Multifunctional integrated intelligent weighing system and method based on artificial intelligence
Through artificial intelligence, analyze object weight data, draw curves and derivative curves, classify objects and calculate range limits, the problem of large errors in existing weighing tools is solved, intelligent range selection and accurate weighing are realized, and the weighing results are ensured. Authenticity and efficiency.
Patent Information
- Application Number
- CN202411270297.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-12-31
AI Technical Summary
Existing weighing tools have problems with ghost scales and fake scales, and the scope of use is limited. Tools with different scales can only be used in limited scenarios, resulting in large weighing errors. How to intelligently judge the appropriate range and reduce errors has become an important issue.
Adopt a multi-functional integrated intelligent weighing system based on artificial intelligence. By collecting historical data, analyzing object weights, drawing curve charts and derivative curves, classifying objects, calculating range division limits and abnormal thresholds, and real-time judgment and selection of suitable ranges for Weigh and reduce errors.
It improves weighing accuracy, reduces instrument errors, ensures the authenticity of weighing results, avoids false measurements, and enhances the system's working efficiency.
Smart Images
Figure CN119223431A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent weighing technology, and in particular to a multifunctional integrated intelligent weighing system and method based on artificial intelligence. Background Art
[0002] With the rapid development of computers and science and technology, weighing objects is no longer done with old tools such as scales, but with various new-age weighing tools, such as smart electronic scales, portable scales, etc. The measurement of object weights has also become more and more intelligent, and weighing technology exists in many fields and industries, such as medical care, transportation, trade, etc. The weight of products is monitored in real time on the production line to ensure that the weight of each product meets the standard and reduce the scrap rate; the goods are automatically weighed at logistics centers or transportation stations to optimize loading plans and improve transportation efficiency; although science and technology have provided more convenient and faster weighing tools in the field of weighing, there are still problems such as ghost scales and fake scales, and the scope of use of weighing tools is relatively limited. Tools with different weighing ranges can only be used in limited scenarios; if tools with inappropriate ranges are used, the weighing error is large; therefore, how to intelligently judge the appropriate range for weighing objects, intelligently select the appropriate range for weighing, and reduce weighing errors; it is crucial to judge the authenticity of the weight of the weighing object. Summary of the invention
[0003] The purpose of the present invention is to provide a multifunctional integrated intelligent weighing system and method based on artificial intelligence to solve the problems raised in the above background technology.
[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: A multifunctional integrated intelligent weighing method based on artificial intelligence, the method comprising the following steps: S100, collecting weight data of objects weighed in history, analyzing all collected weight data, and classifying the weighed objects according to their weights; Furthermore, the specific steps for classifying weighable objects according to their weight are as follows: S101. Collect the weight data of objects in the history when they are weighed. Suppose the weight of each object in the history collected is Indicates collecting the weight of the 1st, 2nd, 3rd, ..., nth object, where n is a positive integer; sorting the weight of each collected object from small to large, using the weight of the sorted objects to plot points on the coordinate axis, customizing the interval m of the plotted points, each of the m horizontal coordinates depicts an object weight point, using the weight of the object as a variable, and the plotted interval m as the horizontal coordinate; drawing a curve graph of the object weight; S102, calculating the derivative curve of the drawn object weight curve, the formula is: ; In the formula, sl represents the derivative of each segment of the object weight curve, and Lim represents the limit m is infinitely close to 0. Indicates the weight of the next point in the weight curve of the object. represents the weight of the previous point in the object weight curve; m represents the weight of the previous point in the object weight curve and Corresponding to the change in the value of the independent variable, ; S103, the slope of each two adjacent points in the object weight curve is calculated, and the slope between all two points in the object weight curve is obtained. Indicates that the slopes between the 1st, 2nd, 3rd, ..., n-1th two points in the object weight curve are calculated; and the object weight derivative curve is drawn using the calculated n-1 slopes; S104, judging the derivative curve graph consisting of the calculated n-1 slopes, when When the weight of the object increases sharply, the slope The corresponding two points are used as the weight classification standard of the object. When the weight of the object increases slightly, The slopes corresponding to the two points are of the same type of weight; the slope judgment is expressed in the derivative curve graph as finding the peak value in the derivative curve graph, judging that the slope at the peak point has a sharp increase in weight, and the corresponding two points are the classification standard; after judgment, the slope with a sharp increase is It represents the 1st, 2nd, 3rd, ..., kth slopes with sharp increases obtained after judgment, and the object weights are classified into k+1 types using the k slopes with sharp increases, and the objects in each object weight are set together to obtain k+1 object sets.
