Method and device for peeling a belt scale

Abstract

The application discloses a method and device for removing skin of a belt scale. The method comprises the following steps: after the belt scale starts to remove skin, collecting data under the condition that the conveying belt runs at a constant speed; using a sliding time window, dividing the collected data into multiple data segments according to the skin weight difference of all skin weight detection data groups, and each data segment comprises multiple skin weight detection data groups; for each data segment, calculating the median of all skin weight detection data in each skin weight detection data group in the data segment, and using the median of all skin weight detection data in the data segment to obtain a first fitting curve and a fitting function of the data segment, and combining all the first fitting curves into a skin weight curve; and outputting the fitting functions of all the data segments and the skin weight curve. The application solves the many disadvantages existing in the single skin weight value for describing the skin weight of the conveying belt, improves the process control level of the state monitoring of the belt scale, and improves the weighing accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of weighing measurement, more particularly, to a method and device for removing tare weight of a belt scale. BACKGROUND

[0002] Currently, the methods for removing tare weight (referred to as "tare removal") of a belt scale generally include positive and negative integration method and threshold method. Both of the two methods need to measure the whole length of the conveying belt first, and then the whole length of the empty conveying belt is integrated to obtain the average linear density value of the conveying belt, which is used as the basis for calculating the tare weight during the data processing of the weighing. The difference between the gross weight and the average tare weight is used as the net weight of the material.

[0003] The positive and negative integration method inevitably causes the phenomenon of "empty scale walking", or the cumulative value may decrease over time, which is against common sense and is difficult for users to accept. The threshold method has the advantage of avoiding the phenomenon of "empty scale walking", but the weighing sensitivity of the belt scale is reduced due to the pre-set weighing sensitivity threshold, which may cause the phenomenon of "no material walking". Since both of the two tare removal modes use the average linear density of the conveying belt, the correct measurement of the length of the conveying belt and the non-uniform dispersion of the linear density are crucial to the accuracy of the weighing measurement of the belt scale.

[0004] Although many new tare removal methods have been improved, most of them are complex and difficult to implement. Generally, since the total length of most industrial belt scales is relatively long, and the conveying belt has non-uniformity in terms of material, thickness, joint, etc., the change in linear density is too large, and the deviation between the average value of multiple measurements and the theoretical optimal value, the tare removal method which describes the tare weight with a certain numerical value has many drawbacks, such as the decrease in weighing accuracy with the increase in the total weight being measured, the regularity deviation of the tare weight value, and the irregularity deviation of the tare weight value caused by unexpected factors. SUMMARY

[0005] The present application provides a tare removal method and device for a belt scale, which divides and collects data using a sliding time window, obtains a tare weight curve based on the median of the data segment, and visually displays the change of the tare weight value with the position of the conveying belt through images, so as to express the tare weight of all unit lengths within the total length of the belt scale in a scientific description manner, solve the many drawbacks of describing the tare weight of the conveying belt with a single tare weight value, improve the process control level of the state monitoring of the belt scale, and improve the weighing accuracy.

[0006] The present application provides a tare removal method for a belt scale, which includes:

[0007] After the belt scale starts to peel, under the condition that the conveying belt runs at a constant speed, the collected data is received, and the collected data includes a plurality of tare weight detection data groups; wherein the detection time of each tare weight detection data group is a detection period, and the detection period includes a plurality of same collection periods, and each detection period corresponds to a tare weight detection data in the tare weight detection data group;

[0008] The tare weight difference value of each tare weight detection data group is calculated;

[0009] Using a sliding time window, the collected data is divided into a plurality of data segments according to the tare weight difference values of all tare weight detection data groups, and each data segment includes a plurality of tare weight detection data groups;

[0010] For each data segment, the median of all tare weight detection data in each tare weight detection data group in the data segment is calculated, and the median of all tare weight detection data in the data segment is used to obtain a first fitting curve and a fitting function of the data segment, and all first fitting curves are combined into a tare weight curve;

[0011] The fitting functions and the tare weight curve of all data segments are output.

[0012] Preferably, the peeling method further comprises:

[0013] The sum of the areas of the fitting curves of all data segments is divided by the total number of detection periods of the collected data to obtain a tare weight value and output.

