CN111721448A

CN111721448A - Granary detection method and device based on bottom surface pressure intensity statistic and reserve equation

Info

Publication number: CN111721448A
Application number: CN202010549653.0A
Authority: CN
Inventors: 张德贤; 张弈晨; 张苗; 邓淼磊
Original assignee: Henan University of Technology
Current assignee: Henan University of Technology
Priority date: 2020-06-16
Filing date: 2020-06-16
Publication date: 2020-09-29
Anticipated expiration: 2040-06-16
Also published as: CN111721448B

Abstract

The invention belongs to the technical field of granary quantity detection, and particularly relates to a granary detection method and device based on bottom surface pressure statistic and reserve equation. The invention utilizes the output value of the two circles of pressure sensors on the bottom surface or the output value of the single circle of pressure sensors on the bottom surface to determine the pressure mean value on the bottom surface and the pressure mean value on the side surface, and the pressure mean values are substituted into the constructed granary storage quantity model, so that the granary storage quantity can be determined. The invention is characterized in that the relationship between the grain bulk height and the grain storage quantity of the granary is introduced to construct a granary grain storage quantity model so as to improve the robustness and generalization capability of the granary grain storage quantity detection model, is suitable for the structural types of horizontal warehouses and the like, is convenient for the remote online detection of the granary quantity and the like, and can meet the requirement of the remote online detection of the grain storage quantity of the horizontal warehouses and the like.

Description

Granary detection method and device based on bottom surface pressure intensity statistic and reserve equation

Technical Field

The invention belongs to the technical field of granary quantity detection, and particularly relates to a granary detection method and device based on bottom surface pressure statistic and reserve equation.

Background

The online detection of the grain quantity is an important guarantee technology for national grain quantity safety, and the development of research and application in the aspect of national grain safety has important significance and can generate great social and economic benefits.

Due to the important position of the grain in national safety, the grain quantity online detection is required to be accurate, rapid and reliable. Meanwhile, the grain quantity is huge, the price is low, and the grain quantity online detection equipment is required to be low in cost, simple and convenient. Therefore, the high precision of detection and the low cost of the detection system are key problems which need to be solved for the development of the online grain quantity detection system.

For example, the chinese patent application publication No. CN110823338A discloses a method and a system for detecting a grain bin based on a single-turn log model of standard deviation of the bottom surface, in which a single-turn pressure sensor is arranged on the bottom surface of the grain bin, the output values of the sensor are used to determine the bottom pressure mean value, the grain pile height, and the average friction force estimation term of the unit area of the side surface, and the grain storage quantity of the grain bin can be determined after the three parameters are determined.

For another example, the chinese patent application publication No. CN110823344A discloses a granary detection method and system based on a bottom surface double-ring standard deviation SVM logarithmic model, in which a double-ring pressure sensor is arranged on the bottom surface of a granary, the output values of the sensor are used to determine the bottom surface pressure mean value, the grain pile height, and the side surface unit area average friction force estimation terms, and after the three parameters are determined, the grain storage quantity of the granary can be determined.

Disclosure of Invention

The invention provides a granary detection method and device based on bottom surface pressure statistic and reserve equation, which are used for detecting the grain storage quantity of a granary.

In order to solve the technical problems, the technical scheme and the beneficial effects of the invention comprise that:

the invention provides a granary reserve detection method based on two circles of pressure intensity statistics of a bottom surface, which determines the bottom by utilizing the output value of a two circles of pressure sensors of the bottom surfaceSubstituting the surface pressure mean value and the side pressure mean value into the constructed granary storage quantity model to determine the granary storage quantity; wherein, the granary grain storage quantity model is as follows: w is A_BQ_BNF(s)，

The invention is characterized in that the relationship between the height of the grain pile and the grain storage quantity of the grain bin is introduced to construct a grain storage quantity model of the grain bin, wherein the model comprises the primary relationship, the secondary relationship, the tertiary relationship and the quartic relationship between the height of the grain pile and the grain storage quantity of the grain bin, so that the robustness and the generalization capability of the grain storage quantity detection model of the grain bin are improved, the detection model is suitable for structural types of horizontal warehouses and the like, the detection of the quantity of the grain bin in a remote online manner is convenient, and the requirements of the remote online detection of the grain storage quantity of the grain bins such as the horizontal warehouses and.

The invention also provides a granary storage capacity detection method based on the bottom surface single-circle pressure intensity statistic, which is characterized in that the output value of the bottom surface single-circle pressure intensity sensor is utilized to determine the bottom surface pressure intensity mean value and the side surface pressure intensity mean value, and the output values are substituted into the constructed granary storage capacity quantity model to determine the granary storage quantity; the model of the quantity of stored grains in the granary is as follows:

H＝b_HQ_BNF(s) or

The invention is characterized in that the relationship between the height of the grain pile and the grain storage quantity of the granary is introduced to construct a granary grain storage quantity model, which comprises the primary relationship and the secondary relationship between the height of the grain pile and the grain storage quantity of the granary, so that the robustness and the generalization capability of the granary grain storage quantity detection model are improved, the granary grain storage quantity detection model is suitable for structural types such as horizontal warehouses, is convenient for the remote online detection of the grain storage quantity and the like, and can meet the requirement of the remote online detection of the grain storage quantity of the granary such as the horizontal warehouses.

The invention also provides a granary reserve detection device based on the bottom surface pressure statistic and the reserve equation, which comprises a memory and a processor, wherein the processor is used for executing instructions stored in the memory to realize the introduced granary reserve detection method based on the bottom surface two-circle pressure statistic or realize the introduced granary reserve detection method based on the bottom surface single-circle pressure statistic.

Drawings

FIG. 1 is a schematic illustration of a horizontal warehouse floor pressure sensor (two-turn) arrangement of the present invention;

FIG. 2 is a schematic diagram of the cartridge floor pressure sensor (two-turn) arrangement of the present invention;

FIG. 3 is a graph of the results of the error in the calculation of the grain weight of the granary modeled using all samples in example 1 of the two-pass method of the present invention;

FIG. 4 is a graph showing the results of the errors in the calculation of the grain weight of the grain bin when samples No. 25 to 30 are used as test samples in example 1 of the two-pass method of the present invention;

FIG. 5 is a graph of the results of the error in the calculation of the grain weight of the granary modeled after the two-pass method of example 1 of the present invention;

FIG. 6 is the error in the calculation of the grain weights of all the samples in example 1 of the two-pass method of the present invention;

FIG. 7 is a flow chart of the granary reserve detection method based on the quadratic equation of the pressure statistics for two circles on the bottom surface of example 1 of the two-circle method of the present invention;

FIG. 8 is a schematic view of a horizontal warehouse floor pressure sensor (single loop) arrangement of the present invention;

FIG. 9 is a schematic diagram of the cartridge floor pressure sensor (single turn) arrangement of the present invention;

FIG. 10 is a graph of the results of the error in the calculation of grain weight for a granary modeled using all samples according to example 1 of the single-turn method of the present invention;

FIG. 11 is a graph of the results of the errors in the calculation of the grain weight of the grain bin when samples No. 19 to 24 of example 1 of the single-turn method of the present invention are used as test samples;

FIG. 12 is a graph of the results of the error in the calculation of the grain weight of the granary modeled after the single-turn method of example 1 of the present invention;

FIG. 13 is a graph of the results of the error in the calculation of the grain weights of the granary for all samples of example 1 of the single-pass method of the present invention;

FIG. 14 is a flow chart of the method of detecting the reserves of a granary based on the equation of once for the offset statistics of the pressure at the single circle on the bottom surface of embodiment 1 of the single circle method of the present invention;

fig. 15 is a block diagram of the grain bin reserves detection apparatus of the present invention based on floor pressure statistics and reserve equations.

Detailed Description

The method is mainly characterized in that the relationship between the grain bulk height and the grain storage quantity of the granary is introduced to construct a granary grain storage quantity model, and the robustness and generalization capability of the granary grain storage quantity detection model are improved. After a grain quantity model of grain delivery and storage is constructed, the model can be used for detecting the grain quantity of the grain storage.

