A method for processing baking curve
By resampling the mean value of the baking curve and processing the sliding window, the baking start conditions are determined, which solves the problem of inaccurate identification of the baking starting point in the prior art, improves the accuracy of identification of the baking starting point, and ensures effective baking Accuracy of the curve.
Patent Information
- Application Number
- CN202410597033.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-14
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-05-14
AI Technical Summary
In the prior art, the identification of the baking starting point is relatively low, especially affected by the influence of the previous baking time or the significant temperature drop at the beginning of the baking, resulting in inaccurate identification of the baking starting point.
By obtaining the baking curve with 5 minutes as the sampling period, performing mean resampling with 1 hour as the cycle, and processing the dry bulb temperature in the sliding window with a 6-hour sliding window, determining the initial starting point data point, and combining with the target baking The minimum dry bulb temperature data point 3 hours before the time corresponding to the initial starting point data point in the curve, determine the baking start conditions.
The accuracy of identification of baking starting points is improved, making the effective baking curve obtained based on the baking starting points more accurate, solving the problem of inaccurate identification of baking starting points, and enhancing the accuracy of subsequent abnormality analysis and baking process display.
Smart Images

Figure CN118537439B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of electrical digital data processing, in particular to a method for processing a baking curve. Background Art
[0002] With the application of IoT baking equipment, baking process data has been accumulated. Flue-cured tobacco is mainly divided into three stages, namely the yellowing stage, the color fixing stage and the dry rib stage; the collected baking process data, in addition to the baking data corresponding to the above three stages, often also includes baking data before the yellowing stage and baking data after the dry rib stage. Based on the collected baking process data, a baking curve corresponding to the baking process can be fitted. The prior art discloses a method for identifying the baking starting point (i.e., the starting point of the yellowing stage) and the baking end point (i.e., the end point of the dry rib stage) in the baking curve. Based on the effective baking curve between the baking starting point and the baking end point in the baking curve, it can be judged whether there is an abnormality in the baking process. However, the method disclosed in the prior art has the problem of low accuracy in identifying the baking starting point. How to improve the accuracy of identifying the baking starting point is a problem that needs to be solved urgently. Summary of the invention
[0003] The object of the present invention is to provide a method for processing a baking curve, so as to improve the accuracy of identifying a baking starting point.
[0004] According to the present invention, a method for processing a baking curve comprises the following steps:
[0005] S100, obtaining a target baking curve, wherein the abscissa of the target baking curve is time, the ordinate of the target baking curve is dry-bulb temperature, and the target baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 5 minutes.
[0006] S200, determining a baking end time according to the target baking curve, where the baking end time is a time corresponding to a data point on the target baking curve that meets a preset baking end condition.
[0007] S300, if there is a first interval that meets the first preset condition and a second interval that meets the second preset condition in the target baking curve, enter S400; the first preset condition is that the time span is greater than or equal to 12 hours and the dry-bulb temperatures corresponding to all the data points included belong to the temperature range of 45°C to 54°C, and the second preset condition is that the time span is greater than or equal to 20 hours and the dry-bulb temperatures corresponding to all the data points included belong to the temperature range of 38°C to 42°C.
[0008] S400, determining a toasting start time according to the target toasting curve, where the toasting start time is a time corresponding to a data point on the target toasting curve that meets a preset toasting start condition.
[0009] S500: Determine a curve between the baking start time and the baking end time in the target baking curve as a valid baking curve.
[0010] The process of acquiring the data points on the target baking curve that meet the preset baking start conditions includes:
[0011] S410, obtaining a first mean resampling baking curve according to the target baking curve, where the first mean resampling baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 1 hour.
[0012] S420, sliding the first mean resampling baking curve with a sliding window length of 6 hours. If the first hour t in the jth sliding window j,1 Dry bulb temperature Tem j,1 Less than 35℃, Tem j,1 Greater than t j,1 Dry bulb temperature Tem in the first hour j,0 , and the 6th hour t in the sliding window j,6 Dry bulb temperature Tem j,6 With Tem j,1 The difference satisfies (Tem j,1 -Tem j,0 ) / (Tem j,6 -Tem j,1 )>30%, the data point corresponding to the first hour in the j-th sliding window is added to the first preset data point set; the first preset data point set is initialized to an empty set.
[0013] S430: Determine the latest data point in the first preset data point set as the initial starting point data point poi sta .
[0014] S440, setting the time corresponding to the target baking curve to [t sta-3 ,t sta ] and the corresponding data point with the lowest dry-bulb temperature is determined as the data point on the target baking curve that meets the preset baking start condition; t sta For poi sta The corresponding time, t sta-3 is t sta 3 hours early.