[0014] Preferably, before obtaining the first fitting curve and the fitting function, it further comprises:

[0015] For each data segment, the maximum value and the minimum value of all tare weight detection data in each tare weight detection data group in the data segment are calculated, and the maximum value of all tare weight detection data in the data segment is used to obtain a second fitting curve of the data segment, and the minimum value of all tare weight detection data in the data segment is used to obtain a third fitting curve of the data segment, and the area of the envelope region between the second fitting curve and the third fitting curve of each data segment is calculated;

[0016] If the area of all data segments is less than the corresponding first threshold value, the first fitting curve and the fitting function are obtained.

[0017] Preferably, if the area of at least one data segment is greater than the corresponding first threshold value, the collected data is re-divided.

[0018] Preferably, the tare weight difference value of each tare weight detection data group is calculated, and specifically comprises:

[0019] The average value of all tare weight detection data in each tare weight detection data group is calculated as a first average value;

[0020] calculating the average of the first average values of all the tare detection data groups as a second average value;

[0021] calculating the difference between the first average value and the second average value of each tare detection data group as a tare difference value of the tare detection data group.

[0022] Preferably, the collected data is segmented into multiple data segments according to the tare difference values of all the tare detection data groups by using a sliding time window, specifically including the following steps in a loop until all the tare detection data groups are segmented into data segments:

[0023] dividing a first window from the collected data in units of a minimum specification sliding time window, the minimum specification sliding time window covering multiple tare detection data groups;

[0024] calculating the average of the absolute values of the tare difference values of all the tare detection data groups in the first window as a third average value;

[0025] if the third average value is greater than a second threshold value, the collected data in the first window is taken as a first data segment;

[0026] dividing a second window from the collected data in units of the minimum specification sliding time window, starting from the tare detection data group next to the first window;

[0027] updating the first window with the second window.

[0028] The application also provides a tare removing device of a belt scale, including a receiving module, a tare difference value calculation module, a segmentation module, a fitting module and an output module;

[0029] The receiving module is used to receive collected data under the condition that the conveying belt runs at a constant speed after the belt scale starts tare removing, the collected data including multiple tare detection data groups; wherein the detection time of each tare detection data group is one detection period, the detection period including multiple same collection periods, and each detection period corresponding to one tare detection data in the tare detection data group;

[0030] The tare difference value calculation module is used to calculate the tare difference value of each tare detection data group;

[0031] The segmentation module is used to segment the collected data into multiple data segments according to the tare difference values of all the tare detection data groups by using a sliding time window, each data segment including multiple tare detection data groups;

[0032] The fitting module is used to calculate the median of all the tare detection data in each tare detection data group in each data segment, and obtain a first fitting curve and a fitting function of the data segment by using the median of all the tare detection data in the data segment, and combine all the first fitting curves into a tare curve.

[0033] The output module is configured to output the fitting function and the tare curve of all data segments.

[0034] Preferably, the tare removing device further comprises an area calculation module and a judgment module.

[0035] The area calculation module is configured to calculate the maximum value and the minimum value of all tare detection data in each tare detection data group in each data segment, obtain a second fitting curve of the data segment by using the maximum value of all tare detection data in the data segment, obtain a third fitting curve of the data segment by using the minimum value of all tare detection data in the data segment, and calculate the area of an envelope region between the second fitting curve and the third fitting curve of each data segment.

[0036] The judgment module is configured to judge whether the area of all data segments is less than a corresponding first threshold value.

[0037] Preferably, the tare difference calculation module comprises a first calculation module, a second calculation module and a third calculation module.

[0038] The first calculation module is configured to calculate the average value of all tare detection data in each tare detection data group as a first average value.

[0039] The second calculation module is configured to calculate the average value of the first average values of all tare detection data groups as a second average value.

[0040] The third calculation module is configured to calculate the difference between the first average value and the second average value of each tare detection data group as the tare difference of the tare detection data group.

[0041] Preferably, the segmentation module comprises a first window obtaining module, a fourth calculation module, a first data segment obtaining module, a second window obtaining module and an updating module.

[0042] The first window obtaining module is configured to divide a first window from the collected data in units of a minimum specification sliding time window, and the minimum specification sliding time window covers a plurality of tare detection data groups.

[0043] The fourth calculation module is configured to calculate the average value of the absolute values of the tare differences of all tare detection data groups in the first window as a third average value.