[0005] By analyzing the weight of objects in history, objects can be classified according to their weight. When weighing objects, different judgment criteria can be selected according to the weight of objects of different types to analyze the weight of objects, which greatly speeds up the time of object weight analysis and improves the efficiency of system work. S200, according to the weight data of each weighing object obtained by classification, analyzing the weight difference amplitude of each object to generalize the weight data, calculating the range division limit when weighing the object, and obtaining different ranges for weighing the object; Furthermore, the specific steps of obtaining different ranges for weighing an object are as follows: S201, extract the weight of the objects in each category set after classifying the weight of the objects in S104 Represents the weight of the 1st, 2nd, 3rd, ..., hth object in each object type set, where h is a positive integer; calculates the average difference in the weight of each object in the set, using the formula: ; In the formula, Am represents the average difference in the weights of h objects in the set. Indicates the weight of the last object in each object weight set. represents the weight of the previous object in each object weight set; h-1 represents the number of calculated object weight differences; S202, taking the calculated average difference in the weight of each set of objects as the change range of the weight of the objects, and calculating the change ranges of the weight of the objects in the k+1 sets to obtain k+1 change ranges of represents the calculated change in the weight of the object in the 1st, 2nd, 3rd, ..., k+1th object weight set; S203: Find the maximum value of the weight of each object in the weight set in advance , according to the calculated weight variation of each object, the weight data is generalized and the range division limit when the object is weighed is calculated. The formula is: ; In the formula, L represents the range division limit when the object is weighed. Indicates the maximum value of each object weight set extracted, Am indicates the variation range of each object weight set, and bit indicates The unit digit of the calculated value; the minimum value of the object weight in the collected history is , calculate the minimum boundary of the range when weighing an object, the formula is: ; In the formula, Indicates the minimum range boundary when weighing the object being calculated. Indicates the minimum object weight in the collected history, Am indicates the change in the weight set of each object, and bit indicates The unit digit of the calculated value; S204, calculating the maximum weight of the objects in the extracted k+1 sets, and obtaining the range division limit values from small to large when weighing the objects. Indicates the calculated 1st, 2nd, 3rd, ..., k+1th range division limits; according to the minimum range boundary and the division limit, the different ranges for weighing the object are obtained from small to large. Indicates the calculated 1st, 2nd, 3rd, ..., k+1th range when weighing an object.
[0006] The weight data in each object weight set is used for analysis, and the variation range of the object weights in each object weight set is calculated. The maximum weight value in each object weight set is generalized using the variation range, and an integer multiple of 10 is taken to calculate the range division limit to obtain different ranges. Using different ranges can make the weighing results more accurate when weighing objects and reduce inaccurate weighing caused by instrument errors.
[0007] S300, collecting data on abnormal weights of objects when weighing objects in history, and calculating an abnormal threshold value for the weight of each object; Furthermore, the specific steps for calculating the abnormal threshold of each object weight are as follows: S301, collecting the data of abnormal weight when weighing each object in the history, assuming that the collected data of abnormal weight when weighing each object is , It means collecting the 1st, 2nd, 3rd, ..., pth data of abnormal weight when weighing each object in the history, where p is a positive integer; the volume of each object with abnormal weight is collected respectively represents the volume of the 1st, 2nd, 3rd, ..., pth objects with abnormal weight collected; S302, using the data of weight anomalies when weighing each object in the collected history, calculate the abnormal threshold of the weight of each object, the formula is: ; In the formula, Z_th represents the calculated abnormal threshold when weighing an object, Z represents the data of abnormal weight when weighing each object in the collected history, Z_st represents the standard deviation of the data of abnormal weight when weighing each object in the collected history, and p represents the number of data of abnormal weight when weighing each object in the collected history; S303: Using the formula in S302, the abnormal threshold of the weight of k+1 objects is: Indicates the calculated abnormal thresholds of the weights of the 1st, 2nd, 3rd, ..., k+1th objects.