[0015] The present invention has at least the following beneficial effects:
[0016] The present invention obtains a target baking curve, which is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 5 minutes; in order to obtain a more accurate baking start time, the present invention first performs mean resampling on the target baking curve, and the sampling period corresponding to the mean resampling is 1 hour; for the obtained mean resampled baking curve, the present invention performs a sliding window with a sliding window length of 6 hours, and the last one that satisfies the corresponding dry-bulb temperature less than 35°C, the corresponding dry-bulb temperature greater than the dry-bulb temperature of the previous hour, and (Tem j,1 -Tem j,0 ) / (Tem j,6 -Tem j,1 )>30% condition is determined as the initial starting point data point. On this basis, the data point with the lowest dry bulb temperature in the three hours before the time corresponding to the initial starting point data point in the target baking curve is determined as the data point that meets the preset baking start condition. Compared with the starting point identification method disclosed in the prior art, the present invention can solve the problem of inaccurate baking starting point identification in the prior art due to the influence of the previous baking or the significant temperature drop at the beginning of baking. The baking starting point identified by the present invention is more accurate, so that the effective baking curve obtained based on the baking starting point is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0018] Figure 1 The present invention provides a flowchart of a method for processing a baking curve. DETAILED DESCRIPTION
[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0020] According to the present invention, a method for processing a baking curve is provided, such as Figure 1 As shown, including:
[0021] S100, obtaining a target baking curve, wherein the abscissa of the target baking curve is time, the ordinate of the target baking curve is dry-bulb temperature, and the target baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 5 minutes.
[0022] In this embodiment, the dry-bulb temperature is obtained by using a temperature sensor arranged in the baking room.
[0023] In this embodiment, the target baking curve corresponds to a baking process. Optionally, the target baking curve is obtained by segmenting a curve including multiple baking times, and the segmentation position is the position of the data point on the curve that meets the preset baking end condition.
[0024] S200, determining a baking end time according to the target baking curve, where the baking end time is a time corresponding to a data point on the target baking curve that meets a preset baking end condition.
[0025] As a preferred specific implementation manner, the process of acquiring the data points on the target baking curve that meet the preset baking end condition includes:
[0026] S210, obtaining the first data point poi corresponding to the dry-bulb temperature of 54° C. on the target baking curve 1 .
[0027] S220, obtaining the position at poi on the target baking curve 1 After and at a distance of poi 1 The corresponding time is 20 hours of data point poi 2 .
[0028] S230: Initialize a first variable k to 0.
[0029] S240, determining the poi on the target baking curve 2+k Whether the dry-bulb temperature of all corresponding specified data points is less than 54℃; poi 2+k The corresponding specified data point is no earlier than poi 2+k The corresponding time and distance poi 2+k The corresponding time is less than or equal to 3 hours; poi 2+k For poi 2 The kth data point after .
[0030] S250, if yes, then poi 2+k It is determined to be a data point on the target baking curve that meets the preset baking end condition; otherwise, update k to k+1, and repeat S240 until poi on the target baking curve is 2+k The dry bulb temperature of all the corresponding specified data points is less than 54℃, and poi2+k It is determined to be a data point on the target baking curve that meets the preset baking end condition.
[0031] Based on S210-S250, the toasting end time (ie, the toasting end starting point) identified by this embodiment is relatively accurate. As a specific implementation, as shown in Table 1.
[0032] Table 1
[0033] Serial number Artificial endpoint Automatic endpoint recognition End point error Remark 1 2023-05-22T01:13:11 2023-05-22T01:18:11 5 minutes 2 2023-05-30T02:12:05 2023-05-30T02:17:05 5 minutes 3 2023-06-06T18:56:08 2023-06-06T18:56:08 0 minutes After a while, the temperature drops 4 2023-05-23T20:28:13 2023-05-23T20:28:13 0 minutes Finished after baking 5 2023-06-25T08:51:52 2023-06-25T08:51:52 0 minutes Finished after baking
[0034] The maximum error of the starting point in Table 1 is 5 minutes, which is relatively small compared with the total baking time of 143 hours for this baking.
[0035] S300, if there is a first interval that meets the first preset condition and a second interval that meets the second preset condition in the target baking curve, enter S400; the first preset condition is that the time span is greater than or equal to 12 hours and the dry-bulb temperatures corresponding to all the data points included belong to the temperature range of 45°C to 54°C, and the second preset condition is that the time span is greater than or equal to 20 hours and the dry-bulb temperatures corresponding to all the data points included belong to the temperature range of 38°C to 42°C.
[0036] In this embodiment, if the target baking curve does not contain a first interval satisfying the first preset condition and a second interval satisfying the second preset condition, it is determined that the baking process corresponding to the target baking curve is abnormal, and S400 and subsequent steps are no longer executed.
[0037] S400, determining a toasting start time according to the target toasting curve, where the toasting start time is a time corresponding to a data point on the target toasting curve that meets a preset toasting start condition.
[0038] As a preferred specific implementation manner, the process of acquiring the data points on the target baking curve that meet the preset baking start conditions includes:
[0039] S410, obtaining a first mean resampling baking curve according to the target baking curve, where the first mean resampling baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 1 hour.