Two-turn method embodiment 1-two-turn method embodiment 4 is granary reserve detection in which two turns of pressure sensors are provided on the bottom surface of a granary, and single-turn method embodiment 1 and single-turn method embodiment 2 are granary reserve detection in which a single-turn pressure sensor is provided on the bottom surface of a granary. It should be noted that the formula numbers in the respective embodiments only refer to the formulas in the corresponding embodiments, unless otherwise specifically explained.

Two-cycle method example 1:

according to the pressure distribution characteristics of the granary, the granary reserve detection model based on the quadratic equation of the two circles of pressure statistics on the bottom surface is provided. The core technology comprises a quadratic relation model of grain storage quantity of the granary and pressure on the bottom surface and the side surface of the granary, an average value and standard deviation calculation method of inner and outer circle pressure sensors based on the skewed distribution characteristic, and a granary storage quantity detection model based on a quadratic equation of two circle pressure statistics on the bottom surface. As described in detail below.

1. Quadratic relation equation of grain storage quantity and pressure of granary

The grain warehouse is of a horizontal warehouse, a silo and the like, after grains are put into the warehouse, the top of a grain pile is required to be flattened, the shape of the horizontal warehouse grain pile is approximately a cube with different sizes, and the shape of the silo grain pile is approximately a cylinder with different sizes. Without loss of generality, the method can be deduced through grain pile stress analysis, and the grain storage quantity of the granary is as follows:

W＝A_BQ_BNF(s) (1)

wherein W is the grain storage quantity/the grain storage weight of the granary; a. the_BThe area of the bottom surface of the granary contacted with the grain pile; q_BNF(s) is the equivalent average pressure of the bottom surface of the granary in contact with the grain pile, and is shown in formula (2):

wherein s is the set of contact points of the surface of the grain pile and the surface of the granary; k_cAs a geometric parameter of the grain heap, K_c＝C_B/A_B，C_BThe perimeter of the bottom surface of the grain pile; h is the grain pile height; f. of_FIs the average coefficient of friction between the side of the grain bulk and the side of the grain bin, f for a given grain bin and grain type_FIs a constant;

is the pressure intensity average value of the bottom surface of the grain pile,

the average value of the pressure at the side surface of the grain pile is shown as the following formulas (3) to (4):

wherein n is_B、n_FThe number of pressure measurement points on the bottom surface and the side surface of the grain pile is n_B→∝、n_F→∝；Q_B(s_i)、Q_F(s_j) For measuring pressure intensity on the bottom surface of grain pile_iSide pressure measurement point s_jThe pressure value of (2).

Experiments and theoretical analysis show that for a horizontal warehouse and a silo in practical application, under the condition of the normal grain loading height, the grain pile height and the grain storage quantity of the granary have a quadratic relationship shown as the following formula:

wherein, b_H1、b_H2Are coefficients. When formula (5) is substituted for formula (2), there are:

wherein, K_FIs the friction force action coefficient of the unit area of the side surface of the granary,

the formula (6) is a quadratic relation equation between the grain storage quantity of the granary and the pressure of the bottom surface and the side surface of the granary. It describes the average value of the grain storage quantity W of the granary and the pressure intensity of the bottom surface of the grain pile under a certain grain loading height

Mean lateral pressure

The theoretical relationship of (1). Solving the quadratic equation shown in equation (6) has:

the substitution formula (1) comprises:

the formula (6) is a quadratic relation model of the grain storage quantity of the granary and the pressure intensity of the bottom surface and the side surface of the granary. The formula (8) is a granary reserve detection theoretical model based on a quadratic equation of the pressure statistics of two circles at the bottom surface.

2. Sensor arrangement model

To effectively reduce

And

according to the detection cost, aiming at the pressure distribution characteristics of the grain pile of the grain bin, for the commonly used horizontal warehouse and silo, the pressure sensors are arranged on the bottom surface of the grain bin according to an outer ring and an inner ring, as shown in fig. 1 and 2, the rings are the arrangement positions of the pressure sensors, the distances between the outer ring pressure sensors and the side wall are D, and the distances between the inner ring pressure sensors and the side wall are D. D is more than 0 m and less than 1m, D is more than 2m, and about 3 m is generally selected. In order to ensure the universality of the detection model, the distances D and D between the inner and outer ring pressure sensors of each granary and the side wall are the same. The number of the two circles of sensors is 6-10, and the distance between the sensors is not less than 1 m.

3. Inner circle mean and standard deviation calculation

Practical test results show that for the arrangement of the two rings of sensors on the bottom surface of the granary shown in the figures 1 and 2, due to the limited mobility of grains, the output values of the inner and outer ring pressure sensors have remarkable volatility and randomness. Repeated experiment results show that the output values of the inner and outer ring pressure sensors obviously have the following characteristics: 1) the output values of the inner and outer ring pressure sensors obviously have the characteristic of off-normal distribution; 2) the fluctuation and randomness of the sensor output values are relatively small in the vicinity of the median, and the fluctuation and randomness of the output values are relatively large in the regions of small and large values. Based on the characteristics, the invention provides an inner ring sensor selection method based on the skewed distribution characteristics and a corresponding inner and outer ring sensor output value mean value and standard deviation calculation method.

For the arrangement of two circles of sensors on the bottom surface of the granary shown in the figures 1 and 2 and the measured values of any group of inner circle pressure sensors, an inner circle sensor output value sequence Q is constructed according to the sorting of the sensor output values_B(s_Inner). Calculating the mean value of the median neighboring points according to the formulas (9) and (10)

Standard deviation SD from inner ring sensor output value_Med(s_Inner)：

Wherein Q is_B(s_Inner(i) Is a sequence of inner ring sensor output values Q_B(s_Inner) I-1, 2, N_I，N_IThe number of the sensors is the inner ring; i.e. i_MIs the sequence number of the median point; n is a radical of_MFor the adjacent points on the left and right sides of the median point, generally take N_M＝2-3。

Construction of an inner ring sensor output value mean value calculation sequence Q according to formula (11)_BS(s_Inner)：

Wherein, C_ISkThe skewing distribution coefficient of the output value of the inner ring sensor; t is_SDThe threshold coefficient is removed for the inner circle sensor points. The main innovation point of the rule is that the inner ring sensor output value skewed distribution coefficient C is introduced_ISkAnd the rationality of selecting the output value points of the inner ring sensor is improved. Calculating the average value of the output values of the inner ring sensor according to the formula (12)

Wherein N is_ISAnd the number of the sequence data of the output value of the inner ring sensor after removal. Equation (10) is an inner ring sensor output value standard deviation calculation equation, and equation (12) is an inner ring sensor output value mean value calculation equation.

Similarly, for the arrangement of two circles of sensors on the bottom surface of the granary shown in fig. 1 and 2 and the measured value of any group of outer ring pressure sensors, the measured values are arranged according to the output values of the sensorsSequence construction of outer ring sensor output value sequence Q_B(s_Outer). Calculating the mean value of the median neighboring points according to the formulas (13) and (14)

Standard deviation SD from outer ring sensor output value_Med(s_Outer)。

Wherein Q is_B(s_Outer(i) Is a sequence of outer ring sensor output values Q_B(s_Outer) I-1, 2, N_O，N_OThe number of the sensors is the inner ring; i.e. i_MIs the sequence number of the median point; n is a radical of_MThe adjacent points on the left and right sides of the median point are counted.

Construction of an outer ring sensor output value mean value calculation sequence Q according to formula (15)_BS(s_Outer)。

Wherein, C_OSkIs the point-off-state distribution coefficient of the outer ring sensor, C_TSDRemoving the threshold coefficient, T, for the outer ring sensor points_SDThe threshold coefficient is removed for the inner circle sensor points. Calculating the average value of the output values of the outer ring sensor according to the formula (16)

Wherein N is_OSAnd removing the number of the sequence data of the output values of the outer circle sensor. Equation (14) is a calculation equation of the standard deviation of the output value of the outer ring sensor, and equation (16) is the output value of the inner ring sensorAnd (5) a mean value calculation formula.

