Grain weight detection device and method based on granary wall pressure

Abstract

The invention discloses a grain weight detection device and method based on granary wall pressure intensity, and relates to the technical field of grain weight detection.Pressure sensors are arranged on the side face of a granary, a networked monitoring system is formed, data of the pressure sensors are obtained, pressure changes in the granary are measured, and the sensors are arranged; the invention relates to a grain storage quantity function detection model and a model modeling method for a granary, which are based on QF perpendicular to 1 (1), QF perpendicular to 2 (2),..., QF perpendicular to N (N), QF angle 1 (1), QF angle 2 (2),..., QF angle N (N), QB1 (1), QB2 (2),..., and QBN (N), cover various heights of the granary, ensure the comprehensiveness of monitoring, and have the characteristics of high detection precision, suitability for various granary structure types, convenience in remote online granary quantity detection and the like. The pressure sensor is installed on the side wall of the granary at a time, normal grain warehousing and warehouse-out of the granary are not affected, the workload of workers is reduced, and managers are willing to accept the method.

Description

Technical Field

[0001] The present invention relates to the technical field of grain weight detection, and in particular to a grain weight detection device and method based on silo wall pressure. Background Art

[0002] Grain weight detection technology is a vital tool that relies on the deep integration of physical and electronic principles to accurately measure the volume or other related characteristics of grains, and then accurately calculate the actual weight of grains. This technology plays an indispensable role in many links such as grain procurement, storage, processing and trade.

[0003] The current technology for detecting grain weight in granaries is still to measure the weight of grain by large scales when the grain enters the warehouse, and record the total weight before entering the warehouse. However, the grain is not moved for a long time after entering the warehouse, during which time the grain in the warehouse may not be supervised by the superiors, and the warehouse managers may cheat for personal gain.

[0004] Now we urgently need a grain weight supervision and testing system for grain warehouses. If the current weight of grain in the warehouse can be measured at any time after the grain enters the warehouse and reported to the management department, it will solve the problem of lack of supervision of grain in the current warehouse and safeguard the safety of national grain storage.

[0005] Some domestic universities and research institutes have also conducted relevant experiments. For example, Chinese patent CN201510766195.5, published (announcement) date: 2016-03-23, discloses a grain silo weight detection device and method based on the relationship between grain pile height and bottom pressure. Some pressure sensors are installed at the bottom of the silo. After the grain is stored, the sensor is not easy to move. The relevant algorithm is used to monitor the grain weight in real time, which plays a very good role. However, this method installs pressure sensors at the bottom of the silo, which makes it very inconvenient for grain to be stored and unstored. Workers are unwilling to accept this method because the sensor is installed before each grain is stored and the sensor is retrieved after the grain is unstored, which greatly increases the workload of the workers. The grain warehouse managers also have resistance, and the implementation of this plan is difficult. Summary of the invention

[0006] The present invention proposes a grain weight detection device and method based on the pressure of the grain silo wall. The pressure sensor is installed not at the bottom of the grain silo but on the side of the grain silo wall. The installation is a one-time installation and will not be moved thereafter, which will not affect the normal grain storage and outbound transportation of the grain silo and reduce the workload of workers.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] The present invention provides a grain weight detection device based on the pressure of a granary wall. The granary is a cylindrical silo or a flat silo. A plurality of pressure sensors are installed on the side walls of the granary. The pressure sensors include a sensor bracket. The sensor bracket has a trapezoidal shape. A vertical measurement sensor and an inclined plane measurement sensor are installed inside the sensor bracket.

[0009] Preferably, the angle α between the inclined plane measuring sensor and the bottom surface of the granary is 25-45 degrees.

[0010] Preferably, a plurality of the pressure sensors are arranged in layers on the side wall of the granary, and only one pressure sensor is arranged on each layer.

[0011] Preferably, the distance between the pressure sensor at the bottom layer and the granary floor is 10-30 centimeters, and the distance between the pressure sensors at each layer is 1-2 meters.