[0040] S420, sliding the first mean resampling baking curve with a sliding window length of 6 hours. If the first hour t in the jth sliding window j,1 Dry bulb temperature Tem j,1 Less than 35℃, Tem j,1 Greater than t j,1 Dry bulb temperature Tem in the first hour j,0 , and the 6th hour t in the sliding windowj,6 Dry bulb temperature Tem j,6 With Tem j,1 The difference satisfies (Tem j,1 -Tem j,0 ) / (Tem j,6 -Tem j,1 )>30%, the data point corresponding to the first hour in the j-th sliding window is added to the first preset data point set; the first preset data point set is initialized to an empty set.
[0041] S430: Determine the latest data point in the first preset data point set as the initial starting point data point poi sta .
[0042] S440, setting the time corresponding to the target baking curve to [t sta-3 ,t sta ] and the corresponding data point with the lowest dry-bulb temperature is determined as the data point on the target baking curve that meets the preset baking start condition; t sta For poi sta The corresponding time, t sta-3 is t sta 3 hours early.
[0043] Based on S410-S440, the toasting start time (ie, the toasting start point) identified by this embodiment is relatively accurate. As a specific implementation, as shown in Table 2.
[0044] Table 2
[0045] Serial number Manual identification starting point Automatically identify the starting point Starting point error Remark 1 2023-05-15T22:06:16 2023-05-15T21:01:16 1 hour 2 2023-05-23T22:35:58 2023-05-24T00:00:58 1.5 hours 3 2023-05-31T14:18:27 2023-05-31T14:18:27 0 hours Temperature drop last time 4 2023-05-17T18:45:43 2023-05-17T21:00:43 2.2 hours Heating interruption 5 2023-06-17T18:23:34 2023-06-17T18:03:34 20 minutes
[0046] The starting point error corresponding to sequence number 4 in Table 2 is relatively large. After analysis, it is found that there is an interruption in the heating process at this point, and there is a temperature drop process in the pre-yellowing period, which causes the recognition position deviation. However, the error of 2.2 hours accounts for 1.5% of the total baking time of 143 hours, which still meets the curve recognition requirements within the pre-yellowing period.
[0047] S500: Determine a curve between the baking start time and the baking end time in the target baking curve as a valid baking curve.
[0048] Compared with the starting point identification method disclosed in the prior art, the present embodiment can solve the problem of inaccurate identification of the baking starting point caused by the influence of the previous baking or the significant temperature drop at the beginning of baking. The baking starting point identified by the present embodiment is more accurate, so that the effective baking curve obtained based on the baking starting point is more accurate, and the subsequent abnormal analysis or baking process display based on the effective baking curve is more accurate.
[0049] As a preferred specific implementation, after S500, the method further includes:
[0050] S510: Obtain a third mean resampling baking curve according to the effective baking curve, wherein the third mean resampling baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 30 minutes.
[0051] In this embodiment, if the dry-bulb temperature is sampled according to a sampling period of 5 minutes, and it is estimated that a single baking time is 130 hours to 180 hours, then the number of data items for a single baking time is generally more than 1500 to 2000, and the corresponding data volume is relatively large; moreover, affected by the sensor equipment, the measured temperature has certain short-term mutations, resulting in certain glitches in the curve and other problems, affecting the visualization of the data.
[0052] In this embodiment, the effective baking curve with a sampling period of 5 minutes is averaged and resampled with a sampling period of 30 minutes, and the data volume is reduced to 200-300 data. Moreover, through the average resampling, the burr situation of the effective baking curve can be reduced, the abnormal temperature jump in a small range can be filtered out, and the overall curve characteristics can be retained.
[0053] S520, simplifying the third mean resampling baking curve to obtain a target simplified baking curve.
[0054] As a preferred specific implementation, S520 includes:
[0055] S521, connecting the first data point and the last data point of the third mean resampling baking curve to obtain an initial straight line L 0 .
[0056] In this embodiment, the first data point is the data point with the earliest corresponding time, and the last data point is the data point with the latest corresponding time.
[0057] S522, if the third mean resampling baking curve is consistent with L 0 The data point with the largest distance pd max,0 With L 0 The distance is greater than or equal to the target distance threshold d 0 , then enter S523.
[0058] In this embodiment, d 0 For experience value.
[0059] S523, with pd max,0 The third mean resampling baking curve is divided for the first time as the curve dividing point to obtain a partial curve set H after the first division. 1 ; Initialize the third variable f to 1.
[0060] In this embodiment, H 1 It includes 2 partial curves, the first data point and the last data point of one curve are the first data point and pd of the third mean resampling baking curve respectively. max,0 , the first and last data points of the other curve are pd max,0 and the third mean resampled baking curve's last data point.
[0061] S524, obtaining the partial curve set H after the f-th division f The number of partial curves included qua f .
[0062] In this embodiment, when f=1, H 1 The number of partial curves included qua 1 is 2. When f>1, for H f-1 Each part of the curve is judged again, for example, for H f-1 The δth part of the curve c δ , connect c δ The first and last data points of c δ The corresponding straight line; if c δ Up and c δ The corresponding data point with the largest straight-line distance is c δ The distance between the corresponding straight lines is greater than or equal to d 0 , then c δ Up and c δ The data point with the largest straight line distance is the dividing point of the curve. δ Divide, thus, c δ It is divided into two partial curves, and the two partial curves are added to H f If c δ Up and c δ The corresponding data point with the largest straight-line distance is c δ The distance between the corresponding straight lines is less than d 0 , then c δ Divide c δ Append to H f . Thus, H f , statistics H f The number of curves in the middle part can be obtained by f .
