factors are combined to affect the grain quality, the corresponding equipment is adjusted based on the combination of factors, and the data of each sensor is collected based on a second preset period, which is less than the first prediction period, a new prediction result is obtained based on the grain quality prediction model, and the environment of the grain depot is continuously adjusted based on the new prediction result.

[0057] Specifically, generally, there are multiple factors affecting the decline of grain quality, that is, when the output of the grain quality prediction model shows that the grain quality level has decreased compared with the last time, and it is determined through comparative analysis that multiple factors (such as temperature, humidity and gas concentration) jointly cause the quality decline, the multi-factor comprehensive adjustment strategy will be started. If temperature and humidity are both influencing factors, the refrigeration equipment and dehumidification equipment will be started at the same time, and the operation of the ventilation equipment may be increased to quickly adjust the temperature and humidity environment in the grain depot. When increasing the operation of the ventilation equipment, the natural ventilation condition is preferred, which not only ensures the grain quality, but also saves resources. If the gas concentration is also an influencing factor, the gas regulation storage technology will be combined to adjust the type and proportion of the filled gas to reduce the oxygen concentration and optimize the storage environment. If the grain weight data shows that it is too high, the grain will be dispatched out of the warehouse or allocated among the grain depots to maintain the appropriate storage amount. By shortening the data collection period, the system can respond more quickly to environmental changes and adjust the equipment in time to reduce the risk of grain quality decline. Comprehensive adjustment of multiple environmental factors makes the grain depot management more refined, improves the stability of the grain storage environment, reduces the grain loss caused by unsuitable environment, and improves the storage quality of the grain.

[0058] Further, the dispatching of grain in and out includes:

[0059] When the first prediction result is higher or the same as the last grain quality level, the same type of grain as that in the grain depot is unloaded into the warehouse, otherwise it is unloaded into other grain depots or newly built grain depots.

[0060] Specifically, when the output of the grain quality prediction model shows that the grain quality level has increased or remained the same compared to the last time, the grain scheduling process is automatically triggered. The quality level of the current grain in the warehouse is compared with the quality level of the grain about to enter the warehouse. If the grain about to enter the warehouse is of the same type as the grain stored in the warehouse, the logistics scheduling unit is notified to unload these grains into the warehouse. If they are not of the same type, the system will instruct the logistics scheduling unit to unload these grains into other warehouses or newly built warehouses. The logistics scheduling unit will dispatch the corresponding transportation equipment (such as conveyors, forklifts, automated guided vehicles, etc.) to transport the grain to the designated warehouse according to the system instructions. At the same time, the scheduling unit will consider the current storage capacity of the warehouse, the shelf life of the grain, and the flow rate of the grain, etc. to optimize the storage space and improve the flow efficiency. Storing grains with similar quality levels together can reduce cross-contamination between different quality grains and maintain the overall quality of the grain. Reasonable grain scheduling can reduce grain loss caused by improper storage, such as avoiding storing perishable grains in unsuitable environments.

[0061] Further, scheduling the entry and exit of grain also includes:

[0062] After the preset time of environmental adjustment of the warehouse, if the quality level of the grain in the warehouse does not change, the warehouse affected by changes in the external environment is obtained as the target warehouse, and the grain in the target warehouse is scheduled for export or part of the grain in the target warehouse is transferred to other warehouses.

[0063] Specifically, after the preset time of environmental adjustment of the warehouse, the quality level of the grain in the warehouse is re-evaluated. If the evaluation result shows that the quality level of the grain does not change, it indicates that the previous environmental adjustment measures have not effectively improved the grain storage conditions. Analyze and identify factors affecting changes in the external environment of the warehouse, such as climate change, geographical location, etc. Based on the influencing factors, the affected warehouse is obtained as the target warehouse, and further grain scheduling measures are taken. For the target warehouse, specific grain scheduling strategies are developed, including grain export and redistribution. The scheduling strategy will consider the quality of the grain in the target warehouse, the storage capacity, the type of grain, and market demand, etc. According to the scheduling strategy, the logistics scheduling unit is instructed to export part or all of the grain in the target warehouse. For grain that does not need to be exported, these grains are transferred to other warehouses with more suitable conditions to optimize the overall grain storage layout. By optimizing the storage and scheduling of grain, the risk of grain during storage can be reduced, and the safety of grain can be improved. Through the implementation of this step, the impact of environmental changes on grain storage can be more effectively addressed, ensuring the quality of grain and improving the efficiency and safety of grain reserves.