4. Granary stored grain quantity detection model

For the granary bottom surface two-circle sensor arrangement model shown in figures 1 and 2, the pressure mean value of the granary bottom surface is constructed

The estimation of (d) is:

wherein, a_B(m) is

Estimate coefficients of the term, m 1_B，N_BIs composed of

Estimating the order of the polynomial;

the average value of the output values of the two circles of pressure sensors on the bottom surface is shown as the following formula:

wherein,

the average value of the output values of the outer ring sensors is obtained;

and the average value of the output values of the inner ring sensors is obtained.

Structural side pressure mean

The estimation of (d) is:

wherein, b_F(n) is

Estimate coefficients of the term, N1_F,N_FIs composed of

Estimating the order of the polynomial; i is_D(s) is the mean lateral pressure

The estimated amount of (a) is shown as follows:

wherein, I_D(s) is the mean lateral pressure

An estimate of (a); SD_Med(s_Inner)、SD_Med(s_Outer) Respectively is the standard deviation of the output values of the inner ring pressure sensor and the outer ring pressure sensor; k_SDIs the coefficient of the difference term of the variance.

Substituting the formula (17) and the formula (19) into the formula (6), coefficient b_H1、b_H2Average friction coefficient f between the side of the grain bulk and the side of the grain bin_FAnd b_F(n) combined, then:

wherein, a_F1(n)、a_F2And (n) is the coefficient of the combined polynomial term.

The actual modeling results show that Q in the formula (21) is used in some cases_BNF(s) first and second terms, taking different I_DSince the(s) order contributes to improvement of the model accuracy, the correction equation (21) is expressed by the following equation:

wherein N is_F1、N_F2Are respectively Q_BNF(s) I of the first and second order terms_D(s) order, N_F1≥N_F2. Solving the quadratic equation shown in equation (22) has:

wherein,

formula (22) is the average value of the grain storage quantity of the granary and the output values of the inner and outer ring pressure sensors

Mean lateral pressure

Is estimated by_D(s) quadratic equation giving an estimate of the quantity of grain stored in the barn

And the average value of the output values of the inner and outer ring pressure sensors

Side pressure mean value estimator I_D(s) polynomial relational description. And the formula (23) is a granary reserve detection model based on a quadratic equation of the pressure statistics of two circles of the bottom surface.

5. Modeling method

Selecting and constructing modeling sample set S from experimental data_Λ：

Wherein, k is the sample point number, and k is 1,2,3,..., M is the number of samples;

sequence of sensor output values for the k-th sample point, i ═ 1,2_I，N_IThe number of the sensors is the inner ring;

the sequence of outer sensor outputs for the k-th sample point, j 1,2_O，N_OThe number of the outer ring sensors is; w^kIs the actual grain feed weight at sample point k,

is the corresponding area contacting with the bottom surface of the granary. According to the sample set S_ΛA modeling sample set S is constructed by the formula (1), the formula (16) and the formula (18)：

Wherein,

average value of output values of inner and outer ring pressure sensors of the kth sample point respectively

Side pressure mean value estimator I_D(s) and equivalent mean pressure Q of the bottom of the granary_BNF(s) is selected. Sample set SIs divided into multiple regression sample set S_MAnd a test sample set S_T。

For a given set of modeling samples S_MAnd test sample set S_TAs can be seen from the equations (22) and (23), the granary reserve detection model based on the quadratic equation of the pressure statistics of the two circles around the bottom surface shown in the equation (23) comprises

Maximum order of a term N_B、I_DMaximum of(s)Large number of orders N_F1And N_F2、I_DTerm(s) parameter K_SDInner and outer circle sensor point removal threshold coefficient T_SDAnd C_TSDInner and outer ring sensor point bias distribution coefficient C_ISkAnd C_OSkAnd the number N of adjacent points on the left and right sides of the median point of the sensor output value sequence of the median adjacent points_MAnd polynomial term coefficient a_B(m)、a_F1(n₁)、a_F2(n₂) And the like. Order:

C_R＝(N_B,N_F1,N_F2,K_SD,T_SD,C_TSD,C_ISk,C_OSk,N_M) (26)

wherein, C_RIs a parameter set. As can be seen from equation (22), given the parameter set C_RThe modeling problem of the model shown in the formula (23) can be simplified to a zero-crossing point multiple linear regression problem shown in the following formula:

regression independent variable

And

total N_B+N_F1+N_F2The modeling problem of the model expressed by equation (22) can be converted into an optimization problem expressed by the following equation:

modeling can be achieved by a method combining optimization and multiple regression.

6. Examples of detection

The flow chart of the granary reserve detection method based on the quadratic equation of the pressure statistics of the two circles on the bottom surface is shown in fig. 7, and the method is applied to specific examples to illustrate the effectiveness of the granary reserve detection method.

1) Test example 1

The length of the horizontal warehouse adopted by the experiment is 9m, the width is 4.2m, and the area is 37.8m²，C_B/A_B0.698. The granaries all belong to small-sized granaries C_B/A_BIs relatively large. According to the pressure sensor arrangement model shown in fig. 1, the pressure sensors are arranged in 2 circles, 6 pressure sensors are arranged in the inner circle, and 16 pressure sensors are arranged in the outer circle, so that 22 pressure sensors are arranged. The height of the wheat grain pile is about 6 meters, data is taken every 1 meter when the grains are fed, and 5 times of experiments are repeated to obtain 30 samples.

For the granary stored grain quantity detection model based on the quadratic equation of the bottom surface two-circle pressure statistic shown in the formula (23), all 30 samples are used as modeling samples. The optimized modeling parameters are shown in Table 1-1, and the obtained parameters are shown in Table 1-2. The error of the calculation of the grain weight in the granary is shown in figure 3, and the maximum percentage error is 1.8%.

For the granary stored grain quantity detection model based on the quadratic equation of the pressure statistics of the two circles on the bottom surface shown in the formula (23), samples 25 to 30 of the experiment 5 are used as test samples, samples 13 to 18 of the experiment 3 are used as parameter optimization samples, and the rest 18 samples are used as modeling samples. The optimized modeling parameters are shown in tables 1-3, and the obtained parameters are shown in tables 1-4. The calculation errors of the grain storage weight of the granary are shown in figure 4, and the prediction errors are all less than 1.9%. Since the maximum test error is large due to too few modeling samples, the prediction error can be further reduced if the number of modeling samples is increased.

2) Detection example 2

For 4 rice barns in the Tongzhou grain depot and 2 surging rice barns, the stored grain weights are 6450 tons, 4420 tons, 3215 tons, 64500 tons, 2455.6 tons and 2099.9 tons respectively. 1231 samples were selected from the long-term test data. 922 samples are selected as modeling samples (308 samples are selected as item maximum order selection samples), and the others are selected as test samples. For the granary stored grain quantity detection model based on the quadratic equation of the pressure statistics of two circles on the bottom surface shown in the formula (23), the optimized modeling parameters are shown in tables 1-5, and the obtained parameters are shown in tables 1-6. The calculated error of the grain weights of the modeled samples is shown in fig. 5, and the calculated error of the grain weights of all samples is shown in fig. 6. From these results, it can be seen that the errors in the calculation of the grain weights of the granary stores for both the modeled samples and the tested samples were less than 0.2%.

The granary storage capacity detection model of the quadratic equation of the two circles of pressure statistics on the bottom surface is mainly characterized in that the quadratic relation between the grain pile height and the grain storage quantity of the granary is directly introduced, the robustness and the generalization capability of the granary storage quantity detection model are improved, the granary storage capacity detection model is suitable for structural types such as horizontal warehouses, is convenient for remote online granary quantity detection, and can meet the requirement of remote online grain storage quantity detection of the horizontal warehouses and the like. Further, based on the skewness distribution characteristic expression of the output values of the inner and outer ring pressure sensors, the inner ring sensor output value average value calculation sequence Q shown in expression (11) is proposed_BS(s_Inner) The construction rule of (1) and the outer ring sensor output value mean value calculation sequence Q shown in formula (15)_BS(s_Outer) And a rule is constructed, the skewed distribution coefficient of the output values of the inner and outer ring sensors is introduced, and the rationality of selecting the output value points of the inner ring sensor is improved.