[0063] S525, if qua f +1 <num 1 , then update f to f+1, repeat S524 until num 1 ≤qua+1≤num 2or qua+1>num 2 or qua f =qua f-1 ;num 1 is the lower limit of the preset target quantity range, num 2 is the upper limit of the preset target quantity range; if num 1 ≤qua+1≤num 2 , then go to S526; if qua+1>num 2 , then enter S527.
[0064] In this embodiment, num 1 and num 2 All are empirical values, preferably, num 1 is 50, num 2 The value is 60, which not only reduces the amount of data that needs to be transmitted, but also solves the burr problem of the curve, making it easier to transmit and display data. It can also ensure as many characteristic inflection points in the curve as possible, so that the characteristics of the target simplified baking curve and the effective baking curve are similar.
[0065] In this embodiment, if qua f =qua f-1 , then enter S5221.
[0066] S526, H f The first data point and the last data point corresponding to each part of the curve are determined as the data points corresponding to the target simplified baking curve, and curve fitting is performed according to the data points corresponding to the target simplified baking curve to obtain the target simplified baking curve.
[0067] S527, obtaining the distance sequence dis after the f-1th division f-1 ,dis f-1 =(dis f-1,1 ,dis f-1,2 ,…,dis f-1,c ,…,dis f-1,v ), dis f-1,c H f-1 The cth curve in the figure is connected to L f-1,c The data point with the largest distance from L f-1,c The distance, L f-1,c is the line connecting the first and last data points of the cth partial curve, where c ranges from 1 to v, and v is H f-1 The number of curves in the middle section.
[0068] S528, get dis f-1 The largest front number in the middle distance 2 -qua f-1The set of partial curves cur corresponding to the distance.
[0069] S529, each cur η Perform a division process and add the first and last data points of the partial curve obtained after the division to U f-1 , and append the U f-1 Determine a set of data points corresponding to the target simplified baking curve, and perform curve fitting according to the set of data points corresponding to the target simplified baking curve to obtain the target simplified baking curve; η is the ηth partial curve in cur, L η For cur η The line connecting the first and last data points of U f-1 H f-1 The set of first and last data points of all included partial curves.
[0070] In this embodiment, each cur η The process of performing a division process is similar to the process of performing a division process on the third mean resampling baking curve, which will not be described in detail here.
[0071] Based on S521-S529, it is possible to set 0 When the value is smaller, a target number of data points in the third mean resampling baking curve are obtained, and these data points can better characterize the trend and characteristics of the third mean resampling baking curve.
[0072] As a preferred specific implementation, S522 further includes: if the third mean resampling baking curve is consistent with L 0 The data point with the largest distance pd max,0 With L 0 The distance is less than the target distance threshold d 0 , then enter S5221.
[0073] S5221, initialize the second variable r to 0.
[0074] S5222, set the distance threshold to d 0 -r×Δd, use the Douglas Peucker algorithm to simplify the third mean resampling baking curve after the rth update to obtain the number of data points after the r+1th simplification; Δd is the preset distance threshold adjustment step size.
[0075] In this embodiment, Δd is an empirical value, preferably, Δd=0.01.
[0076] S5223, if the number of data points after the r+1th simplification is less than num 1, then update r to r+1, and repeat S5222 until the number of data points after the r+1th simplification is greater than or equal to num 1 and less than or equal to num 2 Or the number of data points after the r+1th simplification is greater than num 2 ; If the number of data points after the r+1th simplification is greater than or equal to num 1 and less than or equal to num 2 , enter S5224.
[0077] In this embodiment, if the number of data points after the r+1th simplification is greater than num 2 , then randomly select num from the data points newly added in the r+1th simplification 2 -w r The number of data points is appended to the data point set after the r-th simplification, the data point set after the r-th simplification after the appending is determined as the specified data points, and the curve fitting is performed according to the specified data points to obtain the target simplified baking curve; the data points added in the r+1th simplification are the data points that exist in the data point set after the r+1th simplification but not in the data point set after the r-th simplification, w r is the number of data points in the data point set after the rth simplification.
[0078] S5224, performing curve fitting based on the data points after the r+1th simplification to obtain a target simplified baking curve.
[0079] Based on S5221-S5224, it is possible to set 0 When the value is larger, a target number of data points in the third mean resampling baking curve are obtained, and these data points can better characterize the trend and characteristics of the third mean resampling baking curve.
[0080] As a preferred specific implementation, after S500, the method further includes:
[0081] S600: Obtain a second mean resampling baking curve according to the effective baking curve, where the second mean resampling baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 2 hours.
[0082] S700, if the dry-bulb temperature corresponding to the qth data point on the second mean resampling baking curve is less than the difference between the dry-bulb temperature corresponding to the q-1th data point and the preset temperature rise threshold, or the dry-bulb temperature corresponding to the qth data point on the second mean resampling baking curve is greater than the sum of the dry-bulb temperature corresponding to the q+1th data point and the preset temperature rise threshold, then the qth data point is determined to be an abnormal data point.