[0012] The present invention also provides a method for detecting grain weight based on the granary wall pressure, comprising the following steps:

[0013] Step 1. Detect the output value of the multi-layer pressure sensor installed on the side of the granary;

[0014] Step 2. Use the multi-layer pressure sensor output value Q F⊥1 (1) Q F⊥2 (2) ........, Q F⊥N (N); Q F∠1 (1) Q F∠2 (2),......,Q F∠N (N), calculate the bottom pressure value of the grain pile Q B1 (1) Q B2 (2) ........, Q BN (N), construct the relationship between the output values ​​of the vertical sensor and the inclined plane sensor and the vertical pressure on the same horizontal plane:

[0015] Formula 1: Q B1 (1) = L1Q F∠1 (1)*cos(α)

[0016] Formula 2: Q B2 (2) = L2Q F∠2 (2)*cos(α) ..........

[0018] Formula 3: Q Bn (n) = L n Q F∠n (n)*cos(α)

[0019] Where Q Bn(n) The vertical pressure value of the nth horizontal plane, Q F∠n (n), Q F⊥n (n) are the values ​​of the slope pressure sensor and vertical sensor of the nth layer respectively; L n is the correction factor;

[0020] According to the principle of mechanical decomposition, Q F∠n (n), Q F⊥n (n)There are the following relations:

[0021] Formula 4: Q F⊥1 (1) = L1Q F∠1 (1)*sin(α)

[0022] Formula 5: Q F⊥2 (2) = L2Q F∠2 (2)*sin(α) ..........

[0024] Formula 6: Q F⊥n (n) = L n Q F∠n (n)*sin(α)

[0025] Step 3. Granary grain quantity detection process:

[0026] Grain weight detection below the first level sensor:

[0027] Formula 7: W(0)=A B H0ρ

[0028] Grain weight detection between 1st and 2nd layers:

[0029] Formula 8: W(1)=A B (Q B1 (1)-Q B2 (2))+A B J1f F H N Q F⊥1 (1)

[0030] Grain weight detection between 2nd → 3rd layer:

[0031] Formula 9: W(2)=A B (Q B2 (2)-Q B3 (3))+A B J2f F H (N-1) Q F⊥2 (2) ..........

[0033] Grain weight detection between layers n-1→n:

[0034] Formula 10: W(n)=A B (Q Bn (n)-Q B(n+1) (n+1))+A B J n f F H1Q F⊥n (n)

[0035] Total weight of grain in the granary:

[0036] Formula 11:

[0037] Where W(0), W(1), W(2), ..., W(N) are the weights of grains detected at each layer. is the total weight, A B is the bottom area of ​​the granary; H0 is 0.1-0.3 m, ρ is the grain density; J1, J2, ..., J N is the correction factor, H N ≤2, H (N-1) =2*(N-(N-1))+H N ,......,H1≤2*N+H N , f F It is the coefficient of friction between the side of the grain pile and the side of the granary.

[0038] Preferably, in step 3, the theoretical detection model for the grain storage quantity in the granary is:

[0039] Formula 12:

[0040]

[0041] Substituting formula 1, formula 2, formula 3, and formula 7 into formula 12, we get:

[0042] Formula 13:

[0043]

[0044] Formula 14 can be obtained from Formula 4, Formula 5, and Formula 6:

[0045]

[0046] Among them, H N-(n-1) is the height of the nth sensor.

[0047] Preferably, the detection model optimizes and learns various parameters, specifically

[0048] Formula 15:

[0049]

[0050] Where k is the sample point number, k = 1, 2, 3, ..., M, M is the number of samples; W k is the actual grain weight in the granary; through sample learning, the optimal value of the parameter is obtained:

[0051] when When the error E(J k ) minimum, is the optimal value of coefficient Jn.