Two-cycle method example 2:

according to the pressure distribution characteristics of the granary, the granary reserve volume detection model based on the linear equation of the two-circle pressure statistic on the bottom surface is provided. The core technology comprises a primary relation model of grain storage quantity of the granary and pressure on the bottom surface and the side surface of the granary, a granary storage quantity detection model based on a primary equation of pressure statistics of two circles on the bottom surface, and a modeling algorithm of a granary storage quantity detection model based on a primary equation of pressure statistics of two circles on the bottom surface.

1. Equation of one-time relation between grain storage quantity and pressure of granary

In this embodiment, when a relational equation between the grain storage quantity of the grain bin and the pressure is constructed, the difference from the two-turn method embodiment 1 is only that under the condition of the normal grain loading height, the grain bulk height and the grain storage quantity of the grain bin have a strong linear relationship, namely:

H≈b_H1Q_BNF(s) (1)

wherein, b_H1Is a scaling factor. When formula (7) is introduced into formula (2), formula (3):

the formula (3) is a linear relation equation between the grain storage quantity of the granary and the pressure intensity of the bottom surface and the side surface of the granary. It describes the average value of the grain storage quantity W of the granary and the pressure intensity of the bottom surface of the grain pile under a certain grain loading height

Mean lateral pressure

The theoretical relationship of (1).

2. Granary stored grain quantity detection model

The mean value of the output values of the two-circle sensor on the bottom surface is determined by adopting the same two-circle method and the formulas (17) to (20) in the embodiment 1

Mean value of pressure at bottom

Is estimated by

Mean lateral pressure

Is estimated by_D(s) and side pressure mean

Is estimated by

Substituting the formulas (17) and (20) in the embodiment 1 of the double-circle method into the formula (3) in the embodiment, and combining the average friction coefficient f between the side surface of the grain pile and the side surface of the granary_FCoefficient b_H1And

coefficient b of the estimated term_F(n) then:

wherein, a_FAnd (n) is the coefficient of the combined estimation term. An arrangement (15) having:

formula (4) is the average value of the grain storage quantity of the granary and the output values of the inner and outer ring pressure sensors

Mean lateral pressure

Is estimated by_D(s) of the equation which gives an estimate of the quantity of grain stored in the barn

And average value of output values of two circles of pressure sensors

Mean lateral pressure

Is estimated by_D(s) polynomial relational description. And the formula (5) is a granary reserve detection model based on the linear equation of the pressure statistics of two circles on the bottom surface. The model has the main characteristics that the linear relation between the grain bulk height and the grain storage quantity of the granary is directly introduced, the accuracy and the effectiveness of grain bulk height estimation are improved, and therefore the robustness and the generalization capability of the grain storage quantity detection model are improved.

In this embodiment, the same sensor arrangement model as that in the two-turn method implementation 1 is adopted, and the modeling method is similar to that in the two-turn method implementation 1, and is not described here again. Meanwhile, the method of the embodiment is different from the methods of the embodiments 1 and 2 of the double-circle method only in models, and the steps of the method are basically the same, so that the embodiment does not provide specific examples and detection examples any more.

The method has the main characteristics that the linear relation between the grain bulk height and the grain storage quantity of the granary is directly introduced, the strong robustness and generalization capability of the model and the detection method are improved, the method is suitable for structural types of horizontal warehouses and the like, the remote online detection of the grain storage quantity of the granary is facilitated, and the requirement of the remote online detection of the grain storage quantity of the horizontal warehouses and the like can be met.

Two-cycle method example 3:

according to the pressure distribution characteristics of the granary, the granary reserve detection model based on the three-dimensional equation of the pressure statistic of two circles on the bottom surface is provided. The core technology comprises a granary storage capacity detection model based on a bottom surface two-circle pressure intensity statistic cubic equation and a granary storage capacity detection model modeling method based on the bottom surface two-circle pressure intensity statistic cubic equation. As described in detail below.

1. Cubic relation equation of grain storage quantity and pressure of granary

In this embodiment, when a relation equation of the grain storage quantity of the granary and the pressure is constructed, the difference from the two-circle method embodiment 1 is only that in the case of the normal grain loading height, the grain bulk height and the grain storage quantity of the granary have a cubic relation shown in the following formula:

wherein, b_H1、b_H2、b_H3Are coefficients. When formula (1) is substituted for formula (2), formula (3):

the pressure intensity mean value of the side surface of the grain pile is obtained; k_FIs the friction force action coefficient of the unit area of the side surface of the granary,

the formula (3) is a cubic relation equation between the grain storage quantity of the granary and the pressure intensity of the bottom surface and the side surface of the granary. It describes the average value of the grain storage quantity W of the granary and the pressure on the bottom surface of the grain pile under the condition of higher grain loading height

Mean lateral pressure

The theoretical relationship of (1).

2. Granary stored grain quantity detection model

Mean value of pressure at bottom

Is estimated by

Mean lateral pressure

Is estimated by_D(s) and side pressure mean

Is estimated by

Substituting the formulas (17) and (20) in the embodiment 1 of the double-circle method into the formula (3) in the embodiment, combining the average friction coefficients between the lateral surface of the grain pile and the lateral surface of the granary, and combining the average friction coefficient f between the lateral surface of the grain pile and the lateral surface of the granary_FCoefficient b_H1、b_H2、b_H3And b_F(n) of (a). Then there are:

wherein, a_B(m)、a_F1(n)、a_F2(n)、a_F3(N) is a coefficient of the estimation term, m 1_B，n＝1,...,N_F，N_B、N_FAre respectively as

I_DThe order of the(s) term.

The actual modeling results show that in some cases Q_BNF(s) first, second and third terms, taking different I_DSince the order of(s) is helpful in improving the model accuracy, the correction equation (4) is expressed as follows:

wherein N is_F1、N_F2、N_F3Are respectively Q_BNF(s) I of first, second and third terms_D(s) order. Solving the equation, then:

wherein,

formula (5) is the average value of the grain storage quantity of the granary and the output values of the inner and outer ring pressure sensors

Mean lateral pressure

Is estimated by_D(s) cubic equation giving the estimate of the grain stock quantity in a grain bin at a higher grain loading height

Side pressure mean value estimator I_D(s) polynomial relational description. And the formula (6) is a granary reserve detection model based on a bottom surface two-circle pressure statistic cubic equation provided by the invention.

The method has the main characteristics that the cubic relation between the height of the grain pile and the grain storage quantity of the granary is directly introduced, so that the robustness and the generalization capability of a grain storage quantity detection model of the granary are improved, the method is suitable for the structural types of silos and the like, the remote online detection of the grain storage quantity of the granary is facilitated, and the requirements of the remote online detection of the grain storage quantity of the silos and the like can be met.

Two-cycle method example 4:

according to the pressure distribution characteristics of the granary, the granary reserve detection model based on the quadratic equation of the two-circle pressure statistic of the bottom surface is provided. The core technology comprises a granary storage capacity detection model based on a bottom surface two-circle pressure intensity statistic quadratic equation and a granary storage capacity detection model modeling method based on the bottom surface two-circle pressure intensity statistic quadratic equation. As described in detail below.

1. Quartic relation equation of grain storage quantity and pressure of granary

In this example, when an equation of the relationship between the grain storage quantity of the grain bin and the pressure is constructed, the difference from the two-turn method example 1 is only that in the case of the normal grain loading height, the grain bulk height and the grain storage quantity of the grain bin have a quartic relationship as shown in the following formula:

wherein, b_H1、b_H2、b_H3、b_H4Are coefficients. When formula (1) is substituted for formula (2), formula (3):

the formula (3) is a quartic relation equation between the grain storage quantity of the granary and the pressure intensity of the bottom surface and the side surface of the granary. It describes the average value of the grain storage quantity W of the granary and the pressure of the bottom surface of the grain pile under the condition of large grain loading height

Mean lateral pressure

The theoretical relationship of (1).