[0083] In this embodiment, the qth data point on the second mean resampling baking curve is any data point on the second mean resampling baking curve except the last three data points in the dry tendon period.
[0084] In this embodiment, the preset temperature rise threshold is an empirical value. Optionally, the preset temperature rise threshold is 1°C.
[0085] S800, performing interpolation processing on all abnormal data points on the second mean resampling baking curve to obtain an interpolated second mean resampling baking curve.
[0086] As a preferred specific implementation, S800 includes:
[0087] S810, obtaining the abnormal data point interval G corresponding to the second mean resampling baking curve, G = (g 1 ,g 2 ,…,g a ,…,g b ), g a is the ath abnormal data point interval corresponding to the second mean resampling baking curve, the value range of a is 1 to b, and b is the number of abnormal data point intervals corresponding to the second mean resampling baking curve.
[0088] In this embodiment, g a Includes 1 abnormal data point or includes multiple abnormal data points in succession.
[0089] S820, traverse g a , if g a If the corresponding dry-bulb temperature range includes 55°C, the process proceeds to S830.
[0090] In this embodiment, g a The corresponding dry bulb temperature range is g a The temperature corresponding to the first data point of the earliest abnormal data point is included in g a Includes the temperature corresponding to the data point after the latest abnormal data point.
[0091] S830: If the data point corresponding to 66°C on the second mean resampling baking curve is not an abnormal data point, the data point corresponding to the 11th hour before the data point corresponding to 66°C is set as poi a,0 The corresponding dry-bulb temperature is determined to be 55°C, and the process proceeds to S840.
[0092] S840, g a Located in poi a,0 The dry-bulb temperature corresponding to the previous nth data point is determined as (55-0.5×Δtn 1 )℃, g aLocated in poi a,0 The dry-bulb temperature corresponding to the mth data point is determined as (55+2×Δtm 2 )℃;Δtn 1 For poi a,0 The corresponding time and g a Located in poi a,0 The time difference corresponding to the previous n-th data point, Δtm 2 g a Located in poi a,0 The time corresponding to the mth data point after that is the same as poi a,0 The corresponding time difference.
[0093] In this embodiment, Δtn 1 and Δtm 2 The unit is hour.
[0094] Based on S810-S840, when the data point corresponding to the dry-bulb temperature of 55° C. is abnormal, the abnormal data points near the corresponding dry-bulb temperature of 55° C. can be interpolated, and a more accurate interpolated second mean resampling baking curve can be obtained.
[0095] In this embodiment, if g a The corresponding dry-bulb temperature range includes 35°C, and there is a data point on the second mean resampling baking curve with a corresponding dry-bulb temperature less than 35°C and not an abnormal data point, then the corresponding time is t x +(35-tem x ) / 2 data point poi y The corresponding dry bulb temperature is determined to be 35°C, tem x is the dry-bulb temperature corresponding to the last data point on the second mean resampling baking curve whose corresponding dry-bulb temperature is less than 35°C and is not an abnormal data point, t x is the time corresponding to the last data point on the second mean resampling baking curve whose corresponding dry-bulb temperature is less than 35°C and is not an abnormal data point; g a Located in poi y The dry-bulb temperature corresponding to the zth data point before is determined as (35-2×Δtz 3 )℃, g a Located in poi y The dry-bulb temperature corresponding to the e-th data point is determined as (35+0.5×Δtz 4 )℃; Δtz 3 For poi y The corresponding time and g a Located in poi y The time difference corresponding to the previous z-th data point, Δtz4 g a Located in poi y The time corresponding to the e-th data point after that is the same as poi y The corresponding time difference.
[0096] In this embodiment, Δtz 3 and Δtz 4 The unit is hour.
[0097] This embodiment can interpolate abnormal data points with a corresponding dry-bulb temperature near 35° C. when the data points with a corresponding dry-bulb temperature of 35° C. are abnormal, and can obtain a more accurate interpolated second mean resampling baking curve.
[0098] In this embodiment, if g a The corresponding dry-bulb temperature interval does not include 35°C or 55°C. The dry-bulb temperature corresponding to the abnormal data point is determined based on the time difference between the data point corresponding to 35°C or 55°C and the abnormal data point. Specifically, if g a If the included abnormal data point is located before the data point corresponding to 35℃, then g is obtained based on the time difference between the data point corresponding to 35℃ and the abnormal data point and the amplitude of 2℃ rise per hour. a The dry-bulb temperature corresponding to the abnormal data points included; if g a If the abnormal data point is located after the data point corresponding to 55℃, then g is obtained based on the time difference between the abnormal data point and the data point corresponding to 55℃ and the amplitude of the increase of 1℃ per hour. a The dry-bulb temperature corresponding to the abnormal data points included; if g a The included abnormal data points are between the data points corresponding to 35℃ and 55℃, and g is obtained based on the time difference between the abnormal data point and the data point corresponding to 35℃ and the amplitude of 0.5℃ rise per hour. a The dry bulb temperature corresponding to the abnormal data point included, or the time difference between the data point corresponding to 55℃ and the abnormal data point and the amplitude of 0.5℃ rise per hour to obtain g a Includes dry-bulb temperatures corresponding to anomalous data points.