[0044] The first data segment obtaining module is configured to, if the third average value is greater than a second threshold value, take the collected data in the first window as a first data segment.

[0045] The second window obtaining module is configured to divide a second window from the collected data in units of the minimum specification sliding time window, starting from the tare detection data group next to the first window.

[0046] The updating module is configured to update the first window with the second window.

[0047] Other features of the present application, its nature and advantages will become apparent from the following detailed description of the exemplary embodiments of the application with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the application and, together with the description, serve to explain the principles of the application.

[0049] Figure 1 The flow chart of the method for removing the skin of the belt scale provided in the present application;

[0050] Figure 2 The structural schematic diagram of the belt scale provided in the present application;

[0051] Figure 3 The flow chart of the method for removing the skin of the belt scale provided in the present application;

[0052] Figure 4 The structural diagram of the device for removing the skin of the belt scale provided in the present application. DETAILED DESCRIPTION

[0053] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions, and numerical values set forth in these embodiments are not limiting to the scope of the present application unless specifically stated otherwise.

[0054] The following description of at least one exemplary embodiment is merely exemplary in nature and is in no way intended to limit the application or its application or uses.

[0055] Techniques, methods, and apparatus known to those of ordinary skill in the relevant art can not be discussed in detail herein. However, where appropriate, techniques, methods, and apparatus should be considered as being part of the specification.

[0056] In all of the examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as a limitation. Thus, other examples of the exemplary embodiments can have different values.

[0057] The present application provides a method and device for removing the skin of a belt scale, which divides and collects data by using a sliding time window, obtains a skin weight curve based on the median of a data segment, visually displays the change of the skin weight value with the position of the conveyor belt by an image, and expresses the skin weight of all unit lengths in the total length of the belt scale in a scientific manner, thereby solving the many drawbacks of describing the skin weight of the conveyor belt by a single skin weight value, improving the process control level of the state monitoring of the belt scale, and improving the weighing accuracy.

[0058] As Figure 1 shown, the belt scale peeling method provided by the present application comprises the following steps:

[0059] S110: After the belt scale starts peeling, under the condition that the conveying belt runs at a constant speed, the collected data is received, and the collected data comprises a plurality of tare weight detection data groups; wherein the detection time of each tare weight detection data group is a detection period, and the detection period comprises a plurality of same collection periods, and each detection period corresponds to a tare weight detection data in the tare weight detection data group.

[0060] As an embodiment, as Figure 2 shown, an inductive marker 220 is installed at a specific position on the conveying belt 210 of the belt scale, and is used as a peeling detection starting point. An inductive marker detection device 230 is arranged on the rack of the belt scale, and is used to detect the starting pulse measured by the inductive marker 220 on the conveying belt. The detection signal of the inductive marker detection device 230 is transmitted to the belt scale controller 240, and the belt scale controller 240 is used to control the belt scale to collect tare weight detection data.

[0061] Thus, the conveying belt 210 moves one round every time, the inductive marker detection device 230 detects the presence of the inductive marker 220 once, and the time interval between the adjacent two detections of the inductive marker 220 is a detection period. The belt scale controller 240 divides one detection period into a plurality of (for example, G, which is a manually preset value) same collection periods, and controls the belt scale to collect tare weight detection data once in each collection period.

[0062] The following is described by taking the division of each detection period into G collection periods as an example. After the belt scale starts peeling, the conveying belt is controlled to run at a constant speed at a set speed. The first detection of the inductive marker is taken as the starting point of the tare weight detection data record, G tare weight detection data are obtained in each detection period, and a detection signal of an inductive marker is also obtained, that is, there are G tare weight detection data between the adjacent two detection signals, and the G tare weight detection data constitute a tare weight detection data group. When the tare weight detection data and the detection signal are plotted on the coordinate axes, the detection signal is displayed on the X axis in turn, and the tare weight detection data is displayed on the Y axis in turn. Thus, each X value corresponds to G Y values, and assuming that W tare weight detection data groups are collected, the X1 coordinate corresponds to G Y values: Y 11 , Y 12 , Y 13 …Y 1G ; the X2 coordinate corresponds to G Y values: Y 21 , Y 22 , Y 23 …Y 2G ; and so on, the X W coordinate corresponds to G Y values: Y W1 , YW2 Y W3 …Y WG .

