Grain storage intelligent early warning system based on big data driving

Abstract

The invention discloses a grain storage intelligent early-warning system based on big data driving, and relates to the technical field of granary storage, and the grain storage early-warning system comprises an initial data module, a storage model, an early-warning module, an early-warning module and an early-warning module, a virtual three-dimensional model is constructed through storage data, grain types are obtained through image recognition, and the virtual three-dimensional model is perfected according to grain quality and quantity. The system has the advantages that temperature, humidity, gas components and insect attack activity data of storage grain piles are collected in real time through a sensor monitoring module and transmitted to a data processing module; the data processing module dynamically generates a virtual monitoring point position in a virtual model based on a three-dimensional grid division result, and the virtual monitoring point position and the sensor monitoring module are matched with each other to monitor various data in the grain pile, so that the monitoring range of the stored grain pile is expanded, and the data accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of grain storage technology, and in particular to a grain storage intelligent early warning system driven by big data. Background Art

[0002] Since the 21st century, global climate change has intensified, with more frequent extreme weather events. Grain storage faces increasingly complex environmental fluctuation risks. At the same time, consumers are increasingly demanding higher quality grains, and niche markets such as organic and functional grains are imposing increasingly stringent standards for storage conditions. Against this backdrop, a big data-driven intelligent early warning system has emerged. This system relies on IoT sensors to collect real-time data on grain pile temperature, humidity, gas composition, and pest activity. It combines 3D modeling technology to construct a virtual granary, dynamically simulating grain storage conditions. Machine learning algorithms analyze historical data to predict risks such as mold and pests, and automatically adjust the environment using ventilation and cooling equipment. For example, the system can accurately locate local hotspots based on temperature and humidity gradients, providing early warnings and avoiding the lag of traditional "after-the-fact" remediation methods.

[0003] The existing storage monitoring and early warning system is unable to predict the internal data of the grain pile in a targeted manner, which affects the data accuracy and monitoring range of the grain storage intelligent early warning system. Therefore, we propose a grain storage intelligent early warning system driven by big data. Summary of the Invention

[0004] The purpose of the present invention is to provide a grain storage intelligent early warning system driven by big data.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a grain storage intelligent early warning system driven by big data, the grain storage early warning system comprising:

[0006] The initial data module is used to obtain the origin, type, quality and storage data of stored grain through IoT devices, user input or database integration;

[0007] The warehouse model uses the data from the initial data module to build a virtual 3D model using 3D modeling software, and then improves the virtual 3D model by taking into account the type, quality and quantity of grain;

[0008] A sensor monitoring module is used to install sensors at monitoring points to monitor grain in storage and transmit the monitored grain data to the virtual three-dimensional model;

[0009] The data processing module divides the virtual model into three-dimensional grids, establishes virtual monitoring points according to the spatial coordinates of the grid nodes, and obtains the data of the virtual monitoring points through the formula. The specific formula is as follows:

[0010]

[0011] Among them, Z a is the monitoring data of the virtual monitoring point, Y i are the data of different monitoring points in the i-th sensor monitoring module, f i are the weight coefficients of different monitoring points in the i-th sensor monitoring module, and the weight coefficients of different monitoring points are adjusted according to the distance from the virtual monitoring point, and a distance monitoring distance threshold W is set. When the distance between different monitoring points and the virtual monitoring point is greater than the distance threshold W, the corresponding weight coefficient is set to 0;

[0012] Intelligent early warning module, used to issue early warnings for abnormal data in the warehouse.

[0013] As a further solution of the present invention: the sensor monitoring module is provided with a temperature sensor, a humidity sensor, an air density sensor and a wind direction sensor, and the environmental data in the warehouse model is obtained through different sensors in the sensor monitoring module, and the environmental data is imported into the virtual three-dimensional model. At the same time, the sensor monitoring module is also provided with an adjustment unit, and the adjustment unit is used to control the ventilation equipment, humidification equipment, dehumidification equipment, cooling equipment and warming equipment in the warehouse, and the grain storage environment in the warehouse is adjusted through the adjustment unit.

