A cross flow batch cycle grain dryer drying cycle demarcation point prediction positioning system and method
By real-time detection and the use of data fusion and neural network technology in a cross-flow batch circulating grain dryer, the problem of inaccurate positioning of the cycle boundary point was solved, and precise control and energy consumption optimization of the grain drying process were achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGSU UNIV
- Filing Date
- 2023-06-26
- Publication Date
- 2026-06-12
AI Technical Summary
In the existing technology, the cross-flow batch circulating grain dryer lacks a reliable strategy for positioning the circulation boundary point during the rice drying process, which makes it difficult to achieve the expected drying effect. In addition, the random sampling detection method is often used, which causes the control point to deviate from the actual boundary point.
The internal status of the grain dryer is monitored in real time using a 3D radar level scanner, temperature sensor, speed sensor, and grain moisture sensor. Combined with signal processing circuit and data processing system, a convolutional neural network with fusion attention mechanism is used to accurately locate the drying cycle boundary point. Online positioning is achieved through data fusion and correction optimization.
It improves the positioning accuracy of the drying cycle boundary point, realizes precise control of the grain drying process, and enhances the drying effect and energy efficiency.
Smart Images
Figure CN116818826B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of precise control of the drying process in agricultural drying machinery, and in particular to a system and method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer. Background Technology
[0002] Mechanized grain drying is a crucial step in post-harvest grain processing. Precise control of the grain drying process based on optimized drying technology can effectively reduce energy consumption and ensure grain quality. When using optimized drying technology for mechanized grain drying, given the batch-cycle drying characteristics of cross-flow batch dryers for paddy rice, it is necessary to accurately locate the cycle boundary points between each cycle. Then, at these cycle boundary points, the drying process parameters can be optimized and controlled according to the optimized drying technology.
[0003] Current methods for determining the cycle boundary point in rice drying in grain dryers are outdated, lacking a reliable strategy for locating the cycle boundary point. Consequently, random sampling is often used in actual operations. When using random sampling to control the rice drying process, the grain dryer operator or control system checks the rice moisture content at least three times consecutively. If the average moisture content is close to or lower than the target moisture content at the control node, drying parameters are adjusted or drying is terminated. However, the control point is prone to deviating from the rice drying cycle boundary point. This outdated method for determining the rice drying cycle boundary point in grain dryers makes it difficult to achieve the expected drying effect of the segmented temperature drying process. Therefore, researching a predictive positioning system and method for the drying cycle boundary point in a cross-flow batch grain dryer is of significant practical importance. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides a predictive positioning system and method for the drying cycle boundary point of a cross-flow batch circulating grain dryer, which effectively solves the problem of poor positioning accuracy of the drying cycle boundary point during the grain drying process.
[0005] The present invention achieves the above-mentioned technical objectives through the following technical means.
[0006] A predictive positioning system for the drying cycle boundary point of a crossflow batch circulating grain dryer includes a 3D radar level scanner, a temperature sensor, a speed sensor, a grain moisture sensor, a signal processing circuit, a data processing system, and a human-machine interface; the 3D radar level scanner, temperature sensor, speed sensor, and grain moisture sensor are connected to the signal processing circuit; the signal processing circuit is connected to the data processing system; and the data processing system is connected to the human-machine interface.
[0007] In the above scheme, the 3D radar level scanner is installed on top of the grain dryer.
[0008] In the above scheme, the temperature sensor is installed in the air inlet duct of the drying section of the grain dryer.
[0009] In the above scheme, the speed sensor is installed on the feeding mechanism of the grain dryer.
[0010] In the above scheme, the 3D radar level scanner is used to detect the grain accumulation structure inside the grain dryer in real time; the temperature sensor is used to detect the temperature of the drying hot air in real time; the speed sensor is used to detect the speed of the feeding mechanism in real time; and the grain moisture sensor detects the grain moisture content in real time.
[0011] In the above scheme, the grain moisture sensor is installed in the grain flow channel between the screw conveyor and the elevator of the grain dryer.
[0012] In the above scheme, the signal processing circuit filters out data noise and abnormal data points, and applies a sliding time average filtering algorithm to filter the data.
[0013] In the above scheme, the data processing system uses a convolutional neural network with a fusion attention mechanism to perform data fusion prediction and localization of the drying cycle boundary point.
[0014] A method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer includes the following steps:
[0015] Step 1: When using a cross-flow batch circulating grain dryer to dry grains, a 3D radar level scanner is used to detect the grain accumulation structure inside the grain dryer, a temperature sensor is used to detect the drying hot air temperature of the grain dryer, a speed sensor is used to detect the speed of the feeding mechanism of the grain dryer, and a grain moisture sensor is used to detect the grain moisture content.