[0063] S120: Calculate the tare difference value of each tare detection data group.

[0064] As an embodiment, the tare difference value of each tare detection data group is calculated, specifically including:

[0065] S1201: Calculate the average value of all tare detection data in each tare detection data group as the first average value AVG(X i ):

[0066]

[0067] Wherein, AVG(X i ) represents the first average value corresponding to the i-th X point.

[0068] S1202: Calculate the average value of the first average value of all tare detection data groups as the second average value G avg :

[0069]

[0070] S1203: Calculate the difference value between the first average value and the second average value of each tare detection data group as the tare difference value Diff(X i ) of the tare detection data group:

[0071]

[0072] Wherein, Diff(X i ) represents the tare difference value corresponding to the i-th X point.

[0073] S130: Using a sliding time window, the collected data is divided into multiple data segments according to the tare difference value of all tare detection data groups, and each data segment includes multiple tare detection data groups.

[0074] Specifically, as an embodiment, as shown in Figure 3 , using a sliding time window, the collected data is divided into multiple data segments according to the tare difference value of all tare detection data groups, specifically including the following steps in a loop until all tare detection data groups are divided into data segments:

[0075] S310: Divide the first window from the collected data in units of the smallest specification sliding time window, and the smallest specification sliding time window covers multiple tare detection data groups.

[0076] Wherein, the number of tare detection data groups (i.e. X value) that can be covered by the smallest specification sliding time window is Nmin (N min ∈N + ). The maximum number of tare weight detection data groups (i.e. X values) that can be covered by the sliding time window is N. max (N max ∈N + ).

[0077] If we start from the i-th X value (i.e. X i Point) starts to divide the new sliding time window, and the first window covers the time from X i Click to X i+Nmin The tare weight detection data of the point.

[0078] S320: Calculate the average absolute value of the tare weight difference of all tare weight detection data groups in the first window as the third average value WIN_Diff(m, N L ):

[0079]

[0080] Among them, WIN_Diff(m,N L ) indicates that the sliding time window specification is N L (i.e. the length is N L X value), covering the mth X value point (X m ) to the m+Lth X value point (X m+NL ) is a third average value of the first window. For S310, N L =N min .

[0081] S330: Determine the third average value WIN_Diff (m, N L ) is greater than the second threshold WIN F If so, it indicates that the tare weight offset value in the first window is large and lasts for a long time, and then S340 is executed; otherwise, it indicates that the tare weight offset value in the first window is not large and lasts for a short time, and then S380 is executed.

[0082] S340: Use the collected data in the first window as the first data segment, and execute S350.

[0083] S350: Determine whether the length of the collected data after the first data segment is less than the minimum size N of the sliding time window min If yes, execute S3150; otherwise, execute S360.

[0084] S360: From the tare weight detection data group immediately after the first window (ie X m+NL+1 Point) as the starting point, with the minimum specification N minThe second window is divided from the collected data in units of the sliding time window, at this time N L = N min .

[0085] S370: update the first window with the second window, and return to S320.

[0086] S380: increase the size of the sliding time window by one tare detection data group (i.e. N L = N min + 1), divide all tare detection data groups in the first window and the tare detection data group immediately after the first window into a third window (from X m to X m+Nmin+1 ).

[0087] S390: determine whether the size N min + 1 of the third window is equal to the maximum size N max of the sliding time window. If yes, execute S3100; otherwise, execute S3140.

[0088] S3100: take the collected data in the third window as the second data segment, and execute S3110.

[0089] S3110: determine whether the length of the collected data after the third data segment is less than the minimum size N min of the sliding time window. If yes, execute S3160; otherwise, execute S3120.

[0090] S3120: divide a fourth window from the collected data in units of the minimum size of the sliding time window, starting from the tare detection data group (i.e. X m+Nmin+2 ) immediately after the third window.

[0091] S3130: update the first window with the fourth window, and return to S320.

[0092] S3140: update the first window with the third window, and return to S320.

[0093] S3150: merge the collected data after the first window with the first data segment to form a third data segment.

[0094] S3160: merge the collected data after the third window with the second data segment to form a fourth data segment.

[0095] As an embodiment, after step S130, S160-170 are executed.