2. Granary stored grain quantity detection model

Mean value of pressure at bottom

Is estimated by

Mean lateral pressure

Is estimated by_D(s) and side pressure mean

Is estimated by

The equations (17) and (20) in the example 1 of the double-turn method are substituted into the equation (3) in the present example, and the coefficient b_H1、b_H2、b_H3、b_H4Average friction coefficient f between the side of the grain bulk and the side of the grain bin_FAnd b_F(n) combining. Then there are:

wherein, a_B(m)、a_F1(n)、a_F2(n)、a_F3(n)、a_F4(N) is a coefficient of the estimation term, m 1_B，n＝1,...,N_F，N_B、N_FAre respectively as

I_DThe order of the(s) term.

The actual modeling results show that in some cases Q_BNF(s) first, second, third and fourth terms, not to be takenSame as I_DSince the order of(s) is helpful in improving the model accuracy, the correction equation (4) is expressed as follows:

wherein,

Mean lateral pressure

Is estimated by_D(s) of the quartic equation giving an estimate of the grain reserve in a grain bin at a higher grain fill height

Side pressure mean value estimator I_D(s) polynomial relational description. Formula (5) the invention provides a granary reserve detection model based on a quadratic equation of pressure statistics of two circles at the bottom surface.

In this embodiment, the same sensor arrangement model as that in the two-turn method embodiment 1 is implemented, and the modeling method is similar to that in the two-turn method embodiment 1 and the two-turn method embodiment 2, and is not described here again. Meanwhile, the method of the embodiment is different from the methods of the embodiments 1 and 2 of the double-circle method only in models, and the steps of the method are basically the same, so that the embodiment does not provide specific examples and detection examples any more.

The method has the main characteristics that the quartic relation between the grain bulk height and the grain storage quantity of the granary is directly introduced, the robustness and the generalization capability of a grain storage quantity detection model of the granary are improved, the method is suitable for the structural types of silos and the like, the remote online detection of the grain storage quantity of the granary is facilitated, and the like, and the requirements of the remote online detection of the grain storage quantity of the granary such as the silos and the like can be met.

Single-loop method example 1:

according to the pressure distribution characteristics of the granary, the granary reserve detection model based on the linear equation of single-circle pressure state statistics of the bottom surface is provided. The core technology of the invention comprises two parts, namely single-circle sensor magnitude value sequence division and a mean value and standard deviation calculation method thereof, and a granary reserve detection model based on a bottom surface single-circle pressure intensity state partial statistics linear equation. As described in detail below.

1. Sensor arrangement model

For the commonly used horizontal silos and silos, a single-turn pressure sensor group as shown in fig. 8 and 9 is arranged on the bottom surface of the granary, and the pressure sensor 1 is arranged at the position of the dotted line in the figure. Under the condition of ensuring convenient grain loading and unloading, the distance d between each pressure sensor 1 and the side wall can be 1-2 meters generally. In order to ensure the universality of the detection model, the distance d between each pressure sensor of each granary and the side wall is the same. The number of the pressure sensors of each granary is 10-15, and the distance between every two adjacent pressure sensors is larger than 1 m.

2. Screening of output values of single-ring sensor, division of large and small values and calculation of mean value and standard deviation of output values

Practical test results show that for the granary bottom surface single-loop sensor arrangement models shown in fig. 8 and 9, due to the limited mobility of grains, the output value of the single-loop pressure sensor has remarkable volatility and randomness. Repeated experiment results show that the output value of the single-ring pressure sensor obviously has the following characteristics: 1) the output value of the single-ring pressure sensor obviously has the characteristic of off-normal distribution; 2) the output value fluctuation and randomness of the single-turn pressure sensor are relatively small in the area around the median, and the output value fluctuation and randomness are relatively large in the area with smaller and larger values. Based on the characteristics, the invention provides a filtering selection and large and small value division method of the output value of the single-ring pressure sensor based on the skewed distribution characteristics and a calculation method of the average value and the standard deviation of the output value of the single sensor.

For the arrangement of the single-circle pressure sensors on the bottom surface of the granary and the output values of the single-circle pressure sensors shown in the figures 8 and 9, the single-circle sensor output value sequence Q(s) is constructed by sequencing according to the magnitude of the output values of the sensors. Calculating the mean value of the median neighboring points of the output value sequence Q(s) according to the formula (1) and the formula (2)

Standard deviation SD from output value sequence Q(s)_Med(s)：

Wherein, Q (s (i)) is the i-th element value of the single-loop pressure sensor output value sequence Q(s), i is 1,2_S，N_SIs a sensorThe number of the cells; i.e. i_MIs the sequence number of the median point; n is a radical of_MFor the number of output values adjacent to the left and right of the middle point of the preset output value sequence Q(s), N is generally selected_M＝2-3。

For the output value sequence Q(s) of the single-ring pressure sensor, removing value points with large fluctuation quantity according to the formula (3) to construct an effective output value sequence Q of the single-ring pressure sensor_SE(s)。

Wherein, C_SkIs the off-state distribution coefficient of the output value of the single-turn sensor, T_SDThe threshold coefficient is removed for the sensor points of the single-turn pressure sensor arrangement. The invention provides a method for selecting an effective output value sequence of a single-turn sensor in a formula (3), which has the main innovation point that a sensor output value skewed distribution coefficient C is introduced according to skewed characteristics of the sensor output value on the bottom surface of a granary_SkThe rationality of the selection of the output value of the sensor is improved.

To construct an estimate of the lateral pressure of a grain pile, a sequence of effective output values Q of a single-turn sensor is constructed_SE(s) sequence of small values Q divided into single-turn sensors_SS(s) and large value sequence Q_SL(s)。

Wherein, C_MVAnd dividing an adjusting coefficient for the single-circle sensor output value sequence. The formula (4) and the formula (5) are small value sequences Q of the single-turn sensor provided by the invention_SS(s) and large value sequence Q_SLThe(s) division method is mainly characterized in that a single-turn sensor output value sequence division adjustment coefficient C is introduced according to the skewed characteristic of the output value of the sensor on the bottom surface of the granary_MVThe rationality of large-value and small-value division is improved.

Sequence of small values Q of a single-turn sensor_SSMean value of(s)

And standard deviation SD_SS(s) is:

wherein N is_SSSequence of small values Q for a single-turn sensor_SSNumber of data of(s).

Large value sequence Q of single-turn sensor_SLMean value of(s)

And standard deviation SD_SL(s) is:

wherein N is_SLSequence of large sensor outputs Q for a single-turn sensor_SLNumber of data of(s).

3. Granary stored grain quantity detection model

W＝A_BQ_BNF(s) (10)

wherein W is the quantity of stored grains in the granary or the weight of the stored grains in the granary; a. the_BThe area of the bottom surface of the granary contacted with the grain pile; q_BNF(s) is andequivalent average pressure, Q, at the bottom of a grain bin contacted with a grain pile_BNF(s) is represented by the formula (11).

Wherein s is the set of contact points of the surface of the grain pile and the surface of the granary; k_cAs a geometric parameter of the grain heap, K_c＝C_B/A_B，C_BThe circumference of the bottom surface of the granary contacted with the grain pile; h is the grain pile height; f. of_FIs the average friction coefficient between the side of the grain bulk and the side of the grain bin. For a given grain bin and grain type, f_FIs a constant.

The pressure mean values of the bottom surface and the side surface of the grain pile are respectively shown as formulas (12) to (13).

Wherein n is_B、n_FThe number of pressure measurement points on the bottom surface and the side surface of the grain pile is n_B→∝、n_F→∝；Q_B(s_i)、Q_F(s_j) Respectively as pressure intensity measuring points s on the bottom surface of the grain pile_iSide pressure measurement point s_jThe pressure value of (2).

Experiments and theoretical analysis show that for a horizontal warehouse and a silo in practical application, the grain pile height and the grain storage quantity of the granary have a strong linear relationship, namely:

H≈b_HQ_BNF(s) (14)

wherein, b_HIs a scaling factor. When formula (14) is substituted for formula (11), there are:

for the granary bottom surface single-ring pressure sensor arrangement model shown in fig. 8 and 9, the average value of the output values of the single-ring pressure sensors

Comprises the following steps:

wherein,

respectively a small value sequence mean value and a large value sequence mean value of the single-turn sensor.