[0016] Step 2: Use signal processing circuit to process the output signals of 3D radar level scanner, temperature sensor, speed sensor and grain moisture sensor and upload them to data processing system for processing, to obtain and record grain volume time series data, drying hot air temperature time series data, feeding mechanism speed time series data and grain moisture content time series data.
[0017] Step 3: In the initial stage of grain drying, the data processing system, based on the structural parameters of the grain dryer and the principle of heat and mass exchange during drying, calculates the drying cycle boundary between the first and second cycle based on the grain volume time-series data, the drying hot air temperature time-series data, the feeding mechanism speed time-series data, and the grain moisture content time-series data. This point is the drying cycle boundary between the first and second cycle.
[0018] Step four, the second cycle execution phase: Based on the grain moisture content time-series data, the data processing system uses an early classification and identification method to predict the end point of the second cycle, which is the predicted drying cycle boundary point a between the second and third cycles; During the drying process of the second cycle, based on the start point of the second cycle, the drying hot air temperature time-series data, the feeding mechanism speed time-series data, and the grain moisture content time-series data, the data processing system predicts the end point of the second cycle based on the drying heat and mass exchange principle, which is the predicted drying cycle boundary point b between the second and third cycles;
[0019] Step 5: The data processing system uses a convolutional neural network with a fusion attention mechanism to fuse data at the drying cycle prediction boundary point a and the drying cycle prediction boundary point b, and determines the drying cycle fusion prediction boundary point between the second cycle and the third cycle.
[0020] Step Six: After the second cycle of drying is completed, the data processing system, based on the structural parameters of the grain dryer and the principle of heat and mass exchange during drying, locates the drying cycle correction boundary point between the second and third cycles according to the grain volume time-series data, drying hot air temperature time-series data, feeding mechanism speed time-series data, and grain moisture content time-series data. The key parameters of the precise positioning method for the drying cycle boundary point are then optimized online based on the difference between the drying cycle correction boundary point and the drying cycle fusion prediction boundary point.
[0021] Step 7: The data processing system transmits the grain moisture content time series data, drying cycle fusion prediction boundary point, and drying cycle correction boundary point information of the grain dryer to the human-machine interface for display, and performs real-time positioning and display of the drying cycle boundary point.
[0022] Step 8: When entering the next cycle, repeat steps 4 to 7 until the target termination moisture content is reached, thus realizing the prediction and positioning of the drying cycle boundary point of the entire grain drying process.
[0023] The formula for calculating the drying cycle boundary point between cycles in the above scheme is as follows:
[0024]
[0025]
[0026]
[0027] Among them, V ib— Total volume of grain in the dryer at the start of the i-th cycle; V1— Volume of grain in the tempering section, drying section, and flow channel at the start of the i-th cycle; V2— Volume of grain in the screw conveyor at the start of the i-th cycle; V3— Volume of grain in the elevator at the start of the i-th cycle; M— Time-series data of grain moisture content; T— Time-series data of drying temperature; w— Time-series data of feed mechanism rotation speed; t ib —The starting time of the i-th cycle; f(M,T,w,t) ib ,t i )—t ib Time begins t i The volume of grain flowing through the grain moisture sensor within a given time period; f(M,T,w,t) ib ,t j )—t ib Time begins t j The volume of grain flowing through the grain moisture sensor within a given time period; G(M,T,w,t) ib +t j ,t p )—t ib +t j Time begins t p Predicts the volume of grain flowing through the grain moisture sensor within a given time period; t ie —The drying cycle correction boundary point for the i-th cycle; t iep —Predicted boundary point b for the i-th cycle of the drying cycle;
[0028]
[0029]
[0030]
[0031] Among them, h ib (x,y)—Grain height value detected by 3D radar level scanner at the plane coordinates (x,y) of the grain dryer; V0—Grain volume in the drying section and flow channel; t l —Grain transport time using a screw conveyor; t h —Grain transport time by elevator.
[0032] Beneficial effects:
[0033] This invention continuously detects grain level, drying temperature, feed mechanism rotation speed, and grain moisture during the grain drying process. Based on early classification and identification methods and the principle of drying heat and mass exchange, it uses a data fusion method to accurately locate the drying cycle boundary point in a cross-flow batch circulating grain dryer. This effectively solves the problem of poor positioning accuracy of the drying cycle boundary point in grain dryers, and further improves the control accuracy of grain drying process parameters at the drying cycle boundary point. Attached Figure Description
[0034] Figure 1 This is a schematic diagram of the drying cycle boundary point prediction and positioning system for the cross-flow batch circulating grain dryer involved in the present invention.