[0014] As a further solution of the present invention, the data processing module verifies the deviation between the predicted data of the virtual monitoring point and the actual measurement data through the root mean square error formula. The specific formula is as follows:

[0015]

[0016] Among them, Z b is the deviation between the virtual monitoring point and the actual data, is the data value predicted for the i-th virtual monitoring point, is the actual data periodically measured at the i-th virtual monitoring point.

[0017] As a further solution of the present invention: a monitoring unit is set in the data processing module, and a monitoring threshold X is set in the monitoring unit. At the same time, the monitoring threshold of the virtual monitoring point is X-|Z b |, when one of the temperature, humidity and pest density of the sensor monitoring point exceeds the preset monitoring threshold X and the data of the virtual monitoring point is greater than X-|Z b |When the data processing module controls the regulating unit to process the grain in the warehouse.

[0018] As a further solution of the present invention: a simulation unit is set in the data processing module, the virtual three-dimensional model data is imported into the simulation unit, the sensor in the sensor monitoring module is used to obtain the environmental data of the warehouse, and then the working state parameters of the adjustment unit are simulated in the simulation unit according to the environmental data of the warehouse.

[0019] As a further solution of the present invention, the simulation unit simulates the state data in the warehouse through a formula, and the specific formula is as follows:

[0020] M=α·K+β·P+γ·A

[0021] Among them, M is the simulation data of the simulation unit, K is the pest development prediction data, P is the storage environment prediction data, A is the mildew risk prediction data, α, β and γ are the weight coefficients of pest development prediction, storage environment prediction and mildew risk prediction respectively.

[0022] As a further solution of the present invention: the storage environment prediction data is generated by the grain pile pores and the moisture and heat gradient. The storage environment prediction data is Among them, δ and ε are the weight coefficients of the linear state and nonlinear state in the pores of the grain pile, respectively; L is the air dynamic viscosity data inside the grain pile, which is obtained by the air velocity sensor; S is the permeability data inside the grain pile, which is obtained by the porosity detector; v is the ventilation efficiency data inside the grain pile, which is measured by the wind speed sensor; G is the inertia data of the air flow inside the grain pile; and D is the air density data in the grain.

[0023] As a further solution of the present invention: the intelligent early warning module issues an early warning through the instruction data of the simulation unit. When the simulation unit calculates that the working efficiency of the adjustment unit is lower than the preset threshold, it is determined that it cannot be adjusted. The intelligent early warning module then broadcasts the abnormal data of the grain pile and generates a corresponding processing strategy.

[0024] By adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are:

[0025] 1. The present invention uses a sensor monitoring module to collect real-time data on the temperature, humidity, gas composition, and pest activity of stored grain piles, and transmits it to a data processing module. The data processing module dynamically generates virtual monitoring points in a virtual model based on the three-dimensional grid division results. The virtual monitoring points and the sensor monitoring module cooperate with each other to monitor various data in the grain pile, thereby improving the monitoring range of the stored grain pile and the data accuracy.

[0026] 2. The present invention periodically collects virtual point data to ensure the deviation value of the virtual point data, and then adjusts the range of the monitoring threshold, so that the storage system can quickly adjust the internal grain pile and issue early warnings;

[0027] 3. The present invention simulates the state of the grain pile in storage through the simulation unit, and then when the grain pile is processed, the grain pile state data can be analyzed through the sensor monitoring module and the virtual monitoring point, and then it can be determined whether the grain pile can be adjusted through the adjustment unit so that the temperature, humidity, etc. in the grain pile return to normal. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Schematic diagram of the intelligent warehouse early warning system in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0030] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0031] Please see the attached Figure 1 The present invention provides a grain storage intelligent early warning system based on big data drive, and the grain storage early warning system includes:

[0032] The initial data module is used to obtain the origin, type, quality and storage data of stored grain through IoT devices, user input or database integration;

[0033] The warehouse model uses the data from the initial data module to build a virtual 3D model using 3D modeling software, and then improves the virtual 3D model by taking into account the type, quality and quantity of grain;

[0034] A sensor monitoring module is used to install sensors at monitoring points to monitor grain in storage and transmit the monitored grain data to the virtual three-dimensional model;