[0008] Analyze the abnormal weight data of objects in historical weighing, calculate the abnormal threshold of each range, and use the abnormal threshold to judge and monitor the objects weighed in real time to ensure the real weight of the objects is obtained during weighing and avoid false weight values due to unexpected situations. S400, when weighing an object in real time, using the division limits of different ranges to judge the weight of the object being weighed in real time and select a weighing range; Furthermore, the specific steps of determining the weight of the object being weighed in real time and selecting the weighing range are as follows: S401, when weighing an object in real time, extract the weight of the object in real time as W_s, and use the division limit values of different ranges when weighing the object in S200 to judge the weight of the object being weighed in real time. When the weight of the object is determined in real time, the range is When the weight of the object is determined in real time, the range is ; Indicates the division limits of different ranges when weighing objects. .
[0009] In the existing technology, the automatic adjustment of the measuring range is to first select fixed sensors of different measuring ranges, for example, the measuring ranges are 0-10, 10-30, 30-100, etc.; these measuring ranges are selected outside the system operation and may not fit the system working environment; and only a measuring range including the weight of the object is rigidly selected after the object is preliminarily weighed. The present invention obtains a measuring range that actually fits the weight of the object in the system working environment by analyzing the weight of the object in the history, which can make the weighing of the object more accurate and fit the system operation compared with the prior art.
[0010] S500: Use the weighing threshold of each weight object to judge whether there is an abnormal risk in the weight of the weighing object and issue an early warning.
[0011] Furthermore, the specific steps of weighing the object to determine whether there is an abnormal risk and issuing an early warning are: S501, after judging the range required for the weight of the object to be weighed in real time by the method described in S400, the range required for the weight of the object to be weighed in real time is obtained. ; According to the range required for real-time weighing of the object weight, select the corresponding abnormal threshold value for weighing , ; S502, extract the weight W_s and volume V_s of the real-time weighing object, and calculate the real-time abnormality index, the formula is: ,when When the weight of the object being weighed in real time is judged to be abnormal, an abnormal warning is issued; When the weight of the object being weighed in real time is judged to have no abnormal risk.
[0012] A multifunctional integrated intelligent weighing system based on artificial intelligence, which includes a data collection module, an object weight classification module, a weighing range division module, an abnormal threshold calculation module, a real-time range judgment module and a real-time abnormal judgment module; The data collection module is used to collect weight data of objects weighed in history; The object weight classification module is used to collect the weights of objects weighed in history, draw a curve graph, calculate the derivative curve of the curve graph, and classify the weights of objects according to the increase in the weight of the objects in history; The weighing range division module is used to analyze the weight data in each object weight set after classification, calculate the limit value of the range division, and generalize the object weight to obtain different ranges when weighing the object; The abnormal threshold calculation module is used to calculate the abnormal threshold in each weighing range by using the data of weight abnormalities that occurred when weighing objects in the history; The real-time range determination module is used to extract the weight of the object being weighed in real time, and to determine the range to be used for weighing the object in real time by using the division limits of different ranges; The real-time abnormality judgment module is used to extract the corresponding abnormal threshold according to the range required for the real-time weighing object to be weighed, and use the abnormal threshold to judge whether there is an abnormal risk for the weight of the real-time weighing object and issue an early warning.
[0013] The object weight classification module includes an object weight curve unit and a derivation unit; The object weight curve unit is used to collect the object weights when the objects are weighed in the history, sort the weights of each object collected from small to large, use the sorted object weights to draw points on the coordinate axis, customize the interval m of the points, each m horizontal coordinates depicts an object weight point, use the weight of the object as a variable, and the interval m of the points is the horizontal coordinate; draw a curve graph of the object weight; The derivation unit is used to perform derivation calculation between every two points in the object weight curve diagram, calculate the slope between every two points, judge all the calculated slopes, and use the slope greater than the average value as the standard for classifying the object weight.
[0014] The weighing range division module includes a limit value calculation unit and a range division unit; The limit value calculation unit is used to calculate the average value of the amplitude change of each object weight, extract the maximum value of each object weight, generalize the maximum value and take an integer multiple of 10, and calculate different limit values for dividing the range; The range division unit is used to divide different ranges of weighing according to the calculated division limit to obtain k+1 different ranges.
[0015] The abnormal judgment module is used to determine the range required for real-time weighing of the object, select the abnormal threshold corresponding to the range, and use the abnormal threshold to determine whether there is an abnormal risk in the weight of the real-time weighing object and issue an early warning.