[0096] S160: For each data segment, calculate the median of all tare weight detection data in each tare weight detection data group in the data segment, and use the median of all tare weight detection data in the data segment to obtain a first fitting curve and a corresponding fitting function for the data segment, and combine all first fitting curves into a tare weight curve.

[0097] Assume that the collected data is divided into M data segments. If the first data segment has N1 X points; the second data segment has N2 X points; the third data segment has N3 X points, and so on, the hth data segment has N h There are X points, and the Mth data segment has N M X points.

[0098] For each data segment, the i-th (1≤i≤N L , N L is the length of the data segment) X value points (X i ) is:

[0099] Median(Xi)=Med(Y i1 ,Y i2 ,Y i3 ,...,Y iG ) (5)

[0100] Where Median(X i ) is the statistical median calculation function: G tare weight detection data Y i1 To Y iG Arrange them in order of size. If the number of data G is an odd number, the middle data is the median of the group of data; if the number of data G is an even number, the arithmetic mean of the two middle data is the median of the group of data.

[0101] Perform curve fitting on the median of all tare weight detection data groups in each data segment to obtain the first fitting curve and the corresponding fitting function f h (x). Function f h (x) represents the fitting function of the hth (1≤h≤M) data segment, which represents the fluctuation of the peeling value of the conveyor belt in the hth data segment.

[0102] Specifically, the fitting curve model selects a polynomial model, and the order of the polynomial model is determined by the first average value AVG(X i ) is selected from the range of standard deviation SD values.

[0103]

[0104] Among them, N LThe number of X points in the data segment (i.e., the length of the data segment).

[0105] If 0 < SD≤0.05, the order of the polynomial model is selected as 4; if 0.05 < N≤0.1, the order of the polynomial model is selected as 5; if 0.1 < N≤0.2, the order of the polynomial model is selected as 6; if 0.2 < N≤0.5, the order of the polynomial model is selected as 7; if N > 0.5, in order to save hardware computing resources and improve operation speed, the order of the polynomial model is selected as 8.

[0106] After obtaining the first fitting curve of each data segment, the corresponding first fitting curve is arranged in the order of the data segment to obtain the tare curve.

[0107] S170: output the fitting function of all data segments and the tare curve. Specifically, the combination of the fitting functions of all data segments is the overall fitting function corresponding to the collected data.

[0108] The expression formula of the overall fitting function G(x) is:

[0109]

[0110] As another embodiment, the tare removal method of the present application further comprises:

[0111] S180: take the quotient of the sum of the areas of the fitting curves of all data segments and the total number of detection periods of the collected data as the tare value V(x) and output it.

[0112] Wherein, the area of the fitting curve of each data segment is calculated by the principle of integral area, so as to obtain the tare value V(x):

[0113]

[0114] Preferably, between steps S130 and S160, S140 and S150 are further included.

[0115] S140: for each data segment, calculate the maximum value Max(X i ) and the minimum value Min(X i ) of all tare detection data in each tare detection data group in the data segment, and obtain the second fitting curve of the data segment by using the maximum value of all tare detection data in the data segment, obtain the third fitting curve of the data segment by using the minimum value of all tare detection data in the data segment, and calculate the area S c of the envelope region between the second fitting curve and the third fitting curve of each data segment.

[0116] The second fitting curve and the third fitting curve are obtained by using the polynomial fitting curve model, and the standard deviation SD of each data segment is calculated by referring to formula (6).

[0117] If 0 < SD≤ 0.05, the polynomial model order is selected as 3; if 0.05 < N≤ 0.1, the polynomial model order is selected as 4; if 0.1 < N≤ 0.2, the polynomial model order is selected as 5; if 0.2 < N≤ 0.5, the polynomial model order is selected as 6; if N > 0.5, in order to save hardware computing resources and improve operation speed, the polynomial model order is selected as 7.

[0118] After the second fitting curve and the third fitting curve are obtained, the area value S of the envelope region between the second fitting curve and the third fitting curve of each data segment is calculated by using an integral formula c :

[0119]

[0120] wherein, x i is the first X value point of the data segment, and x i+NL is the last X value point of the data segment. h S150: judging whether the areas of all data segments are less than the corresponding first threshold value H

[0121] If yes, S160 is executed; otherwise, at least one data segment has an area greater than the corresponding first threshold value,

[0122] indicating that the fluctuation of the skin weight value is large, and the skin needs to be removed again, and then returning to S110.