[0035] The data processing module divides the virtual model into three-dimensional grids, establishes virtual monitoring points according to the spatial coordinates of the grid nodes, and obtains the data of the virtual monitoring points through the formula. The specific formula is as follows:

[0036]

[0037] Among them, Z a is the monitoring data of the virtual monitoring point, Y iare the data of different monitoring points in the i-th sensor monitoring module, f i are the weight coefficients of different monitoring points in the i-th sensor monitoring module, and the weight coefficients of different monitoring points are adjusted according to the distance from the virtual monitoring point, and the distance monitoring distance threshold W is set. When the distance between different monitoring points and the virtual monitoring point is greater than the distance threshold W, the corresponding weight coefficient is set to 0; that is, f i >W, then f i =0.

[0038] Intelligent early warning module, used to issue early warnings for abnormal data in the warehouse.

[0039] In one embodiment of the present invention: a temperature sensor, a humidity sensor, an air density sensor and a wind direction sensor are provided in the sensor monitoring module, and environmental data in the storage model are obtained through different sensors in the sensor monitoring module, and the environmental data are imported into the virtual three-dimensional model. At the same time, an adjustment unit is also provided in the sensor monitoring module, and the adjustment unit is used to control the ventilation equipment, humidification equipment, dehumidification equipment, cooling equipment and warming equipment in the warehouse, and the grain storage environment in the warehouse is adjusted through the adjustment unit.

[0040] In one embodiment of the present invention, the data processing module verifies the deviation between the predicted data of the virtual monitoring point and the actual measured data using the root mean square error formula. The specific formula is as follows:

[0041]

[0042] Among them, Z b is the deviation between the virtual monitoring point and the actual data, is the data value predicted for the i-th virtual monitoring point, is the actual data periodically measured at the i-th virtual monitoring point.

[0043] In one embodiment of the present invention, a monitoring unit is set in the data processing module, a monitoring threshold X is set in the monitoring unit, and the monitoring threshold of the virtual monitoring point is X-|Z b |, when one of the temperature, humidity and pest density of the sensor monitoring point exceeds the preset monitoring threshold X and the data of the virtual monitoring point is greater than X-|Z b |, the data processing module controls the regulating unit to process the grain in the storage;

[0044] Obtain the moisture and heat gradient data of the stored grain pile that changes over time, and import the changing moisture and heat gradient data into the virtual monitoring point X-|Z b |, and then dynamically adjust the monitoring range of the virtual monitoring point.

[0045] In one embodiment of the present invention: a simulation unit is set in the data processing module, the virtual three-dimensional model data is imported into the simulation unit, the sensor in the sensor monitoring module is used to obtain the environmental data of the warehouse, and then the working state parameters of the adjustment unit are simulated in the simulation unit according to the environmental data of the warehouse.

[0046] In one embodiment of the present invention, the simulation unit simulates the state data in the warehouse through a formula, and the specific formula is as follows:

[0047] M=α·K+β·P+γ·A

[0048] Where M is the simulation data of the simulation unit, K is the pest development prediction data, P is the storage environment prediction data, A is the mildew risk prediction data, α, β and γ are the weight coefficients of pest development prediction, storage environment prediction and mildew risk prediction respectively;

[0049] The weight coefficients of pest development prediction, storage environment prediction, and mildew risk prediction are 0.3, 0.5, and 0.2, respectively, and the simulation unit is simulated using a CFD simulation system.

[0050] In one embodiment of the present invention, the storage environment prediction data is generated by the grain pile porosity and the moisture and heat gradient. Wherein, δ and ε are the weight coefficients of the linear state and nonlinear state in the grain pile pores, respectively; L is the air dynamic viscosity data inside the grain pile, obtained by the air velocity sensor; S is the permeability data inside the grain pile, obtained by the porosity detector; v is the ventilation efficiency data inside the grain pile, measured by the wind speed sensor; G is the inertia data of the air flow inside the grain pile; and D is the air density data in the grain.

[0051] The ventilation efficiency in the pores of the grain pile is obtained through linear and nonlinear methods, and then the cooling data and mildew inhibition data in the grain pile are determined based on the ventilation efficiency. The weight coefficients of the linear and nonlinear methods in the pores of the grain pile are 0.9 and 0.1, respectively.