[0099] S900, smoothing the interpolated second mean resampling baking curve to obtain a smoothed second mean resampling baking curve.
[0100] As a specific implementation, a low-pass filter is used to smooth the interpolated second mean resampling baking curve, and the filter parameter cutoff frequency is set to 0.1 and the filter order is set to 8.
[0101] S1000: Determine whether there is an abnormality in the baking process corresponding to the target baking curve according to the second mean resampled baking curve after the smoothing process.
[0102] As a preferred specific implementation, S1000 includes:
[0103] S1010, obtaining an effective baking time according to the second mean resampling baking curve after the smoothing process.
[0104] In this embodiment, the effective baking time is the difference between the time corresponding to the last data point and the time corresponding to the first data point of the second mean resampling baking curve after the smoothing process.
[0105] S1020, if the effective baking time is greater than or equal to the preset baking time threshold, proceed to S1030; otherwise, it is determined that there is an abnormality in the baking process corresponding to the target baking curve.
[0106] In this embodiment, if the effective baking time is less than a preset baking time threshold, it is determined that an abnormality exists in the baking process corresponding to the target baking curve, specifically, an abnormality that the baking time is too short.
[0107] In this embodiment, the preset toasting time threshold is an empirical value. Optionally, the preset toasting time threshold is 144 hours.
[0108] S1030, obtaining a time proportion sequence [β 1 ,β 2 ,β 3 ]; β 1 is the proportion of yellowing period, β 2 is the proportion of color fixation period, β 3 It is the proportion of dry muscle period.
[0109] In this embodiment, the first data point corresponding to the dry bulb temperature of 43°C in the second mean resampling curve after smoothing is determined as the data point for dividing the yellowing period and the color fixing period, and the first data point corresponding to the dry bulb temperature of 56°C in the second mean resampling curve after smoothing is determined as the data point for dividing the color fixing period and the dry tendon period. Thus, the curve corresponding to the yellowing period, the curve corresponding to the color fixing period, and the curve corresponding to the dry tendon period in the second mean resampling baking curve after smoothing can be obtained, and then the duration corresponding to the yellowing period, the duration corresponding to the color fixing period, and the duration corresponding to the dry tendon period can be obtained. The ratio of the duration corresponding to the yellowing period to the total duration corresponding to the second mean resampling curve after smoothing is determined as the yellowing period duration ratio, the ratio of the duration corresponding to the color fixing period to the total duration corresponding to the second mean resampling curve after smoothing is determined as the color fixing period duration ratio, and the ratio of the duration corresponding to the dry tendon period to the total duration corresponding to the second mean resampling curve after smoothing is determined as the dry tendon period duration ratio.
[0110] S1040, if β 1 Belongs to the preset yellowing period duration ratio interval, β 2 Belongs to the preset color period duration ratio interval, and β 3 If the baking process is within the preset dry tendon period duration ratio interval, it is determined that there is no abnormality in the baking process corresponding to the target baking curve; otherwise, it is determined that there is an abnormality in the baking process corresponding to the target baking curve.
[0111] In this embodiment, if β 1 It does not belong to the preset yellowing period duration ratio interval, and it is judged that the baking process corresponding to the yellowing period is abnormal. Specifically, if β 1 If β is less than the lower limit of the preset yellowing period duration ratio, it is determined that the baking time corresponding to the yellowing period is too short; 1 If β is greater than the upper limit of the preset yellowing period duration ratio, it is determined that the baking time corresponding to the yellowing period is too long. 2 If it does not belong to the preset color setting period duration ratio interval, it is judged that the baking process corresponding to the color setting period is abnormal. Specifically, if β 2 If β is less than the lower limit of the preset color setting period duration ratio, it is judged that the baking time corresponding to the color setting period is too short; if 2 If β is greater than the upper limit of the preset color setting period duration ratio, it is judged that the baking time corresponding to the color setting period is too long. 3 It does not belong to the preset dry gluten period duration ratio interval, and it is judged that the baking process corresponding to the dry gluten period is abnormal. Specifically, if β 3 If β is less than the lower limit of the preset dry gluten period duration ratio, it is judged that the baking time corresponding to the dry gluten period is too short; 3 If the value is greater than the upper limit of the preset drying period duration ratio, it is judged that the baking time corresponding to the drying period is too long.
[0112] In this embodiment, the preset yellowing period duration ratio interval, the preset coloring period duration ratio interval and the preset dry tendon period duration ratio interval are all empirical values. Optionally, the preset yellowing period duration ratio interval is 40%-45%, the preset coloring period duration ratio interval is 30%-40%, and the dry tendon period is 20%-25%.
[0113] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.