[0035] Figure 2 Flowchart of the predictive positioning system for the drying cycle boundary point of a crossflow batch circulating grain dryer;
[0036] Figure 3 This is a schematic diagram of the grain volume distribution inside a crossflow batch circulating grain dryer.
[0037] Figure label:
[0038] 1-3D radar level scanner; 2-Temperature sensor; 3-Speed sensor; 4-Grain moisture sensor; 5-Signal processing circuit; 6-Data processing system; 7-Human-machine interface. Detailed Implementation
[0039] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.
[0040] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "axial," "radial," "vertical," "horizontal," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.
[0041] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0042] Combined with appendix Figure 1-3 As shown, a predictive positioning system for the drying cycle boundary point of a cross-flow batch circulating grain dryer includes a 3D radar level scanner 1, a temperature sensor 2, a speed sensor 3, a grain moisture sensor 4, a signal processing circuit 5, a data processing system 6, and a human-machine interface 7. Figure 2 As shown, the 3D radar level scanner 1, temperature sensor 2, speed sensor 3, and grain moisture sensor 4 are connected to the signal processing circuit 5; the signal processing circuit 5 is connected to the data processing system 6; and the data processing system 6 is connected to the human-machine interface 7. Figure 1 As shown, the 3D radar level scanner 1 is installed on the top of the grain dryer; the temperature sensor 2 is installed in the air inlet duct of the drying section of the grain dryer; the speed sensor 3 is installed on the feeding mechanism of the grain dryer; and the grain moisture sensor 4 is installed in the grain flow channel between the screw conveyor and the elevator of the grain dryer.
[0043] A method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer includes the following steps:
[0044] Step 1: When using a cross-flow batch circulating grain dryer to dry grains, a 3D radar level scanner 1 is used to detect the grain accumulation structure inside the grain dryer, a temperature sensor 2 is used to detect the drying hot air temperature of the grain dryer, a speed sensor 3 is used to detect the speed of the feeding mechanism of the grain dryer, and a grain moisture sensor 4 is used to detect the grain moisture content.
[0045] Step 2: The signal processing circuit 5 processes the output signals of the 3D radar level scanner 1, temperature sensor 2, speed sensor 3, and grain moisture sensor 4 and uploads them to the data processing system 6 to record the grain volume time-series data, drying hot air temperature time-series data, feeding mechanism speed time-series data, and grain moisture content time-series data.
[0046] Step 3: In the initial stage of grain drying, the data processing system 6, based on the structural parameters of the grain dryer and the principle of heat and mass exchange during drying, calculates the drying cycle boundary between the first and second cycle based on the grain volume time-series data, the drying hot air temperature time-series data, the feeding mechanism speed time-series data, and the grain moisture content time-series data. This point is the drying cycle boundary between the first and second cycle.
[0047] Step four, the second cycle execution phase: Based on the grain moisture content time series data, the data processing system 6 uses an early classification and identification method to predict the end point of the second cycle, which is the drying cycle prediction boundary point a between the second and third cycles; During the drying process of the second cycle, based on the start point of the second cycle, the drying hot air temperature time series data, the feeding mechanism speed time series data, and the grain moisture content time series data, the data processing system 6 predicts the end point of the second cycle based on the drying heat and mass exchange principle, which is the drying cycle prediction boundary point b between the second and third cycles;
[0048] Step 5: The data processing system 6 uses a convolutional neural network with a fusion attention mechanism to fuse data at the drying cycle prediction boundary point a and the drying cycle prediction boundary point b, and determines the drying cycle fusion prediction boundary point between the second cycle and the third cycle.
[0049] Step 6: After the second cycle of drying is completed, the data processing system 6, based on the structural parameters of the grain dryer and the principle of heat and mass exchange during drying, locates the drying cycle correction boundary point between the second and third cycles according to the grain volume time-series data, drying hot air temperature time-series data, feeding mechanism speed time-series data, and grain moisture content time-series data. The key parameters of the precise positioning method for the drying cycle boundary point are optimized online based on the difference between the drying cycle correction boundary point and the drying cycle fusion prediction boundary point.
[0050] Step 7: The data processing system 6 transmits the grain moisture content time series data, drying cycle fusion prediction boundary point, and drying cycle correction boundary point information of the grain dryer to the human-machine interface 7 for display, and performs real-time positioning and display of the drying cycle boundary point.
[0051] Step 8: When entering the next cycle, repeat steps 4 to 7 until the target termination moisture content is reached, thus realizing the prediction and positioning of the drying cycle boundary point of the entire grain drying process.