Obviously, the pressure intensity mean value of the bottom surface of the grain pile

The actual detection result shows that the detection result shows that,

has strong linear relation with the grain storage quantity of the granary, and has strong detection repeatability and low detection cost. By using

Polynomial construction of bottom pressure mean value of grain pile

The estimation of (d) is:

wherein, a_B(m) is

Estimate coefficients of the term, m 1_B，N_BIs the estimated polynomial order.

The pressure distribution at the side surface of the grain pile has obvious non-uniformity and randomness, so that the pressure at the side surfaceMean value

More pressure sensors are required for detection. In addition, theoretically, for the granary bottom surface single-circle sensor arrangement model shown in fig. 8 and 9, the changes of the mean value and the standard deviation of the small-value sequence and the large-value sequence of the single-circle sensor are inevitably caused due to the side pressure effect,

the increase will increase the difference between the mean and standard deviation of the small value sequence and the large value sequence. Therefore, the mean value, the standard deviation and the difference of the standard deviation of the small-value sequence and the large-value sequence of the single-circle sensor can be embodied

Can be used to construct a side pressure mean using these statistics

Is estimated. Order:

wherein, I_DS(s) is the mean lateral pressure

An estimate of (a); SD_SS(s)、SD_SL(s) are respectively the standard deviation of the output values of the inner ring pressure sensor and the outer ring pressure sensor; k_SDIs the coefficient of the difference term of the variance. By the use of I_D(s) polynomial construction of side pressure mean

The estimation of (d) is:

wherein, b_F(n) is

Estimate coefficients of the term, N1_F，N_FIs composed of

The order of the polynomial is estimated.

Formula (17) or formula (19) is substituted for formula (15), and coefficient b is set_HAverage coefficient of friction f between the side of the grain bulk and the side of the grain bin_FAnd b_F(n) combining. Then there are:

wherein, a_FAnd (n) is the coefficient of the combined estimation term. Substituted into formula (10) and finished with:

formula (21) is the average value of the grain storage quantity of the granary and the output value of the single-circle pressure sensor

Mean lateral pressure

Is estimated by_DS(s) a first order equation giving an estimate of the quantity of grain stored in the barn

And the average value of the output values of the single-ring pressure sensor

Side pressure mean value estimator I_DS(s) polynomial relational description. The formula (21) is the granary reserve detection model based on the bottom surface single-turn pressure off-normal statistic linear equation provided by the invention. The model is mainly characterized in that the height of the grain pile is directly introduced into the modelThe linear relation of the grain storage quantity of the granary does not need to estimate the height of the grain pile based on the output value of the pressure sensor, so that the calculation amount is reduced, the accuracy and the effectiveness of the estimation of the height of the grain pile are improved, and the robustness and the generalization capability of the grain storage quantity detection model of the granary are improved.

In this embodiment, the modeling method is similar to that in the dual-turn method embodiment 1, and details are not repeated here.

4. Specific examples

The granary reserve detection method based on the bottom surface single-turn pressure off-state statistics linear equation is shown in FIG. 14 and comprises the following specific steps:

step one, system configuration. And selecting a specific pressure sensor, and configuring corresponding systems for data acquisition, data transmission and the like.

And step two, installing a bottom surface pressure sensor. The arrangement of the sensors of the horizontal warehouse is shown in fig. 8, the arrangement of the bottom pressure sensors of the silo is shown in fig. 9, the sensors are arranged in a single circle, and the average distance D between the sensors and the side wall is more than 1 meter. The number of the single sensors is 6-10, and the distance between the sensors is not less than 1 m.

Step three, for given sensors, grain types and bin types, if the system is not calibrated, arranging pressure sensors in more than 6 grain bins, feeding grains to full bins, collecting the output values of the pressure sensors in all the bins after the output values of the pressure sensors are stable, and constructing a modeling sample set

Wherein k is a sample point number, k is 1,2,3, M, and M is the number of samples;

a sequence of single-turn sensor output values for the kth sample point, i ═ 1,2_S，N_SThe number of the single-ring pressure sensors on the bottom surface of the granary is arranged; w^kIs the actual grain feed weight at sample point k,

is the corresponding area of the bottom surface of the granary. According to sample sets and corresponding granariesSubstituting the full grain storage quantity into a model represented by a formula (21) to calculate a constant term, and finishing the calibration of the model.

Step four, after obtaining the calibrated model, acquiring the output value sequence of the single-circle sensor on the bottom surface, screening and filtering the output values and establishing a magnitude value sequence according to the method of the invention, and determining the mean value of the magnitude value sequence of the single-circle pressure sensor

Mean of large value series

Small value sequence standard deviation SD_SS(s) and Large value sequence Standard deviation

Further determining the average value of the output values of the single-turn pressure sensor

Determining the side pressure mean value according to equation (18)

Is estimated by_DS(s) subjecting

And I_DS(s) substituting the model shown in the formula (21) to obtain the grain storage quantity of the granary.

5. Examples of detection

1) Test example 1

The length of the horizontal warehouse adopted by the experiment is 9m, the width is 4.2m, and the area is 37.8m²，C_B/A_B0.698. The granaries all belong to small-sized granaries C_B/A_BIs relatively large. According to the pressure sensor arrangement model shown in fig. 8, the pressure sensors are arranged in a single turn for 16 pressure sensors. The height of the wheat grain pile is about 6 meters, data is taken every 1 meter when the grains are fed, and 5 times of experiments are repeated to obtain 30 samples.

For the granary reserve detection model based on the bottom surface single-turn pressure off-state statistics linear equation shown in formula (21), all 30 samples are used as modeling samples. The optimized modeling parameters are shown in Table 5-1, and the obtained parameters are shown in Table 5-2. The calculated error of the grain weight in the granary is shown in fig. 10, and the maximum percentage error of the built model is 1.6%.

For the granary reserves detection model based on the bottom surface single-turn pressure state deviation statistic linear equation shown in formula (21), samples 19 to 24 of experiment 4 are used as test samples, samples 7 to 12 of experiment 4 are used as parameter optimization samples, and the rest 18 samples are used as modeling samples. The optimized modeling parameters are shown in Table 5-3, and the obtained parameters are shown in Table 5-4. The calculation error of the grain weight in the granary is shown in fig. 11, and the prediction error of all samples is less than 1.1%. Because the modeling samples are too few, the maximum testing error is larger, and if the number of the modeling samples is increased, the prediction error can be further reduced.

2) Detection example 2

For 4 rice barns in the Tongzhou grain depot and 2 surging rice barns, the stored grain weights are 6450 tons, 4420 tons, 3215 tons, 64500 tons, 2455.6 tons and 2099.9 tons respectively. 1231 samples were selected from the long-term test data. 922 samples are selected as modeling samples (308 samples are selected as item maximum order selection samples), and the others are selected as test samples. For the granary reserve detection model based on the equation of once for the single-turn pressure off-state statistics of the bottom surface shown in formula (21), the optimized modeling parameters are shown in tables 5-5, and the obtained parameters are shown in tables 5-6. The calculated error of the grain weights of the modeled samples is shown in fig. 12, and the calculated error of the grain weights of all the samples is shown in fig. 13. From these results, it can be seen that the errors in the calculation of the grain weights of the granary stores for both the modeled samples and the tested samples were less than 1.6%.

The invention directly introduces the quadratic relation between the grain bulk height and the grain storage quantity of the granary to construct a granary storage quantity detection model of single-circle pressure intensity skewed state statistic on the bottom surface, and improves the robustness and generalization capability of the granary storage quantity detection model.

Single-loop method example 2:

according to the pressure distribution characteristics of the granary, the granary reserve detection model based on the quadratic equation of the single-circle pressure state statistics of the bottom surface is provided. The core technology comprises a granary reserve detection model based on a quadratic equation of single-circle pressure intensity state-of-deviation statistics of the bottom surface. As described in detail below.

1. Sensor arrangement model

In this example, the same sensor arrangement model as that of the single-turn example 1 was employed.