[0123] wherein, H h represents the first threshold value corresponding to the hth data segment, and H h = S F / N L , wherein, S F is an envelope area threshold value.

[0124] Based on the above skin removing method, the application further provides a skin removing device of a belt scale. As shown in the figure, the skin removing device includes a receiving module 410, a skin weight difference calculation module 420, a segmentation module 430, a fitting module 440, and an output module 450. Figure 4

[0125] The receiving module 410 is used to receive collected data under the condition that the belt scale starts to remove the skin and the conveying belt runs at a constant speed.

[0126] The collected data includes a plurality of skin weight detection data groups; wherein the detection time of each skin weight detection data group is a detection period, and the detection period includes a plurality of same collection periods.

[0127] Each detection period corresponds to a skin weight detection data in the skin weight detection data group.

[0128] ​The tare weight difference calculation module 420 is configured to calculate the tare weight difference of each tare weight detection data group.

[0129] The segmentation module 430 is configured to segment the collected data into a plurality of data segments according to the tare weight differences of all the tare weight detection data groups by using a sliding time window, each data segment including a plurality of tare weight detection data groups.

[0130] The fitting module 440 is configured to calculate the median of all the tare weight detection data in each tare weight detection data group in each data segment, and obtain a first fitting curve and a fitting function of the data segment by using the medians of all the tare weight detection data in the data segment, and combine all the first fitting curves into a tare weight curve.

[0131] The output module 450 is configured to output the fitting functions of all the data segments and the tare weight curve.

[0132] Preferably, the tare weight removing device further comprises an area calculation module 460 and a judgment module 470.

[0133] The area calculation module 460 is configured to calculate the maximum value and the minimum value of all the tare weight detection data in each tare weight detection data group in each data segment, and obtain a second fitting curve of the data segment by using the maximum values of all the tare weight detection data in the data segment, obtain a third fitting curve of the data segment by using the minimum values of all the tare weight detection data in the data segment, and calculate the area of the envelope region between the second fitting curve and the third fitting curve of each data segment.

[0134] The judgment module 470 is configured to judge whether the areas of all the data segments are less than a corresponding first threshold value.

[0135] Preferably, the tare weight difference calculation module 420 comprises a first calculation module 4201, a second calculation module 4202 and a third calculation module 4203.

[0136] The first calculation module 4201 is configured to calculate the average value of all the tare weight detection data in each tare weight detection data group as a first average value.

[0137] The second calculation module 4202 is configured to calculate the average value of the first average values of all the tare weight detection data groups as a second average value.

[0138] The third calculation module 4203 is configured to calculate the difference between the first average value and the second average value of each tare weight detection data group as the tare weight difference of the tare weight detection data group.

[0139] Preferably, the segmentation module 430 comprises a first window obtaining module 4301, a fourth calculation module 4302, a first data segment obtaining module 4303, a second window obtaining module 4304 and an updating module 4305.

[0140] The first window obtaining module 4301 is configured to divide the first window from the collected data in units of the minimum size of the sliding time window, and the minimum size of the sliding time window covers a plurality of tare weight detection data groups.

[0141] The fourth calculating module 4302 is configured to calculate the average value of the absolute values of the tare weight difference values of all the tare weight detection data groups in the first window as a third average value.

[0142] The first data segment obtaining module 4303 is configured to, if the third average value is greater than the second threshold value, take the collected data in the first window as a first data segment.

[0143] The second window obtaining module 4304 is configured to divide a second window from the collected data in units of the minimum size of the sliding time window, starting from the tare weight detection data group next to the first window.

[0144] The updating module 4305 is configured to update the first window with the second window, the third window or the fourth window.

[0145] Preferably, the dividing module 430 further comprises a judging module 4306, a third window obtaining module 4307, a second data segment obtaining module 4308, a fourth window obtaining module 4309 and a merging module 4310.

[0146] The judging module 4306 is configured to judge whether the third average value is greater than the second threshold value, whether the length of the collected data after the divided data segment is less than the minimum size of the sliding time window, and whether the size of the divided window is equal to the maximum size of the sliding time window.