[0052] The formula for calculating the drying cycle boundary point between the cycle periods is as follows:
[0053]
[0054]
[0055]
[0056] Among them, V ib — Total volume of grain in the dryer at the start of the i-th cycle; V1— Volume of grain in the tempering section, drying section, and flow channel at the start of the i-th cycle; V2— Volume of grain in the screw conveyor at the start of the i-th cycle; V3— Volume of grain in the elevator at the start of the i-th cycle; M— Time-series data of grain moisture content; T— Time-series data of drying temperature; w— Time-series data of feed mechanism rotation speed; t ib —The starting time of the i-th cycle; f(M,T,w,t) ib ,t i )—t ib Time begins t i The volume of grain flowing through the grain moisture sensor within a given time period; f(M,T,w,t) ib ,t j )—t ib Time begins t j The volume of grain flowing through grain moisture sensor 4 within a given time period; G(M,T,w,t) ib +t j ,t p )—t ib +t j Time begins t p Predict the volume of grain flowing through grain moisture sensor 4 within a given time period; t ie —The drying cycle correction boundary point for the i-th cycle; t iep —Predicted dividing point b for the i-th cycle drying cycle.
[0057]
[0058]
[0059]
[0060] Where: h ib (x,y)—Grain height value detected by 3D radar level scanner 1 at the plane coordinates (x,y) of the grain dryer; V0—Grain volume in the drying section and flow channel; t l —Grain transport time using a screw conveyor; t h —Grain transport time by elevator.
[0061] The cross-flow batch circulating grain dryer drying cycle boundary point prediction and positioning system provided by this invention uses temperature sensor 2, speed sensor 3, and grain moisture sensor 4 to monitor the drying hot air temperature, feeding mechanism speed, and grain moisture content in real time during the grain drying process. This data is uploaded to the data processing system 6 via signal processing circuit 5 and recorded, establishing a time-series dataset of drying temperature, feeding mechanism speed, and grain moisture content. At the beginning of any cycle, a 3D radar level scanner 1 detects the grain accumulation structure inside the grain dryer and uploads this data to the data processing system 6 via signal processing circuit 5. Based on the time-series datasets from each sensor, the grain volume inside the grain dryer is numerically calculated. The data processing system 6, based on the drying hot air time-series data, feeding mechanism speed time-series data, grain moisture content time-series data, and grain volume at the beginning of the cycle, uses a data fusion method based on early classification and identification methods and the principle of drying heat and mass exchange to accurately locate the drying cycle boundary point during the rice drying process in the cross-flow batch circulating grain dryer. The system also corrects and optimizes the key parameters of the accurate positioning method for the drying cycle boundary point, improving the positioning accuracy of the drying cycle boundary point. The data processing system 6 transmits the time-series data of grain moisture content in the grain dryer, the predicted boundary point of the drying cycle fusion, and the information of the correction boundary point of the drying cycle to the human-machine interface 7 for display, and performs real-time positioning and display of the drying cycle boundary point, so as to realize the online and accurate positioning of the drying cycle boundary point during the grain drying process of the grain dryer.
[0062] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0063] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.
Claims
1. A method for predicting and locating the drying cycle boundary point of a crossflow batch circulating grain dryer, characterized in that, The system includes a 3D radar level scanner (1), a temperature sensor (2), a rotation speed sensor (3), a grain moisture sensor (4), a signal processing circuit (5), a data processing system (6), and a human-machine interface (7); the 3D radar level scanner (1), temperature sensor (2), rotation speed sensor (3), and grain moisture sensor (4) are connected to the signal processing circuit (5); the signal processing circuit (5) is connected to the data processing system (6); the data processing system (6) is connected to the human-machine interface (7); the predictive positioning method includes the following steps: Step 1: When the grain is dried using a cross-flow batch circulating grain dryer, a 3D radar level scanner (1) is used to detect the grain accumulation structure inside the grain dryer, a temperature sensor (2) is used to detect the drying hot air temperature of the grain dryer, a speed sensor (3) is used to detect the speed of the feeding mechanism of the grain dryer, and a grain moisture sensor (4) is used to detect the grain moisture content. Step 2: Use signal processing circuit (5) to process the output signals of 3D radar level scanner (1), temperature sensor (2), speed sensor (3), and grain moisture sensor (4) and upload them to data processing system (6) for processing, and obtain and record grain volume time series data, drying hot air temperature time series data, feeding mechanism speed time series data, and grain moisture content time series data. Step 3: In the initial stage of grain drying, the data processing system (6) calculates the drying cycle boundary between the first cycle and the second cycle based on the structural parameters of the grain dryer and the drying heat and mass exchange principle, according to the grain volume time sequence data, drying hot air temperature time sequence data, feeding mechanism speed time sequence data, and grain moisture content time sequence data. This is the drying cycle boundary between the first cycle and the