2. Screening of single-ring sensor output values, large and small value division and calculation of mean value and standard deviation of single-ring sensor output values

The sequence of magnitude values was divided in the same manner as in the single-turn example 1, and the mean and standard deviation of the sequence of magnitude values and the sequence of magnitude values were calculated, respectively.

As other embodiments, other prior art methods may be used to screen the output values of the single-turn sensor, or not to screen the output values of the single-turn sensor. Other prior art methods may also be employed to divide the single-turn sensor output values into large-value sequences and small-value sequences. For example, chinese patent publications No. CN110823343A, 0058 to 0067, describe a method for screening output values of a single-turn sensor on the bottom surface of a granary (sensor removal rule); [0069] [0071] discloses a method for dividing a large-value sequence and a small-value sequence of output values of a single-turn sensor on the bottom surface of a granary.

Finally obtaining a small-value sequence Q of the single-turn sensor_SSMean value of(s)

And standard deviation SD_SS(s) is:

Large value sequence Q of single-turn sensor_SLMean value of(s)

And standard deviation SD_SL(s) is:

3. Granary stored grain quantity detection model

W＝A_BQ_BNF(s) (10)

wherein,w is the quantity of stored grains in the granary or the weight of the stored grains in the granary; a. the_BThe area of the bottom surface of the granary contacted with the grain pile; q_BNF(s) is the equivalent average pressure of the bottom of the granary in contact with the grain pile, Q_BNF(s) is represented by the formula (11).

Experiments and theoretical analysis show that for a horizontal warehouse and a silo in practical application, under the condition of the normal grain loading height, the grain pile height and the grain storage quantity of the granary have a quadratic relationship shown as the following formula.

Wherein, b_H1、b_H2Are coefficients. By substituting formula (14) for formula (11), there are

the formula (15) is a quadratic relation equation between the grain storage quantity of the granary and the pressure of the bottom surface and the side surface of the granary. It describes the average value of the grain storage quantity W of the granary and the pressure intensity of the bottom surface of the grain pile under a certain grain loading height

Mean lateral pressure

The theoretical relationship of (1).

For the granary bottom surface single-circle sensor arrangement model shown in fig. 8 and 9, the average value of the output values of the single-circle pressure sensors

Comprises the following steps:

wherein,

The actual detection result shows that the detection result shows that,

Polynomial construction of bottom pressure mean value of grain pile

The estimation of (d) is:

wherein, a_B(m) is

The pressure distribution on the side surface of the grain pile has obvious non-uniformity and randomness, and the average value of the pressure on the side surface

Can be used to construct a side pressure mean using these statistics

Is estimated. Order:

wherein, I_D(s) is the mean lateral pressure

The estimation of (d) is:

wherein, b_F(n) is

Estimate coefficients of the term, N1_F，N_FIs composed of

The order of the polynomial is estimated.

Substituting the formula (17) and the formula (19) into the formula (15), coefficient b_H1、b_H2Average friction coefficient f between the side of the grain bulk and the side of the grain bin_FAnd b_F(n) combining. Then there are:

Practical modeling results show that in some cases Q is in equation (20)_BNF(s) first and second terms, taking different I_DSOrder of(s)In a case of a few cases, it is helpful to improve the model accuracy, and the correction formula (20) is expressed by the following formula.

Wherein N is_F1、N_F2Are respectively Q_BNF(s) I of the first and second order terms_DS(s) order, N_F1≥N_F2. Solving the quadratic equation shown in equation (21) includes:

wherein,

Mean lateral pressure

And the average value of the output values of the single-ring pressure sensor

Side pressure mean value estimator I_D(s) polynomial relational description. The formula (22) is a granary reserve detection model based on a quadratic equation of the bottom surface single-circle pressure statistic. The model has the main characteristics that the secondary relation between the grain bulk height and the grain storage quantity of the granary is directly introduced, and the robustness and the generalization capability of the granary grain storage quantity detection model are improved.

4. Modeling method

Selecting and constructing modeling sample set S from experimental data_Λ：

is the corresponding area contacting with the bottom surface of the granary.

According to the sample set S_ΛA modeling sample set S is constructed from the equations (10), (17) and (19)

Wherein,

mean value of output values of single-turn pressure sensors respectively at kth sample point

For a given set of modeling samples S_MAnd test sample set S_TAs can be seen from the equations (21) and (22), the granary reserve detection model based on the quadratic equation of the bottom surface single-circle pressure statistic shown in the equation (22) comprises

Maximum order of a term N_B、I_DSMaximum order N of(s) term_F1And N_F2，I_DSTerm(s) parameter K_SDAnd a sensor point removal threshold coefficient T for a single-turn pressure sensor arrangement_SDOff-state distribution coefficient C of sensor output value_SkOutput value sequence division adjusting coefficient C of single-loop sensor_MVThe number N of the left and right adjacent points of the adopted median point_MAnd polynomial term coefficient a_B(m) and a_F(n, m), etc. Order:

C_R＝(N_B,N_F1,N_F2,K_SD,T_SD,C_Sk,C_MV,N_M) (25)

wherein, C_RIs a parameter set. As can be seen from equation (21), if the parameter set C is given_RThe modeling problem of the model shown in the formula (21) can be simplified to a zero-crossing point multiple linear regression problem shown in the following formula.

The regression independent variables include

And

total N_B+N_F1+N_F2An item. Therefore, the modeling problem of the model shown in the formula (22) can be converted into an optimization problem shown in the following formula, and the modeling can be realized by a method combining optimization and multiple regression.

Wherein, C (T)_SD,K_SD,C_Sk) Model parameter penalty term.

The method of this embodiment is different from the method of embodiment 1 of the single-turn method only in the model, and the steps of the method are basically the same, so that the embodiment also does not provide a specific example and a detection example. The method of the embodiment is directed at the arrangement of the bottom surface single-circle sensors, the quadratic relation between the grain pile height and the grain storage quantity of the granary is directly introduced, the granary storage quantity detection model of the single-circle pressure intensity skewed statistic on the bottom surface is constructed, and the robustness and the generalization capability of the granary storage quantity detection model are improved.

The embodiment of the device is as follows:

an embodiment of the granary reserve detection device based on the floor pressure statistic and the reserve equation is shown in fig. 15 and comprises a memory, a processor and an internal bus, wherein the processor and the memory are communicated with each other through the internal bus.

The processor can be a microprocessor MCU, a programmable logic device FPGA and other processing devices. The memory can be various memories for storing information by using an electric energy mode, such as RAM, ROM and the like; various memories for storing information by magnetic energy, such as a hard disk, a floppy disk, a magnetic tape, a core memory, a bubble memory, a usb disk, etc.; various types of memory that store information optically, such as CDs, DVDs, etc., are used. Of course, other forms of memory are possible, such as quantum memory, graphene memory, and the like.

The processor may invoke logic instructions in the memory to implement a method for detecting a grain bin reserve based on the two-turn pressure statistic of the floor or a method for detecting a grain bin reserve based on the single-turn pressure statistic of the floor. In the two-turn method embodiment 1 to the two-turn method embodiment 4, a granary reserve amount detection method based on the bottom surface two-turn pressure statistic is described in detail, and in the single-turn method embodiment 1 to the single-turn method embodiment 2, a granary reserve amount detection method based on the bottom surface single-turn pressure statistic is described in detail.

Claims

1. A granary reserve detection method based on two circles of pressure intensity statistics on the bottom surface is characterized by comprising the following steps:

1) detecting the output values of two circles of pressure sensors arranged on the bottom surface of the granary and determining the average value of the output values of the inner circles of pressure sensors

Mean value of output values of outer ring pressure sensor

Standard deviation SD(s) of output value of inner ring pressure sensor_Inner) And standard deviation SD(s) of output value of outer ring pressure sensor_Outer)；

2) According to the mean value of the output values of the inner ring pressure sensor

And average value of output values of outer ring pressure sensor

Determining bottom surface pressure mean

3) According to the mean value of the output values of the inner ring pressure sensor

Mean value of output values of outer ring pressure sensor

Standard deviation SD(s) of output value of inner ring pressure sensor_Inner) And standard deviation SD(s) of output value of outer ring pressure sensor_Outer) Determining the mean lateral pressure

4) Averaging the pressure intensity of the bottom surface

And side pressure mean value

Substituting the model into the constructed grain storage quantity model to obtain the grain storage quantity W of the granary; the granary grain storage quantity model is as follows:

W＝A_BQ_BNF(s)

or 4

In the formula, A_BThe area of the bottom surface of the grain pile; q_BNF(s) is the equivalent average pressure of the bottom surface of the granary; h is the grain pile height; k_cAs a geometric parameter of the grain heap, K_c＝C_B/A_B，C_BThe perimeter of the bottom surface of the grain pile; f. of_FThe average friction coefficient between the side surface of the grain pile and the side surface of the granary; b_H1、b_H2、…、b_HkAre coefficients.