[0052] Preferably, the optimal value calculation method of the parameter is:

[0053] Step 101. Detect and obtain sample data k=1 in bin 1. Substitute the known data into formula 14 to obtain:

[0054] Grain weight data below the first layer:

[0055] W(0)=A B H0ρ can be calculated from actual data;

[0056] Grain weight detection between 1st and 2nd layers:

[0057] Let n = 1, we can get formula 16:

[0058] W(1)=A B (Q F⊥1 (1)-Q F⊥2 (2))cot(α)+A B J1f F H N Q F⊥1 (1)

[0059] Grain weight detection between 2nd → 3rd layer:

[0060] Let n = 2, we can get formula 17:

[0061] W(2)=A B (Q F⊥2 (2)-Q F⊥3 (3))cot(α)+A B J2f F H (N-1) Q F⊥2 (2) ..........

[0063] Let n = N, we can get formula 18:

[0064] W(N)=A B Q F⊥N(N)cot(α)+A B J N f F H1Q F⊥N (N)

[0065] By actually measuring the actual weight of grain in each layer: W(0), W(1), W(2), ..., W(N), substituting the actual data, and solving the equations with known data from Formula 16 and Formula 17, it is easy to obtain J1 and J2. Similarly, continue to calculate to obtain J2, J3, ..., J N ;

[0066] Step 102. Detect and obtain sample data k=2 of bin 2. Repeat step 101. Obtain the parameters of the second sample: J1, J2, ..., J N ;

[0067] Step 103. Finally, perform multiple regression on the M sample data parameters to obtain the optimal parameter set

[0068] Step 104. Substitute the optimal parameters into formula 16, formula 17, and formula 18, and they can be used for actual granary data detection. The weight of grain in each layer is obtained, W(0), W(1), W(2), ..., W(N), so as to obtain the total weight of grain in the granary:

[0069] Grain weight detection between 1st and 2nd layers:

[0070] Let n = 1, we can get formula 21:

[0071]

[0072] Grain weight detection between 2nd → 3rd layer:

[0073] Let n = 2, we can get formula 22:

[0074] ..........

[0076] Let n = N, we can get formula 23:

[0077]

[0078] Further we get formula 24:

[0079]

[0080] Formula 24 is the final granary quantity detection model, and all system parameters have been calibrated.

[0081] Compared with the prior art, the pressure sensor used in the present invention is not installed at the bottom of the granary, but on the side of the granary wall. The installation is a one-time installation and will not be moved in the future. It will not affect the normal storage and delivery of grain in the granary, reducing the workload of workers. Management personnel are also willing to accept this method. The present invention establishes a granary weight detection model by arranging N pressure sensors on the side of the granary wall. The measuring sensor includes a sensor bracket, a housing, a level sensor, and an inclined sensor. The sensor has a simple structure, is easy to manufacture, and has a low cost, and is suitable for large-scale promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 This is a cross-sectional view of a single sensor fixed to a granary;

[0083] Figure 2 This is the main view of a single sensor fixed to a granary;

[0084] Figure 3 This is a top view of the sensor fixed on the granary;

[0085] Figure 4 Cross-sectional view of multiple sensors fixed to a granary.

[0086] In the figure, 1-side wall of granary; 2-pressure sensor; 21-sensor bracket; 22-vertical measurement sensor; 23-inclined measurement sensor; 3-bottom of granary; DETAILED DESCRIPTION

[0087] The technical solutions in the embodiments of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0088] like Figure 1-4 As shown, this embodiment provides a grain weight detection device based on the silo wall pressure, a pressure sensor 2 is installed on the silo side wall 1, the sensor bracket 21 is trapezoidal in shape, and two pressure sensors are installed inside, namely a vertical measurement sensor 22 and an inclined measurement sensor 23, which have a certain height distance from the silo bottom 3, wherein the angle α is 25-45 degrees, which does not hinder the normal operation of workers in feeding and discharging grain, and does not move after installation.