second cycle. Step 4, the second cycle execution stage: the data processing system (6) predicts the end point of the second cycle based on the grain moisture content time series data and the early classification and identification method, which is the drying cycle prediction boundary point a between the second cycle and the third cycle; during the drying process of the second cycle, the data processing system (6) predicts the end point of the second cycle based on the drying heat and mass exchange principle, based on the starting point of the second cycle, the drying hot air temperature time series data, the material feeding mechanism speed time series data, and the grain moisture content time series data, which is the drying cycle prediction boundary point b between the second cycle and the third cycle. Step 5, the data processing system (6) uses a convolutional neural network with a fusion attention mechanism to fuse data on the drying cycle prediction boundary point a and the drying cycle prediction boundary point b, and determines the drying cycle fusion prediction boundary point between the second cycle and the third cycle. Step 6: After the second cycle of drying is completed, the data processing system (6) locates the drying cycle correction boundary point between the second and third cycles based on the structural parameters of the grain dryer and the drying heat and mass exchange principle, according to the grain volume time series data, drying hot air temperature time series data, feeding mechanism speed time series data, and grain moisture content time series data. The key parameters of the drying cycle boundary point accurate positioning method are optimized online based on the difference between the drying cycle correction boundary point and the drying cycle fusion prediction boundary point. Step 7: The data processing system (6) transmits the grain moisture content time series data, drying cycle fusion prediction boundary point, and drying cycle correction boundary point information of the grain dryer to the human-machine interface (7) for display, and performs real-time positioning display of the drying cycle boundary point. Step 8: When entering the next cycle, repeat steps 4 to 7 until the target termination moisture content is reached, thus achieving the prediction and positioning of the drying cycle boundary point in the entire grain drying process. The formula for calculating the drying cycle boundary point between cycles is as follows: , , Among them, V ib — Total volume of grain in the dryer at the start of the i-th cycle; V1— Volume of grain in the tempering section, drying section, and flow channel at the start of the i-th cycle; V2— Volume of grain in the screw conveyor at the start of the i-th cycle; V3— Volume of grain in the elevator at the start of the i-th cycle; M— Time-series data of grain moisture content; T— Time-series data of drying temperature; w— Time-series data of feed mechanism rotation speed; t ib —The starting time of the i-th cycle; f(M,T,w,t) ib ,t i )—t ib Time begins t i The volume of grain flowing through the grain moisture sensor within a given time period; f(M,T,w,t) ib ,t j )—t ib Time begins t j The volume of grain flowing through the grain moisture sensor (4) within a given time period; G(M,T,w,t) ib +t j ,t p )—t ib +t j Time begins t p Predict the volume of grain flowing through the grain moisture sensor (4) within a given time period; t ie —The drying cycle correction boundary point for the i-th cycle; t iep —Predicted boundary point b for the i-th cycle of the drying cycle; , , Among them, h ib (x,y)—Grain height value detected by 3D radar level scanner (1) at the plane coordinates (x,y) of the grain dryer; V0—Grain volume in the drying section and flow channel; t l —Grain transport time using a screw conveyor; t h —Grain transport time by elevator.
2. The method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer according to claim 1, characterized in that, The 3D radar level scanner (1) is installed on top of the grain dryer.
3. The method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer according to claim 1, characterized in that, The temperature sensor (2) is installed in the air inlet duct of the drying section of the grain dryer.
4. The method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer according to claim 1, characterized in that, The speed sensor (3) is installed on the feeding mechanism of the grain dryer.
5. The method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer according to claim 1, characterized in that, The 3D radar level scanner (1) is used to detect the grain accumulation structure inside the grain dryer in real time; the temperature sensor (2) is used to detect the temperature of the drying hot air in real time; the speed sensor (3) is used to detect the speed of the feeding mechanism in real time; and the grain moisture sensor (4) is used to detect the grain moisture content in real time.
6. The method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer according to claim 1, characterized in that, The grain moisture sensor (4) is installed in the grain flow channel between the screw conveyor and the elevator of the grain dryer.
7. The method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer according to claim 1, characterized in that, The signal processing circuit (5) filters out data noise and abnormal data points, and applies a sliding time average filtering algorithm to filter the data.
8. The method for predicting and locating the drying cycle boundary point of a cross-flow batch circulating grain dryer according to claim 1, characterized in that, The data processing system (6) uses a convolutional neural network with a fusion attention mechanism to perform data fusion prediction and localization of the drying cycle boundary point.