[0052] In one embodiment of the present invention: the intelligent early warning module issues an early warning through the instruction data of the simulation unit. When the simulation unit calculates that the working efficiency of the adjustment unit is lower than the preset threshold, it is determined that it cannot be adjusted. Then the intelligent early warning module broadcasts the abnormal data of the grain pile and generates a corresponding processing strategy.

[0053] Example 1: Simulating a scene where the grain pile in the warehouse is wheat

[0054] When the grain stored in the warehouse is wheat, the sensor monitoring module is used to obtain the wheat pile density, grain specific heat capacity, grain pile temperature, grain pile effective thermal conductivity, moisture phase change latent heat, grain moisture content, moisture diffusion coefficient, etc., so that the simulation unit can perform high-precision simulation based on multi-dimensional data, thereby improving the accuracy of virtual monitoring points.

[0055] Example 2: Warehousing Management Scenario for Small and Medium-sized Farmers

[0056] For the decentralized storage facilities of small and medium-sized farmers or cooperatives, the system can use low-cost sensors and lightweight models to monitor key indicators such as temperature and humidity of grain piles, pest activity, etc. in real time, and push early warning information through mobile terminals to help farmers take timely measures such as ventilation, drying or fumigation to reduce food losses.

[0057] Example 3: Dynamic Monitoring Scenario for Emergency Rescue Granaries

[0058] In areas prone to disasters such as earthquakes and floods, the system is deployed in temporary disaster relief granaries. It monitors the status of grain piles in real time through wireless sensor networks, and combines meteorological data to predict environmental changes (such as a sudden increase in humidity caused by heavy rain) to trigger emergency ventilation or transfer instructions to ensure disaster relief food security.

[0059] Specifically, the sensor monitoring module collects data on temperature, humidity, gas composition and pest activity of the stored grain pile in real time and transmits it to the data processing module. Based on the three-dimensional grid division results, the data processing module dynamically generates virtual monitoring points in the virtual model, and uses the cooperation between the virtual monitoring points and the sensor monitoring module to monitor various data in the grain pile, thereby improving the monitoring range of the stored grain pile and the data accuracy.

[0060] Specifically, by periodically collecting virtual point data, the deviation value of the virtual point data is ensured, and the range of the monitoring threshold is adjusted, so that the storage system can quickly adjust the internal grain pile and issue early warnings.

[0061] Specifically, the state of the grain pile in the storage is simulated by the simulation unit, and then when the grain pile is processed, the grain pile state data can be analyzed through the sensor monitoring module and the virtual monitoring point to determine whether the grain pile can be adjusted through the adjustment unit so that the temperature, humidity, etc. in the grain pile return to normal.

[0062] Working principle:

[0063] First, the basic data in the warehouse is obtained through the initial data module, and then the basic virtual three-dimensional model is established using the warehouse model. The grain pile in the warehouse is monitored through the sensor monitoring module, and the grain pile is planned using the three-dimensional grid. Virtual monitoring points are established in the planned three-dimensional grid. By verifying the accuracy of the virtual monitoring points, the monitoring thresholds of the sensor monitoring module and the virtual monitoring points are generated. Then, within the corresponding threshold range, the temperature, humidity and other conditions of the grain inside the grain pile are controlled through the adjustment unit. When the simulation unit simulates that the grain pile cannot be adjusted through the adjustment unit, an early warning is issued through the intelligent early warning module. At this point, the entire workflow ends.

[0064] Although the present invention is disclosed above with reference to preferred embodiments, this is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present invention. Therefore, any modifications, equivalent variations, and modifications made to the above embodiments in accordance with the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the scope of protection defined by the claims of the present invention.