[0147] The third window obtaining module 4307 is configured to increase the size of the sliding time window by one tare weight detection data group, and divide all the tare weight detection data groups in the first window and the tare weight detection data group next to the first window into a third window.

[0148] The second data segment obtaining module 4308 is configured to take the collected data in the third window as a second data segment.

[0149] The fourth window obtaining module 4309 is configured to divide a fourth window from the collected data in units of the minimum size of the sliding time window, starting from the tare weight detection data group next to the third window.

[0150] The merging module 4310 is configured to merge the collected data after the first window with the first data segment to form a third data segment, and merge the collected data after the third window with the second data segment to form a fourth data segment.

[0151] Compared with the prior art, the application has at least the following beneficial effects:

[0152] 1. The application improves the weighing accuracy as much as possible (especially the weighing accuracy of large value cumulative weighing application scenarios) through the sliding window segmentation method and curve fitting, and approximates all points in the total length of the belt scale to a function-based expression in a scientific description manner to solve the error caused by the single tare value obtained based on cumulative summation.

[0153] 2. When collecting data by using the sliding time window segmentation, the application reflects the size of the tare offset value and the length of the tare fluctuation duration through the size relationship between the mean value of the absolute value of the tare difference of each X value in the window and the threshold value, divides windows of different lengths with less hardware computing resources, and improves the accuracy of the tare description obtained based on the formula.

[0154] 3. The application compares the area value of the envelope region between the maximum value fitting curve and the minimum value fitting curve based on the integral formula calculation area with the set threshold value to judge the accuracy of the tare removal process and the reliability of the data.

[0155] 4. The application fits the curve based on the segmented median and draws the tare value curve with the function expression of the segmented fitting curve, which can visually show the influence of the position change of the entire conveyor belt on the tare value through the image, and improves the process control level of the belt scale state monitoring.

[0156] 5. The application determines the order of the polynomial fitting curve model according to the standard deviation value of the tare mean value in each sliding window, improves the calculation efficiency and the accuracy of the fitting curve, and further improves the accuracy of the entire method for measuring the tare.

[0157] 6. The application obtains the tare value based on the method of integrating the area sum of the segmented function and then averaging, which is significantly more accurate than the traditional method of accumulating and then averaging.

[0158] Although some specific embodiments of the application have been described in detail through examples, those skilled in the art should understand that the above examples are only for illustration and are not intended to limit the scope of the application. Those skilled in the art should understand that the above embodiments can be modified without departing from the scope and spirit of the application. The scope of the application is defined by the appended claims.

Claims

1. A method of skinning a belt scale, characterized by, The method comprises the following steps: After the belt scale starts to peel, under the condition that the conveying belt runs at a constant speed, collecting data is received, and the collected data comprises a plurality of tare weight detection data groups; wherein the detection time of each tare weight detection data group is a detection period, the detection period comprises a plurality of same collection periods, and each detection period corresponds to a tare weight detection data in the tare weight detection data group; The tare weight difference value of each tare weight detection data group is calculated; Using a sliding time window, the collected data is divided into a plurality of data segments according to the tare weight difference values of all tare weight detection data groups, and each data segment comprises a plurality of tare weight detection data groups; For each data segment, the median of all tare weight detection data in each tare weight detection data group in the data segment is calculated, the median of all tare weight detection data in the data segment is used to obtain a first fitting curve and a fitting function of the data segment, and all first fitting curves are combined into a tare weight curve; The fitting functions of all data segments and the tare weight curve are outputted; The quotient of the sum of the areas of all fitting curves of the data segments and the total number of detection periods of the collected data is taken as a tare weight value and outputted; Using a sliding time window, the collected data is divided into a plurality of data segments according to the tare weight difference values of all tare weight detection data groups, and the division specifically comprises the following steps which are repeated until all belt detection data groups are divided into data segments: A first window is divided from the collected data in units of a minimum specification sliding time window, and the minimum specification sliding time window covers a plurality of tare weight detection data groups; The average value of the absolute values of the tare weight difference values of all tare weight detection data groups in the first window is calculated as a third average value; If the third average value is greater than a second threshold value, the collected data in the first window is taken as a first data segment; A second window is divided from the collected data in units of the minimum specification sliding time window, starting from the tare weight detection data group next to the first window; The first window is updated with the second window.