[0089] This embodiment takes a silo as an example. A silo is a special storage facility, which is characterized by its circular cylindrical structure and high degree of mechanization. The design of the silo allows a large amount of grain to be stored in a relatively small area. This structure not only improves the efficiency of land use, but also facilitates the mechanized storage and retrieval of grain.

[0090] The storage capacity of the silo is very strong. Its cylindrical structure can accommodate a large amount of grain. Due to its shape, the grain is more evenly distributed in the silo, which is conducive to improving storage efficiency. In addition, the silo also has good safety performance. Due to its closed structure, it can effectively prevent the grain from getting damp and moldy, thereby ensuring the quality of the grain.

[0091] Based on the structural characteristics of the silo, this measurement method is arranged on the side of the silo, which will make the measurement more accurate. There is no sensor at the bottom of the silo, only on the side wall. The silo has N layers of sensors, 1 sensor per layer, and a total of N sensors; the distance H1 between the lower layer and the silo ground is 10-30 cm; the height difference H of other adjacent layers is 1-2 meters.

[0092] This embodiment provides a method for detecting grain weight based on silo wall pressure, comprising the following steps:

[0093] Step 1. Detect the output value of the multi-layer pressure sensor installed on the side of the granary;

[0094] Step 2. Use the multi-layer pressure sensor output value Q F⊥1 (1) Q F⊥2 (2) ........, Q F⊥N (N); Q F∠1 (1) Q F∠2 (2),......,Q F∠N (N), calculate the bottom pressure value of the grain pile Q B1 (1) Q B2 (2) ........, Q BN (N), construct the relationship between the output values ​​of the vertical sensor and the inclined plane sensor and the vertical pressure on the same horizontal plane:

[0095] Formula 1: Q B1 (1) = L1Q F∠1 (1)*cos(α)

[0096] Formula 2: Q B2 (2) = L2Q F∠2 (2)*cos(α) ..........

[0098] Formula 3: Q Bn (n) = L n Q F∠n (n)*cos(α)

[0099] Where Q Bn (n) The vertical pressure value of the nth horizontal plane, Q F∠n (n), Q F⊥n(n) are the values ​​of the slope pressure sensor and vertical sensor of the nth layer respectively; L n is the correction factor;

[0100] According to the principle of mechanical decomposition, Q F∠n (n), Q F⊥n (n)There are the following relations:

[0101] Formula 4: Q F⊥1 (1) = L1Q F∠1 (1)*sin(α)

[0102] Formula 5: Q F⊥2 (2) = L2Q F∠2 (2)*sin(α) ..........

[0104] Formula 6: Q F⊥n (n) = L n Q F∠n (n)*sin(α)

[0105] Step 3. Granary grain quantity detection process:

[0106] Grain weight detection below the first level sensor:

[0107] Formula 7: W(0)=A B H0ρ

[0108] Grain weight detection between 1st and 2nd layers:

[0109] Formula 8: W(1)=A B (Q B1 (1)-Q B2 (2))+A B J1f F H N Q F⊥1 (1)

[0110] Grain weight detection between 2nd → 3rd layer:

[0111] Formula 9: W(2)=A B (Q B2 (2)-Q B3 (3))+A B J2f F H (N-1) Q F⊥2 (2) ..........

[0113] Grain weight detection between layers n-1→n:

[0114] Formula 10: W(n)=AB (Q Bn (n)-Q B(n+1) (n+1))+A B J n f F H1Q F⊥n (n)

[0115] Total weight of grain in the granary:

[0116] Formula 11:

[0117] Where W(0), W(1), W(2), ..., W(N) are the weights of grains detected at each layer. is the total weight, A B is the bottom area of ​​the granary; H0 is 0.1-0.3 m, ρ is the grain density; J1, J2, ..., J N is the correction factor, H N ≤2, H (N-1) =2*(N-(N-1))+H N ,......,H1≤2*N+H N , f F It is the coefficient of friction between the side of the grain pile and the side of the granary.