Claims

1. A grain storage intelligent early warning system driven by big data, characterized by: The grain storage early warning system includes: The initial data module is used to obtain the origin, type, quality and storage data of stored grain through IoT devices, user input or database integration; The warehouse model uses the data from the initial data module to build a virtual 3D model using 3D modeling software, and then improves the virtual 3D model by taking into account the type, quality and quantity of grain; A sensor monitoring module is used to install sensors at monitoring points to monitor grain in storage and transmit the monitored grain data to the virtual three-dimensional model; The data processing module divides the virtual model into three-dimensional grids, establishes virtual monitoring points according to the spatial coordinates of the grid nodes, and obtains the data of the virtual monitoring points through the formula. The specific formula is as follows: Among them, Z a is the monitoring data of the virtual monitoring point, Y i are the data of different monitoring points in the i-th sensor monitoring module, f i are the weight coefficients of different monitoring points in the i-th sensor monitoring module, and the weight coefficients of different monitoring points are adjusted according to the distance from the virtual monitoring point, and a distance monitoring distance threshold W is set. When the distance between different monitoring points and the virtual monitoring point is greater than the distance threshold W, the corresponding weight coefficient is set to 0; Intelligent early warning module, used to issue early warnings for abnormal data in the warehouse.

2. The big data-driven intelligent early warning system for grain storage according to claim 1 is characterized by: The sensor monitoring module is provided with a temperature sensor, a humidity sensor, an air density sensor and a wind direction sensor. The environmental data in the storage model is obtained through the different sensors in the sensor monitoring module, and the environmental data is imported into the virtual three-dimensional model. At the same time, the sensor monitoring module is also provided with an adjustment unit. The adjustment unit is used to control the ventilation equipment, humidification equipment, dehumidification equipment, cooling equipment and heating equipment in the warehouse, and the grain storage environment in the warehouse is adjusted through the adjustment unit.

3. The big data-driven grain storage intelligent early warning system according to claim 2 is characterized by: The data processing module verifies the deviation between the predicted data of the virtual monitoring point and the actual measured data through the root mean square error formula. The specific formula is as follows: Among them, Z b is the deviation between the virtual monitoring point and the actual data, is the data value predicted for the i-th virtual monitoring point, is the actual data periodically measured at the i-th virtual monitoring point.

4. The big data-driven grain storage intelligent early warning system according to claim 3 is characterized by: The data processing module is provided with a monitoring unit, and the monitoring unit is provided with a monitoring threshold X. Meanwhile, the monitoring threshold of the virtual monitoring point is X-|Z b |, when one of the temperature, humidity and pest density of the sensor monitoring point exceeds the preset monitoring threshold X and the data of the virtual monitoring point is greater than X-|Z b |When the data processing module controls the regulating unit to process the grain in the warehouse.

5. The big data-driven intelligent early warning system for grain storage according to claim 4 is characterized by: A simulation unit is set in the data processing module, the virtual three-dimensional model data is imported into the simulation unit, the environmental data of the warehouse is obtained by using the sensor in the sensor monitoring module, and then the working state parameters of the adjustment unit are simulated in the simulation unit according to the environmental data of the warehouse.

6. The big data-driven intelligent early warning system for grain storage according to claim 5 is characterized by: The simulation unit simulates the state data in the warehouse through a formula, and the specific formula is as follows: M=α·K+β·P+γ·A Among them, M is the simulation data of the simulation unit, K is the pest development prediction data, p is the storage environment prediction data, A is the mildew risk prediction data, α, β and γ are the weight coefficients of pest development prediction, storage environment prediction and mildew risk prediction respectively.

7. The big data-driven intelligent early warning system for grain storage according to claim 6 is characterized by: The storage environment prediction data is generated by the grain pile porosity and moisture and heat gradient. The storage environment prediction data is Among them, δ and ε are the weight coefficients of the linear state and nonlinear state in the pores of the grain pile, respectively; L is the air dynamic viscosity data inside the grain pile, which is obtained by the air velocity sensor; S is the permeability data inside the grain pile, which is obtained by the porosity detector; v is the ventilation efficiency data inside the grain pile, which is measured by the wind speed sensor; G is the inertia data of the air flow inside the grain pile; and D is the air density data in the grain.

8. The big data-driven intelligent early warning system for grain storage according to claim 7 is characterized by: The intelligent early warning module issues an early warning through the instruction data of the simulation unit. When the simulation unit calculates that the working efficiency of the adjustment unit is lower than the preset threshold, it is determined that it cannot be adjusted. The intelligent early warning module then broadcasts the abnormal data of the grain pile and generates a corresponding processing strategy.

Images