[0064] Further, obtaining the quality level of the grain includes:

[0065] The image of the grain taken by the camera is obtained from the storage unit, the grain image is denoised and enhanced, and the preprocessed grain image is obtained, the grain sample is extracted from the grain image based on the characteristics of the grain, the characteristics of the grain include the color, texture and shape of the grain, the feature parameters related to the quality of the grain are identified from the grain sample, and the quality grade of the grain is obtained based on the feature parameters related to the quality of the grain, the feature parameters related to the quality of the grain include the area, perimeter and aspect ratio of the grain particles.

[0066] Specifically, the image of the grain is periodically taken by the camera installed in the grain depot for quality grade evaluation. The collected grain image is denoised to eliminate random noise in the image and improve image quality. The image is enhanced, such as adjusting contrast, brightness and sharpness, so that the characteristics of the grain are more obvious. The image segmentation technology is used to extract the grain sample from the preprocessed grain image to distinguish the grain and the background. For example, the color, texture and shape features of the grain are extracted, the color feature can be extracted by color histogram or color moment method; the texture feature can be extracted by gray level co-occurrence matrix, local binary pattern method; the shape feature can be extracted by contour tracking, shape descriptor method. The feature parameters related to the quality of the grain, such as the area, perimeter and aspect ratio of the grain particles, are identified from the grain sample, which can be automatically measured by image analysis software, such as using the functions in the image processing library (for example, OpenCV), which belongs to the prior art and will not be repeated here. Based on the identified quality-related feature parameters, combined with the preset quality evaluation standard or machine learning model, the quality grade of the grain is evaluated. If a machine learning model is used, the model may have been trained by historical data to identify the feature parameter distribution of grain of different quality grades. The output quality grade evaluation result of the grain can be used to trigger the above-mentioned grain depot environment regulation or grain scheduling operation. Through this step, the accuracy of quality evaluation is improved; at the same time, real-time grain quality monitoring can also be provided, the change of grain quality can be found in time, the risk of grain deterioration can be reduced, and the safety of grain can be improved.

[0067] Further, obtaining the quality grade of the grain further comprises:

[0068] The first quality threshold and the second quality threshold are preset, the mass of each grain particle is calculated, the first number of grain particles with mass greater than or equal to the first quality threshold in the grain sample is counted, the second number of grain particles with mass less than the first quality threshold and greater than or equal to the second quality threshold in the grain sample is counted, the third number of grain particles with mass less than the second quality threshold is counted, the first proportion of the first number in the total number of grain particles in the grain sample is calculated, the second proportion of the second number in the total number of grain particles in the grain sample is calculated, the third proportion of the third number in the total number of grain particles in the grain sample is calculated, and the quality grade of the grain corresponding to the grain sample is calculated based on the first proportion, the second proportion and the third proportion.

[0069] Specifically, the first quality threshold and the second quality threshold are preset to distinguish the quality grade of the grain particles, the size information of the grain particles is extracted from the image, and the mass of each grain particle is calculated using image analysis technology. The mass distribution of the grain particles in the grain sample is counted: the first number is the number of grain particles with mass greater than or equal to the first quality threshold. The second number is the number of grain particles with mass less than the first quality threshold but greater than or equal to the second quality threshold. The third number is the number of grain particles with mass less than the second quality threshold. The proportions of the first number, the second number and the third number in the total number of grain particles in the grain sample are calculated: the first proportion is the proportion of the first number in the total number. The second proportion is the proportion of the second number in the total number. The third proportion is the proportion of the third number in the total number. Based on the first proportion, the second proportion and the third proportion, the quality grade of the grain corresponding to the grain sample is calculated to more accurately evaluate the quality of the grain, thereby triggering the above-mentioned grain storehouse environment regulation or grain scheduling operation. Through the implementation of this step, the quality of the grain can be more effectively managed and monitored to ensure the safety of the grain while improving management efficiency and economic benefits.

[0070] Further, the calculation of the mass of each grain particle includes:

[0071] Based on the grain sample, the area, the perimeter and the aspect ratio of each grain particle are calculated, and the mass of the grain particle is calculated by the following formula:

[0072] M = S × a1 + L × a2 + K × a3

[0073] wherein M represents the mass of the grain particle, S represents the area of the grain particle, L represents the perimeter of the grain particle, K represents the aspect ratio of the grain particle, a1 represents the first coefficient, a2 represents the second coefficient, and a3 represents the third coefficient, the sum of the first coefficient, the second coefficient and the third coefficient is equal to 1, the first coefficient is greater than the third coefficient, and the third coefficient is greater than the second coefficient.