[0118] In step 3, the theoretical detection model of grain storage quantity in granary is:

[0119] Formula 12:

[0120]

[0121] Substituting formula 1, formula 2, formula 3, and formula 7 into formula 12, we get:

[0122] Formula 13:

[0123]

[0124] Formula 14 can be obtained from Formula 4, Formula 5, and Formula 6:

[0125]

[0126] Among them, H N-(n-1) is the height of the nth sensor.

[0127] Preferably, the detection model optimizes and learns various parameters, specifically

[0128] Formula 15:

[0129]

[0130] Where k is the sample point number, k = 1, 2, 3, ..., M, M is the number of samples; W k is the actual grain weight in the granary; through sample learning, the optimal value of the parameter is obtained:

[0131] when When the error E(J k ) minimum, is the optimal value of coefficient Jn.

[0132] The optimal value of the parameter is calculated as:

[0133] Step 101. Detect and obtain sample data k=1 in bin 1. Substitute the known data into formula 14 to obtain:

[0134] Grain weight data below the first layer:

[0135] W(0)=A B H0ρ can be calculated from actual data;

[0136] Grain weight detection between 1st and 2nd layers:

[0137] Let n = 1, we can get formula 16:

[0138] W(1)=A B (Q F⊥1 (1)-Q F⊥2 (2))cot(α)+A B J1f F H N Q F⊥1 (1)

[0139] Grain weight detection between 2nd → 3rd layer:

[0140] Let n = 2, we can get formula 17:

[0141] W(2)=A B (Q F⊥2 (2)-Q F⊥3 (3))cot(α)+A B J2f F H (N-1) Q F⊥2 (2) ..........

[0143] Let n = N, we can get formula 18:

[0144] W(N)=A B Q F⊥N (N)cot(α)+A B J N f F H1Q F⊥N(N)

[0145] By actually measuring the actual weight of grain in each layer: W(0), W(1), W(2), ..., W(N), substituting the actual data, and solving the equations with known data from Formula 16 and Formula 17, it is easy to obtain J1 and J2. Similarly, continue to calculate to obtain J2, J3, ..., J N ;

[0146] Step 102. Detect and obtain sample data k=2 of bin 2. Repeat step 101. Obtain the parameters of the second sample: J1, J2, ..., J N ;

[0147] Step 103. Finally, perform multiple regression on the M sample data parameters to obtain the optimal parameter set

[0148] Step 104. Substitute the optimal parameters into formula 16, formula 17, and formula 18, and they can be used for actual granary data detection. The weight of grain in each layer is obtained, W(0), W(1), W(2), ..., W(N), so as to obtain the total weight of grain in the granary:

[0149] Grain weight detection between 1st and 2nd layers:

[0150] Let n = 1, we can get formula 21:

[0151]

[0152] Grain weight detection between 2nd → 3rd layer:

[0153] Let n = 2, we can get formula 22:

[0154] ..........

[0156] Let n = N, we can get formula 23:

[0157]

[0158] Further we get formula 24:

[0159]

[0160] Formula 24 is the final granary quantity detection model, and all system parameters have been calibrated.

[0161] For the method provided in this embodiment, the following tests are performed:

[0162] The cylindrical silo used has a diameter of 25m, an actual grain loading height of 20m, and a bottom area of ​​490.625m2 For the silo, the pressure sensors are arranged vertically, starting from the bottom, the first sensor is 0.1m from the ground, the second sensor is 2m from the first sensor, and so on, for a total of 10 pressure sensors. The experimental grain type is wheat, with a density of 750 kg / m3.