[0016] Compared with the prior art, the beneficial effects achieved by the present invention are: 1. The present invention classifies the weight of objects, so that when a series of weight analyses are performed on objects, a wide range of judgment and analysis can be performed according to the types of object weights, without having to analyze the weights of objects one by one, which greatly reduces the amount of calculation of the system during analysis and enhances the working efficiency of the system.
[0017] 2. The present invention calculates different ranges for weighing different objects, and selects a range suitable for the weight of the object according to the different weights of the object for flexibly weighing, thereby reducing the weighing error caused by the error of the instrument system and making the weight obtained by the system for weighing the object more accurate.
[0018] 3. The present invention calculates the abnormal threshold value of each range, and uses the abnormal threshold value to judge and monitor the objects being weighed in real time, so as to ensure that the real weight of the object is obtained during weighing, and avoid false weight values measured due to abnormal conditions, which will cause unnecessary economic losses. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is a module distribution diagram of the multifunctional integrated intelligent weighing system based on artificial intelligence of the present invention; Figure 2 It is a schematic diagram of the steps of the multifunctional integrated intelligent weighing method based on artificial intelligence of the present invention; Figure 3 It is a graph of object weight of the multifunctional integrated intelligent weighing method based on artificial intelligence of the present invention; Figure 4 It is a derivative curve diagram of the multifunctional integrated intelligent weighing method based on artificial intelligence of the present invention. DETAILED DESCRIPTION
[0020] 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0021] See also Figure 1-Figure 4 , the present invention provides a technical solution: A multifunctional integrated intelligent weighing method based on artificial intelligence, the method comprising the following steps: S100, collecting weight data of objects weighed in history, analyzing all collected weight data, and classifying the weighed objects according to their weights; The specific steps for classifying weighable objects according to weight are: S101. Collect the weight data of objects in the history when they are weighed. Suppose the weight of each object in the history collected is Indicates collecting the weight of the 1st, 2nd, 3rd, ..., nth object, where n is a positive integer; sorting the weight of each collected object from small to large, using the weight of the sorted objects to plot points on the coordinate axis, customizing the interval m of the plotted points, each of the m horizontal coordinates depicts an object weight point, using the weight of the object as a variable, and the plotted interval m as the horizontal coordinate; drawing a curve graph of the object weight; S102, calculating the derivative curve of the drawn object weight curve, the formula is: ; In the formula, sl represents the derivative of each segment of the object weight curve, and Lim represents the limit m is infinitely close to 0. Indicates the weight of the next point in the weight curve of the object. represents the weight of the previous point in the object weight curve; m represents the weight of the previous point in the object weight curve and Corresponding to the change in the value of the independent variable, ; S103, the slope of each two adjacent points in the object weight curve is calculated, and the slope between all two points in the object weight curve is obtained. Indicates that the slopes between the 1st, 2nd, 3rd, ..., n-1th two points in the object weight curve are calculated; and the object weight derivative curve is drawn using the calculated n-1 slopes; S104, judging the derivative curve graph consisting of the calculated n-1 slopes, when When the weight of the object increases sharply, the slope The corresponding two points are used as the weight classification standard of the object. When the weight of the object increases slightly, The slopes corresponding to the two points are of the same type of weight; the slope judgment is expressed in the derivative curve graph as finding the peak value in the derivative curve graph, judging that the slope at the peak point has a sharp increase in weight, and the corresponding two points are the classification standard; after judgment, the slope with a sharp increase is It represents the 1st, 2nd, 3rd, ..., kth slopes with sharp increases obtained after judgment, and the object weights are classified into k+1 types using the k slopes with sharp increases, and the objects in each object weight are set together to obtain k+1 object sets.