[0074] Specifically, by implementing the step, the area, the perimeter and the aspect ratio of the grain particles are comprehensively considered, the quality of each grain can be more accurately evaluated, the change of the quality of the grain can be found in time, corresponding measures can be taken, and the safety of the grain can be improved.

[0075] The above describes a method for intelligently monitoring the inventory of a grain reserve warehouse according to an embodiment of the application, and the following describes a system for intelligently monitoring the inventory of a grain reserve warehouse according to an embodiment of the application. Please refer to Figure 2 An embodiment of the system for intelligently monitoring the inventory of a grain reserve warehouse according to an embodiment of the application includes:

[0076] A temperature sensor, a humidity sensor and a camera are respectively installed in each granary, a gas concentration sensor is installed at a preset height from the bottom of the granary, and a weight sensor is installed on the bottom of the granary or the support structure.

[0077] The acquisition unit 201 is configured to acquire historical data of each sensor from the storage unit, pre-process the historical data, and acquire pre-processed historical data. The historical data includes temperature data, humidity data, gas concentration data, grain weight data and quality grade of the grain.

[0078] The prediction unit 202 is configured to construct a grain quality prediction model, train the grain quality prediction model based on the historical data, acquire the trained grain quality prediction model, collect data of each sensor in the granary as first data, pre-process the first data, input the first data into the grain quality prediction model, acquire a first prediction result, and based on the first prediction result, the environment adjustment unit adjusts the environment of the granary and / or the logistics scheduling unit schedules the entry and exit of the grain.

[0079] The adjustment unit 203 is configured to collect data of each sensor again as second data based on a first preset period, acquire a second prediction result predicted based on the second data based on the grain quality prediction model, compare the difference between the second prediction result and the first prediction result, and based on the difference, the environment adjustment unit re-adjusts the environment of the granary and / or the logistics scheduling unit re-schedules the entry and exit of the grain.

[0080] Through the cooperation of the above-mentioned components, the application realizes intelligent monitoring and management of the inventory of the grain reserve warehouse, and improves the safety of grain storage.

[0081] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, system and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0082] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0083] The above description and the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for intelligently monitoring the inventory of a grain reserve warehouse, wherein a temperature sensor, a humidity sensor and a camera device are respectively installed in each grain bin, a gas concentration sensor is installed at a preset height from the bottom of the grain bin, and a weight sensor is installed on the bottom of the grain bin or the support structure, characterized in that, The grain reserve warehouse inventory intelligent management method comprises: Step S1: obtain historical data of each sensor from the storage unit, preprocess the historical data, and obtain preprocessed historical data, the historical data including temperature data, humidity data, gas concentration data, grain weight data, and grain quality grade; Step S2: build a grain quality prediction model, train the grain quality prediction model based on the historical data, and obtain the trained grain quality prediction model, collect data of each sensor in the granary as first data, preprocess the first data, and input the first data into the grain quality prediction model to obtain a first prediction result, based on the first prediction result, the environment adjusting unit adjusts the environment of the granary and / or the logistics scheduling unit schedules the entry and exit of grain; Step S3: based on a first preset period, collect data of each sensor again as second data, and based on the grain quality prediction model, obtain a second prediction result predicted based on the second data, compare the difference between the second prediction result and the first prediction result, and based on the difference, the environment adjusting unit re-adjusts the environment of the granary and / or the logistics scheduling unit re-schedules the entry and exit of grain.

2. The method for intelligent monitoring of the inventory of a grain storage depot according to claim 1, characterized in that, Training the grain quality prediction model: Divide the historical data into a training data set, a validation data set, and a test data set, use the training data set to train the grain quality prediction model, use the validation data set to adjust the parameters of the grain quality prediction model, use the test data set to evaluate the performance of the grain quality prediction model, and obtain the grain quality prediction model when the accuracy rate of the output result of the grain quality prediction model is greater than or equal to a preset threshold.

3. The method of claim 1, wherein the method further comprises: Adjusting the environment of the granary includes: When the first prediction result is lower than the quality grade of the grain in the last time, compare the temperature data, humidity data, gas concentration data, and grain weight data respectively, obtain the factors causing the decrease of the first prediction result, open the refrigeration equipment close to the granary when the temperature data is the decreasing factor, open the dehumidification equipment and ventilation equipment close to the granary when the humidity data is the decreasing factor, use the controlled atmosphere storage technology when the gas concentration data is the decreasing factor, and schedule the grain in the granary for export or transfer part of the grain in the granary to other granaries when the grain weight data is the decreasing factor.