[0163] For a given sensor, grain type, and silo type, if the model parameters have not been calibrated, pressure sensors are placed in more than 6 silos, and grain is added in layers. After the pressure sensor output value stabilizes, the next layer is added until the silo is full. The output values ​​of the pressure sensors in each silo are collected to form a sample set S = {Q F⊥i (i), Q F∠i (i)}, where i is the sample point number, i = 1, 2, 3, ... M,

[0164] M is the number of samples, Q F⊥i (i) is the value of the side pressure sensor at the i-th sample point, Q F∠i (i) is the value of the inclined pressure sensor at the i-th sample point, W i is the actual stratified feed weight at sample point i, A B i is the corresponding granary bottom area, and the sample set S is substituted into equations (21), (22), and (23) respectively, and the optimization and multivariate regression sample set S and the test sample S are optimized according to the optimization and multivariate regression sample set S, and the optimization algorithm is used to optimize the parameters. The actual detection is carried out according to the granary weight detection model shown in formula (24).

[0165] The parameter estimates of the constructed model are shown in Table 1:

[0166] Table 1 Parameters

[0167]

[0168] Through these data detection and calculation, the amount of grain storage inside the granary can be understood in real time, thus realizing networked grain storage quantity monitoring.

[0169] Silo side pressure detection technology provides an efficient and convenient monitoring method for the grain storage industry, helping to ensure safe grain storage. In addition, silo side pressure detection technology has many advantages. First, it has the characteristics of non-contact measurement, which means that no damage will be caused to the grain during the detection process, ensuring the integrity and quality of the grain. Secondly, by monitoring the pressure changes on the side of the silo in real time, the technology can promptly detect and warn of potential safety hazards, such as uneven grain stacking and damaged silo walls.

[0170] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A grain weight detection device based on the pressure of the granary wall, wherein the granary is a cylindrical silo or a flat silo, characterized in that: A plurality of pressure sensors are installed on the side wall of the granary. The pressure sensors include a sensor bracket. The sensor bracket has a trapezoidal shape. A vertical measurement sensor and an inclined plane measurement sensor are installed inside the sensor bracket.

2. A grain weight detection device based on silo wall pressure according to claim 1, characterized in that: The included angle α between the inclined plane measuring sensor and the bottom surface of the granary is 25-45 degrees.

3. The grain weight detection device based on the granary wall pressure according to claim 1 is characterized in that: A plurality of the pressure sensors are arranged in layers on the side wall of the granary, and only one pressure sensor is arranged in each layer.

4. The grain weight detection device based on the granary wall pressure according to claim 3 is characterized in that: The distance between the pressure sensor on the bottom layer and the granary floor is 10-30 cm, and the distance between the pressure sensors on each layer is 1-2 meters.

5. A method for detecting grain weight based on silo wall pressure, using the detection device provided in claims 1-4, characterized in that: The following steps are involved: Step 1. Detect the output value of the multi-layer pressure sensor installed on the side of the granary; Step 2. Use the multi-layer pressure sensor output value Q F⊥1 (1) Q F⊥2 (2) ........, Q F⊥N (N); Q F∠1 (1) Q F∠2 (2),......,Q F∠N (N), calculate the bottom pressure value of the grain pile Q B1 (1) Q B2 (2) ........, Q BN (N), construct the relationship between the output values ​​of the vertical sensor and the inclined plane sensor and the vertical pressure on the same horizontal plane: Formula 1: Q B1 (1) = L1Q F∠1 (1)*cos(α) Formula 2: Q B2 (2) = L2Q F∠2 (2)*cos(α) .......... Formula 3: Q Bn (n) = L n Q F∠n (n)*cos(α) Where Q Bn (n) The vertical pressure value of the nth horizontal plane, Q F∠n (n), Q F⊥n (n) are the values ​​of the slope pressure sensor and vertical sensor of the nth layer respectively; L n is the correction factor; According to the principle of mechanical decomposition, Q F∠n (n), Q F⊥n (n)There are the following relations: Enclosure 4:Q F⊥1 (1)DL1Q F∠1 (1)*sin(α) Enclosure 5:Q F⊥2 (2)DL2Q F∠2 (2)*sin(α) .......... Formula 6: Q F⊥n L(n) = L n Q F∠n L(n) * sin(α) Step 3. Granary grain quantity detection process: Grain weight detection below the first level sensor: Formula 7: W(0)=A B H0ρ Grain weight detection between 1st → 2nd layer: Formula 8: W(1)=A B (Q B1 (1)-Q B2 (2))+A B J1f F H N Q F⊥1 (1) Grain weight detection between 2nd → 3rd layer: Formula 9: W(2)=A B (Q B2 (2)-Q B3 (3))+A B J2f F H (N-1) Q F⊥2 (2) .......... Grain weight detection between layers n-1→n: Formula 10: W(n) = A B (Q Bn (n) - Q B(n+1) (n + 1)) + A B J n f F H1Q F⊥n (n) Total weight of grain in the granary: Formula 11: Where W(0), W(1), W(2), ..., W(N) are the weights of grains detected at each layer. is the total weight, A B is the bottom area of ​​the granary; H0 is 0.1-0.3 m, ρ is the grain density; J1, J2, ..., J N is the correction factor, H N ≤2, H (N-1) =2*(N-(N-1))+H N ,......,H1≤2*N+H N , f F It is the coefficient of friction between the side of the grain pile and the side of the granary.