Claims
1. A method for processing a baking curve, characterized in that: The following steps are involved: S100, obtaining a target baking curve, wherein the abscissa of the target baking curve is time, the ordinate of the target baking curve is dry-bulb temperature, and the target baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 5 minutes; S200, determining a baking end time according to the target baking curve, wherein the baking end time is a time corresponding to a data point on the target baking curve that meets a preset baking end condition; S300, if there is a first interval satisfying a first preset condition and a second interval satisfying a second preset condition in the target baking curve, enter S400; the first preset condition is that the time span is greater than or equal to 12 hours and the dry bulb temperatures corresponding to all the data points included belong to the temperature range of 45°C to 54°C, and the second preset condition is that the time span is greater than or equal to 20 hours and the dry bulb temperatures corresponding to all the data points included belong to the temperature range of 38°C to 42°C; S400, determining a baking start time according to the target baking curve, wherein the baking start time is a time corresponding to a data point on the target baking curve that meets a preset baking start condition; S500, determining a curve between the baking start time and the baking end time in the target baking curve as a valid baking curve; The process of acquiring the data points on the target baking curve that meet the preset baking start conditions includes: S410, acquiring a first mean resampling baking curve according to the target baking curve, where the first mean resampling baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 1 hour; S420, sliding the first mean resampling baking curve with a sliding window length of 6 hours. If the first hour t in the jth sliding window j,1 Dry bulb temperature Tem j,1 Less than 35℃, Tem j,1 Greater than t j,1 Dry bulb temperature Tem in the first hour j,0 , and the 6th hour t in the sliding window j,6 Dry bulb temperature Tem j,6 With Tem j,1 The difference satisfies (Tem j,1 -Tem j,0 ) / (Tem j,6 -Tem j,1 )>30%, then the data point corresponding to the first hour in the j-th sliding window is added to the first preset data point set; the first preset data point set is initialized to an empty set; S430: Determine the latest data point in the first preset data point set as the initial starting point data point poi sta ; S440, setting the time corresponding to the target baking curve to [t sta-3 ,t sta ] and the corresponding data point with the lowest dry-bulb temperature is determined as the data point on the target baking curve that meets the preset baking start condition; t sta For poi sta The corresponding time, t sta-3 is t sta 3 hours early.
2. The method for processing a baking curve according to claim 1, characterized in that: The process of acquiring the data points on the target baking curve that meet the preset baking end conditions includes: S210, obtaining the first data point poi1 on the target baking curve corresponding to a dry-bulb temperature of 54°C; S220, obtaining a data point poi2 on the target baking curve that is located after poi1 and is 20 hours away from poi1; S230, initializing a first variable k to 0; S240, determining the poi on the target baking curve 2+k Whether the dry-bulb temperature of all corresponding specified data points is less than 54℃; poi 2+k The corresponding specified data point is no earlier than poi 2+k The corresponding time and distance poi 2+k The corresponding time is less than or equal to 3 hours; poi 2+k is the kth data point after poi2; S250, if yes, then poi 2+k It is determined to be a data point on the target baking curve that meets the preset baking end condition; otherwise, update k to k+1, and repeat S240 until poi on the target baking curve is 2+k The dry bulb temperature of all the corresponding specified data points is less than 54℃, and poi 2+k It is determined to be a data point on the target baking curve that meets the preset baking end condition.
3. The method for processing a baking curve according to claim 1, characterized in that: After S500, the method further includes: S600, obtaining a second mean resampling baking curve according to the effective baking curve, where the second mean resampling baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 2 hours; S700: If the dry-bulb temperature corresponding to the qth data point on the second mean resampling baking curve is less than the difference between the dry-bulb temperature corresponding to the q-1th data point and the preset temperature rise threshold, or the dry-bulb temperature corresponding to the qth data point on the second mean resampling baking curve is greater than the sum of the dry-bulb temperature corresponding to the q+1th data point and the preset temperature rise threshold, then the qth data point is determined to be an abnormal data point; S800, performing interpolation processing on all abnormal data points on the second mean resampling baking curve to obtain an interpolated second mean resampling baking curve; S900, performing smoothing processing on the interpolated second mean resampling baking curve to obtain a smoothed second mean resampling baking curve; S1000: Determine whether there is an abnormality in the baking process corresponding to the target baking curve according to the second mean resampled baking curve after the smoothing process.
4. The method for processing a baking curve according to claim 3, characterized in that: S800 includes: S810, obtaining an abnormal data point interval G corresponding to the second mean resampling baking curve, G = (g1, g2, ..., g a ,…,g b ), g a is the ath abnormal data point interval corresponding to the second mean resampling baking curve, where a ranges from 1 to b, and b is the number of abnormal data point intervals corresponding to the second mean resampling baking curve; S820, traverse g a , if g a If the corresponding dry-bulb temperature range includes 55°C, then enter S830; S830: If the data point corresponding to 66°C on the second mean resampling baking curve is not an abnormal data point, the data point corresponding to the 11th hour before the data point corresponding to 66°C is set as poi a,0 The corresponding dry bulb temperature is determined to be 55°C, and the process proceeds to S840; S840, g a Located in poi a,0 The dry-bulb temperature corresponding to the previous n-th data point is determined to be (55-0.5×Δtn1)℃. a Located in poi a,0 The dry-bulb temperature corresponding to the mth data point is determined to be (55+2×Δtm2)℃; Δtn1 is poi a,0 The corresponding time and g a Located in poi a,0 The time difference corresponding to the previous n-th data point, Δtm2 is g a Located in poi a,0 The time corresponding to the mth data point after that is the same as poi a,0 The corresponding time difference.