4. The method of claim 3, wherein the method further comprises: Adjusting the environment of the granary also includes: When there are multiple factors combined, adjust the corresponding equipment based on the combination of the factors, collect data of each sensor based on a second preset period, obtain a new prediction result based on the grain quality prediction model, and continue to adjust the environment of the granary based on the new prediction result, wherein the second preset period is less than the first preset period.

5. The method of claim 1, wherein, Scheduling the entry and exit of grain includes: If the first prediction result is that the quality grade of the grain is higher than or equal to the last time, the same type of grain entering the warehouse is unloaded into the grain depot, otherwise, the grain is unloaded into other grain depots or newly built grain depots.

6. The method of claim 5, wherein the method further comprises: The scheduling of the grain in and out further includes: After the preset time of the environment adjustment of the grain depot, if the quality grade of the grain in the grain depot does not change, a grain depot that is affected by the change of the external environment is obtained as a target grain depot, and the grain in the target grain depot is scheduled to be unloaded, or part of the grain in the target grain depot is transferred to other grain depots.

7. The method of claim 1, wherein the method further comprises: The obtaining of the quality grade of the grain includes: The grain image captured by the camera is obtained from the storage unit, the grain image is denoised and enhanced, and the preprocessed grain image is obtained, the grain sample is extracted from the grain image based on the characteristics of the grain, the characteristics of the grain include the color, texture and shape of the grain, the characteristic parameters related to the quality of the grain are identified from the grain sample, and the quality grade of the grain is obtained based on the characteristic parameters related to the quality of the grain, the characteristic parameters related to the quality of the grain include the area, perimeter and aspect ratio of the grain particles.

8. The method of claim 7, wherein the method further comprises: The obtaining of the quality grade of the grain further includes: The first quality threshold and the second quality threshold are preset, the mass of each grain particle is calculated, the first number of the grain particles with a mass greater than or equal to the first quality threshold in the grain sample is counted, the second number of the grain particles with a mass less than the first quality threshold and greater than or equal to the second quality threshold in the grain sample is counted, the third number of the grain particles with a mass less than the second quality threshold in the grain sample is counted, the first proportion of the first number in the total number of grain particles in the grain sample is calculated, the second proportion of the second number in the total number of grain particles in the grain sample is calculated, the third proportion of the third number in the total number of grain particles in the grain sample is calculated, and the quality grade of the grain corresponding to the grain sample is calculated based on the first proportion, the second proportion and the third proportion.

9. The method of claim 8, wherein the method further comprises: The calculation of the mass of each grain particle includes: Based on the grain sample, the area, perimeter and aspect ratio of each grain particle are calculated, and the mass of the grain particle is calculated by the following formula: M=S×α1+L×α2+K×α3 Wherein, M represents the mass of the grain particle, S represents the area of the grain particle, L represents the perimeter of the grain particle, K represents the aspect ratio of the grain particle, α1 represents the first coefficient, α2 represents the second coefficient, and α3 represents the third coefficient, the sum of the first coefficient, the second coefficient and the third coefficient is equal to 1, the first coefficient is greater than the third coefficient, and the third coefficient is greater than the second coefficient.

10. A system for intelligent monitoring of the inventory of a grain storage depot, wherein a temperature sensor, a humidity sensor and a camera are installed in each grain bin, a gas concentration sensor is installed at a predetermined height from the bottom of the grain bin, and a weight sensor is installed on the bottom of the grain bin or the support structure, for implementing the method for intelligent monitoring of the inventory of a grain storage depot according to any one of claims 1-9. The inventory intelligent monitoring system of the grain reserve warehouse includes: The acquisition unit is configured to acquire historical data of each sensor from the storage unit, pre-process the historical data, and acquire pre-processed historical data, wherein the historical data comprises temperature data, humidity data, gas concentration data, grain weight data, and grain quality grade; The prediction unit is configured to construct a grain quality prediction model, train the grain quality prediction model based on the historical data, acquire the trained grain quality prediction model, collect data of each sensor in the granary as first data, pre-process the first data, input the first data into the grain quality prediction model, acquire a first prediction result, and based on the first prediction result, the environment adjustment unit adjusts the environment of the granary and / or the logistics scheduling unit schedules the entry and exit of the grain. The adjustment unit is configured to collect data of each sensor again as second data based on a first preset period, acquire a second prediction result predicted based on the second data based on the grain quality prediction model, compare a difference between the second prediction result and the first prediction result, and based on the difference, the environment adjustment unit re-adjusts the environment of the granary and / or the logistics scheduling unit re-schedules the entry and exit of the grain.

Images