6. A method for detecting grain weight based on silo wall pressure according to claim 5, characterized in that: In step 3, the theoretical detection model for the amount of grain stored in the granary is: Formula 12: Substituting formula 1, formula 2, formula 3, and formula 7 into formula 12, we get: Formula 13: Formula 14 can be obtained from Formula 4, Formula 5, and Formula 6: Among them, H N-(n-1) is the height of the nth sensor.

7. A method for detecting grain weight based on silo wall pressure according to claim 6, characterized in that: In the detection model, various parameters are optimized and learned, specifically as Formula 15: Where k is the sample point number, k = 1, 2, 3, ..., M, M is the number of samples; W k is the actual grain weight in the granary; through sample learning, the optimal value of the parameter is obtained: when When the error E(J k ) minimum, is the optimal value of coefficient Jn.

8. The method for detecting grain weight based on silo wall pressure according to claim 7, characterized in that: The optimal value of the parameter is calculated as: Step 101. Detect and obtain sample data k=1 in bin 1. Substitute the known data into formula 14 to obtain: Grain weight data below the first layer: W(0)=A B H0ρ can be calculated from actual data; Grain weight detection between 1st → 2nd layer: Let n = 1, we can get formula 16: W(1)=A B (Q F⊥1 (1)-Q F⊥2 (2))cot(α)+A B J1f F H N Q F⊥1 (1) Grain weight detection between 2nd → 3rd layer: Let n = 2, we can get formula 17: W(2)=A B (Q F⊥2 (2)-Q F⊥3 (3))cot(α)+A B J2f F H (N-1) Q F⊥2 (2) .......... Let n = N, we can get formula 18: W(N)=A B Q F⊥N (N)cot(α)+A B J N f F H1Q F⊥N (N) By actually measuring the actual weight of grain in each layer: W(0), W(1), W(2), ..., W(N), substituting the actual data, and solving the equations with known data from Formula 16 and Formula 17, it is easy to obtain J1 and J2. Similarly, continue to calculate to obtain J2, J3, ..., J N ; Step 102. Detect and obtain sample data k=2 of bin 2. Repeat step 101. Obtain the parameters of the second sample: J1, J2, ..., J N ; Step 103. Finally, perform multiple regression on the M sample data parameters to obtain the optimal parameter set Step 104. Substitute the optimal parameters into formula 16, formula 17, and formula 18, and they can be used for actual granary data detection. The weight of grain in each layer is obtained, W(0), W(1), W(2), ..., W(N), so as to obtain the total weight of grain in the granary: Grain weight detection between 1st → 2nd layer: Let n = 1, we can get formula 21: Grain weight detection between 2nd → 3rd layer: Let n = 2, we can get formula 22: .......... Let n = N, we can get formula 23: Further we get formula 24: Formula 24 is the final granary quantity detection model, and all system parameters have been calibrated.

Images