2. The method of claim 1, wherein the floor pressure mean value is a floor pressure mean value

Is estimated by

Comprises the following steps:

wherein, a_B(m) is

Estimate coefficients of the term, m 1_B，N_BIs composed of

Estimating the order of the polynomial;

the average value of the output values of the two circles of pressure sensors on the bottom surface is shown.

3. The method of claim 1, wherein the side pressure mean value is a measure of the volume in the grain bin based on two circles of pressure on the floor

Is estimated by

Comprises the following steps:

wherein, b_F(n) is

Estimate coefficients of the term, N1_F,N_FIs composed of

Estimating the order of the polynomial, K_SDIs the coefficient of the difference term of the variance.

4. The grain bin reserve detection method based on two-turn pressure statistic at floor according to claim 3, wherein when k is 2,3, or 4,neutralization coefficient b in granary grain storage quantity model_H1、b_H2、…、b_HkCorresponding multiplied side pressure mean value

Is estimated by

In (II)_DOrder N of(s) term_FDifferent.

5. The grain bin reserve detection method based on the bottom surface two-circle pressure statistic amount according to claim 1, characterized in that in the step 1), the output values of the two-circle pressure sensor obtained through detection are further filtered; the filtering method comprises the following steps: retaining only Q_BS(s_Inner) The output value of the inner ring pressure sensor; wherein Q is_BS(s_Inner) Comprises the following steps:

Q_BS(s_Inner)＝{Q_B(s_Inner(j))|-T_SDSD_Med(s_Inner)≤Q_B(s_Inner(j))-Q_Med(s_Inner)≤C_ISkT_SDSD_Med(s_Inner)}

wherein, C_ISkThe output value of the inner ring pressure sensor is deviated from the state distribution coefficient; t is_SDRemoving a threshold coefficient for the inner ring pressure sensor point;

Q_B(s_Inner(i) is a sequence Q of inner ring pressure sensor output values_B(s_Inner) I-1, 2, N_I，N_IThe number of the pressure sensors is the inner ring; SD_Med(s_Inner) Is the standard deviation of the filtered inner ring pressure sensor output value; q_Med(s_Inner) Obtaining a median value according to the output values of the inner ring pressure sensors;

retaining only Q_BS(s_Outer) The output value of the outer ring pressure sensor; wherein Q is_BS(s_Outer) Comprises the following steps:

Q_BS(s_Outer)＝{Q_B(s_Outer(i))|-C_TSDT_SDSD_Med(s_Outer)≤Q_B(s_Outer(i))-Q_Med(s_Outer)≤C_OSkC_TSDT_SDSD_Med(s_Outer)}

wherein, C_OSkThe distribution coefficient of the deviation of the outer ring sensor points is shown; c_TSDRemoving a threshold coefficient for the outer ring sensor points; t is_SDRemoving a threshold coefficient for the inner ring sensor points; q_B(s_Outer(i) Is a sequence of outer ring sensor output values Q_B(s_Outer) I-1, 2, N_O，N_OThe number of the sensors is the inner ring; SD_Med(s_Outer) Is the standard deviation of the outer ring pressure sensor output value based on filtering; q_Med(s_Outer) And obtaining the output value median of the outer ring pressure sensor.

6. A granary reserve detection method based on single-circle pressure intensity statistic of a bottom surface is characterized by comprising the following steps:

1) detecting the output value of a single-ring pressure sensor arranged on the bottom surface of the granary;

2) determining mean of small values for single-turn pressure sensors

Mean value of large value

Small value standard deviation SD_SS(s) and large standard deviation SD_SL(s); the small value of the single-turn pressure sensor is the output value of the single-turn pressure sensor of which the sensor output value is smaller than a set value, and the large value of the single-turn pressure sensor is the output value of the single-turn pressure sensor of which the sensor output value is larger than the set value;

3) according to the mean of small values

Sum large value mean

Determining bottom surface pressure mean

4) According to the mean of small values

Mean value of large value

Small value standard deviation SD_SS(s) and large standard deviation SD_SL(s) determining the mean lateral pressure

5) The pressure intensity mean value of the bottom surface of the grain pile

And side pressure mean value

W＝A_BQ_BNF(s)

H＝b_HQ_BNF(s) or

Wherein A is_BThe area of the bottom surface of the grain pile; q_BNF(s) is the equivalent average pressure of the bottom surface of the granary; h is the grain pile height; k_cThe geometric shape parameters of the grain pile are obtained; k_c＝C_B/A_B；A_BThe area of the bottom surface of the grain pile; c_BThe perimeter of the bottom surface of the grain pile; f. of_FThe average friction coefficient between the side surface of the grain pile and the side surface of the granary; b_H、b_H1、b_H2Are coefficients.

7. The method for detecting the reserve volume of the granary based on the single-turn pressure statistic of the bottom surface of claim 6, wherein in the step 2), the set value is obtained according to the median value of the output values of the single-turn pressure sensor.

8. The method of claim 7, wherein the set value is the bottom surface single-turn pressure statistic-based granary reserve volume detection method

Wherein,

the average value of the output values of the detected single-ring pressure sensor and the output values of the left and right set numbers is obtained; c_MVThe coefficients are adjusted for the division.

9. The method for detecting the storage capacity of the granary based on the bottom surface single-circle pressure statistic amount according to claim 6, wherein in the step 1), the output value of the single-circle pressure sensor obtained through detection is filtered; the filtering method comprises the following steps: retaining only Q_SE(s) the output value of the single-turn pressure sensor; wherein Q is_SE(s) is:

Q_SE(s)＝{Q(s(j))|-T_SDSD_Med(s)≤Q(s(j))-Q_Med≤C_SkT_SDSD_Med(s)}

wherein, Q (s (j)) is the j element value of the single-ring pressure sensor output value, j is 1,2_S，N_SThe number of the sensors is; t is_SDRemoving coefficients for preset single-ring pressure sensor output values; SD_Med(s) single-turn pressure sensor output for detectionStandard deviation of the values; c_SkThe single-ring sensor output value is the off-normal distribution coefficient; q_MedAnd obtaining the median value according to the output value of the single-circle pressure sensor.

10. The method of claim 9, wherein the pressure mean value of the floor of the grain pile is determined based on the floor single-turn pressure statistic

Is estimated by

Comprises the following steps:

wherein, a_B(m) is

Estimate coefficients of the term, m 1_B，N_BIs composed of

Polynomial order of (1).

11. The method of claim 6, wherein the mean side pressure value is a mean value of the pressure of the grain bin based on the single-turn pressure statistic of the floor

Is estimated by

Comprises the following steps:

wherein, K_SDIs the coefficient of difference of variance; b_F(n) is

Estimate coefficients of the term, N1_F，N_FIs composed of

The order of the polynomial.

12. The method of claim 11, wherein the grain bin reserves are modeled in a grain quantity model

Time, and coefficient b_H1、b_H2Corresponding multiplied side pressure mean value

Is estimated by

In (II)_DOrder N of(s) term_FDifferent.

13. A granary reserve detection device based on a floor pressure statistic and a reserve equation, comprising a memory and a processor, wherein the processor is used for executing instructions stored in the memory to realize the granary reserve detection method based on the floor two-turn pressure statistic according to any one of claims 1-5 or the granary reserve detection method based on the floor one-turn pressure statistic according to any one of claims 6-12.