2. The method of claim 1, wherein, Before the first fitting curve and the fitting function are obtained, the following steps are further included: For each data segment, the maximum value and the minimum value of all tare weight detection data in each tare weight detection data group in the data segment are calculated, the maximum value of all tare weight detection data in the data segment is used to obtain a second fitting curve of the data segment, the minimum value of all tare weight detection data in the data segment is used to obtain a third fitting curve of the data segment, and the area of the envelope region between the second fitting curve and the third fitting curve of each data segment is calculated; If the areas of all data segments are less than corresponding first threshold values, the first fitting curve and the fitting function are obtained.

3. The method of skinning a belt scale according to claim 2, wherein, If the area of at least one data segment is greater than the corresponding first threshold value, the collected data is re-divided.

4. The method of claim 1, wherein, The tare weight difference value of each tare weight detection data group is calculated, and the calculation specifically comprises the following steps: The average value of all tare weight detection data in each tare weight detection data group is calculated as a first average value; The average value of the first average values of all tare weight detection data groups is calculated as a second average value; The difference value between the first average value of each tare weight detection data group and the second average value is calculated as the tare weight difference value of the tare weight detection data group.

5. A debarking device of a belt scale using the debarking method according to any one of claims 1 to 4, characterized by The method comprises a receiving module, a tare difference calculation module, a segmentation module, a fitting module and an output module. The receiving module is configured to receive collected data under the condition that the belt conveyor runs at a constant speed after the belt scale starts to remove the skin, the collected data comprising a plurality of tare detection data groups; wherein the detection time of each tare detection data group is a detection period, the detection period comprising a plurality of identical collection periods, and each detection period corresponds to a tare detection data in the tare detection data group; The tare difference calculation module is configured to calculate the tare difference of each tare detection data group; The segmentation module is configured to divide the collected data into a plurality of data segments according to the tare differences of all tare detection data groups by using a sliding time window, each data segment comprising a plurality of tare detection data groups; The fitting module is configured to calculate the median of all tare detection data in each tare detection data group in each data segment, and obtain a first fitting curve and a fitting function of the data segment by using the median of all tare detection data in the data segment, and combine all first fitting curves into a tare curve; The output module is configured to output the fitting functions of all data segments and the tare curve, and output the quotient of the sum of areas of all fitting curves of data segments and the total number of detection periods of collected data as a tare value.

6. The belt scale debarking device of claim 5, wherein, The method further comprises an area calculation module and a judgment module. The area calculation module is configured to calculate the maximum value and the minimum value of all tare detection data in each tare detection data group in each data segment, and obtain a second fitting curve of the data segment by using the maximum value of all tare detection data in the data segment, obtain a third fitting curve of the data segment by using the minimum value of all tare detection data in the data segment, and calculate the area of an envelope region between the second fitting curve and the third fitting curve of each data segment; The judgment module is configured to judge whether the areas of all data segments are less than a corresponding first threshold value.

7. The belt scale debarking device of claim 5, wherein, The tare difference calculation module comprises a first calculation module, a second calculation module and a third calculation module. The first calculation module is configured to calculate the average value of all tare detection data in each tare detection data group as a first average value; The second calculation module is configured to calculate the average value of the first average values of all tare detection data groups as a second average value; The third calculation module is configured to calculate the difference between the first average value of each tare detection data group and the second average value as the tare difference of the tare detection data group.

8. The belt scale debarking device of claim 5, wherein, The segmentation module comprises a first window obtaining module, a fourth calculation module, a first data segment obtaining module, a second window obtaining module and an updating module. The first window obtaining module is configured to divide a first window from the collected data in units of a minimum specification sliding time window, the minimum specification sliding time window covering a plurality of tare detection data groups; The fourth calculation module is configured to calculate the average value of the absolute values of the tare differences of all tare detection data groups in the first window as a third average value; The first data segment obtaining module is configured to obtain a first data segment by using the third average value as a threshold value; The first data segment obtaining module is configured to, if the third average value is greater than a second threshold value, take the collected data in the first window as a first data segment; The second window obtaining module is configured to, starting from a tare weight detection data group next to the first window, divide a second window from the collected data in units of the minimum specification sliding time window; The updating module is configured to update the first window with the second window.

Images