5. The method for processing a baking curve according to claim 3, characterized in that: S1000 includes: S1010, obtaining an effective baking time according to the second mean resampling baking curve after the smoothing process; S1020, if the effective baking time is greater than or equal to the preset baking time threshold, proceed to S1030; otherwise, it is determined that there is an abnormality in the baking process corresponding to the target baking curve; S1030, obtaining a sequence [β1, β2, β3] of the duration proportions of the yellowing period, the color fixing period, and the tendon drying period according to the second mean resampling baking curve after the smoothing process; β1 is the duration proportion of the yellowing period, β2 is the duration proportion of the color fixing period, and β3 is the duration proportion of the tendon drying period; S1040, if β1 belongs to the preset yellowing period duration ratio interval, β2 belongs to the preset coloring period duration ratio interval, and β3 belongs to the preset dry period duration ratio interval, then it is judged that there is no abnormality in the baking process corresponding to the target baking curve; otherwise, it is judged that there is an abnormality in the baking process corresponding to the target baking curve.
6. The method for processing a baking curve according to claim 1, characterized in that: After S500, the method further includes: S510, obtaining a third mean resampling baking curve according to the effective baking curve, wherein the third mean resampling baking curve is a baking curve obtained by sampling the dry-bulb temperature with a sampling period of 30 minutes; S520, simplifying the third mean resampling baking curve to obtain a target simplified baking curve.
7. The method for processing a baking curve according to claim 6, characterized in that: S520 includes: S521, connecting the first data point and the last data point of the third mean resampling baking curve to obtain an initial straight line L0; S522, if the data point pd on the third mean resampling baking curve with the largest distance from L0 max,0 If the distance to L0 is greater than or equal to the target distance threshold d0, then proceed to S523; S523, with pd max,0 The third mean resampling baking curve is divided for the first time as the curve dividing point to obtain a partial curve set H1 after the first division; the third variable f is initialized to 1; S524, obtaining the partial curve set H after the f-th division f The number of partial curves included qua f ; S525, if qua f +1 < num1, then update f to f + 1, repeat S524 until num1 ≤ qua + 1 ≤ num2 or qua + 1 > num2 or qua f = qua f-1 ; num1 is the lower limit value of the preset target quantity range, num2 is the upper limit value of the preset target quantity range; if num1 ≤ qua + 1 ≤ num2, then enter S526; if qua + 1 > num2, then enter S527; S526, H f The first data point and the last data point corresponding to each part of the curve are determined as the data points corresponding to the target simplified baking curve, and curve fitting is performed according to the data points corresponding to the target simplified baking curve to obtain the target simplified baking curve; S527, obtaining the distance sequence dis after the f-1th division f-1 ,dis f-1 =(dis f-1,1 ,dis f-1,2 ,…,dis f-1,c ,…,dis f-1,v ), dis f-1,c H f-1 The cth curve in the figure is connected to L f-1,c The data point with the largest distance from L f-1,c The distance, L f-1,c is the line connecting the first and last data points of the cth partial curve, where c ranges from 1 to v, and v is H f-1 The number of curves in the middle part; S528, get dis f-1 The largest front num2-qua in the middle distance f-1 The partial curve set cur corresponding to the distance; S529, each cur η Perform a division process and add the first and last data points of the partial curve obtained after the division to U f-1 , and append the U f-1 Determine a set of data points corresponding to the target simplified baking curve, and perform curve fitting according to the set of data points corresponding to the target simplified baking curve to obtain the target simplified baking curve; η is the ηth partial curve in cur, L η For cur η The line connecting the first and last data points of U f-1 H f-1 The set of first and last data points of all included partial curves.
8. The method for processing a baking curve according to claim 7, characterized in that: S522 also includes: if the data point pd on the third mean resampling baking curve with the largest distance from L0 max,0 If the distance to L0 is less than the target distance threshold d0, then enter S5221; S5221, initialize the second variable r to 1; S5222, setting the distance threshold to d0-r×Δd, using the Douglas Peucker algorithm to simplify the third mean resampling baking curve after the r-th update, to obtain the number of data points after the r+1-th simplification; Δd is the preset distance threshold adjustment step size; S5223, if the number of data points after the r+1th simplification is less than num1, update r to r+1, and repeat S5222 until the number of data points after the r+1th simplification is greater than or equal to num1 and less than or equal to num2 or the number of data points after the r+1th simplification is greater than num2; if the number of data points after the r+1th simplification is greater than or equal to num1 and less than or equal to num2, proceed to S5224; S5224, performing curve fitting based on the data points after the r+1th simplification to obtain a target simplified baking curve.
9. The method for processing a baking curve according to claim 8, characterized in that: Δd=0.01.