[0022] By analyzing the weight of objects in history, objects can be classified according to their weight. When weighing objects, different judgment criteria can be selected according to the weight of objects of different types to analyze the weight of objects, which greatly speeds up the time of object weight analysis and improves the efficiency of system work. S200, according to the weight data of each weighing object obtained by classification, analyzing the weight difference amplitude of each object to generalize the weight data, calculating the range division limit when weighing the object, and obtaining different ranges for weighing the object; The specific steps to obtain different ranges for weighing an object are: S201, extract the weight of the objects in each category set after classifying the weight of the objects in S104 Represents the weight of the 1st, 2nd, 3rd, ..., hth object in each object type set, where h is a positive integer; calculates the average difference in the weight of each object in the set, using the formula: ; In the formula, Am represents the average difference in the weights of h objects in the set. Indicates the weight of the last object in each object weight set. represents the weight of the previous object in each object weight set; h-1 represents the number of calculated object weight differences; S202, taking the calculated average difference in the weight of each set of objects as the change range of the weight of the objects, and calculating the change ranges of the weight of the objects in the k+1 sets to obtain k+1 change ranges of represents the calculated change in the weight of the object in the 1st, 2nd, 3rd, ..., k+1th object weight set; S203: Find the maximum value of the weight of each object in the weight set in advance , according to the calculated weight variation of each object, the weight data is generalized and the range division limit when the object is weighed is calculated. The formula is: ; In the formula, L represents the range division limit when the object is weighed. Indicates the maximum value of each object weight set extracted, Am indicates the variation range of each object weight set, and bit indicates The unit digit of the calculated value; the minimum value of the object weight in the collected history is , calculate the minimum boundary of the range when weighing an object, the formula is: ; In the formula, Indicates the minimum range boundary when weighing the object being calculated. Indicates the minimum object weight in the collected history, Am indicates the change in the weight set of each object, and bit indicates The unit digit of the calculated value; S204, calculating the maximum weight of the objects in the extracted k+1 sets, and obtaining the range division limit values from small to large when weighing the objects. Indicates the calculated 1st, 2nd, 3rd, ..., k+1th range division limits; according to the minimum range boundary and the division limit, the different ranges for weighing the object are obtained from small to large. Indicates the calculated 1st, 2nd, 3rd, ..., k+1th range when weighing an object.
[0023] The weight data in each object weight set is used for analysis, and the variation range of the object weights in each object weight set is calculated. The maximum weight value in each object weight set is generalized using the variation range, and an integer multiple of 10 is taken to calculate the range division limit to obtain different ranges. Using different ranges can make the weighing results more accurate when weighing objects and reduce inaccurate weighing caused by instrument errors.
[0024] S300, collecting data on abnormal weights of objects when weighing objects in history, and calculating an abnormal threshold value for the weight of each object; The specific steps to calculate the abnormal threshold of each object weight are: S301, collecting the data of abnormal weight when weighing each object in the history, assuming that the collected data of abnormal weight when weighing each object is , It means collecting the 1st, 2nd, 3rd, ..., pth data of abnormal weight when weighing each object in the history, where p is a positive integer; the volume of each object with abnormal weight is collected respectively represents the volume of the 1st, 2nd, 3rd, ..., pth objects with abnormal weight collected; S302, using the data of weight anomalies when weighing each object in the collected history, calculate the abnormal threshold of the weight of each object, the formula is: ; In the formula, Z_th represents the calculated abnormal threshold when weighing an object, Z represents the data of abnormal weight when weighing each object in the collected history, Z_st represents the standard deviation of the data of abnormal weight when weighing each object in the collected history, and p represents the number of data of abnormal weight when weighing each object in the collected history; S303: Using the formula in S302, the abnormal threshold of the weight of k+1 objects is: Indicates the calculated abnormal thresholds of the weights of the 1st, 2nd, 3rd, ..., k+1th objects.
[0025] Analyze the abnormal weight data of objects in historical weighing, calculate the abnormal threshold of each range, and use the abnormal threshold to judge and monitor the objects weighed in real time to ensure the real weight of the objects is obtained during weighing and avoid false weight values due to unexpected situations. S400, when weighing an object in real time, using the division limits of different ranges to judge the weight of the object being weighed in real time and select a weighing range; The specific steps for judging the weight of an object to be weighed in real time and selecting the weighing range are as follows: S401, when weighing an object in real time, extract the weight of the object in real time as W_s, and use the division limit values of different ranges when weighing the object in S200 to judge the weight of the object being weighed in real time. When the weight of the object is determined in real time, the range is When the weight of the object is determined in real time, the range is ; Indicates the division limits of different ranges when weighing objects. .
[0026] In the existing technology, the automatic adjustment of the measuring range is to first select fixed sensors of different measuring ranges, for example, the measuring ranges are 0-10, 10-30, 30-100, etc.; these measuring ranges are selected outside the system operation and may not fit the system working environment; and only a measuring range including the weight of the object is rigidly selected after the object is preliminarily weighed. The present invention obtains a measuring range that actually fits the weight of the object in the system working environment by analyzing the weight of the object in the history, which can make the weighing of the object more accurate and fit the system operation compared with the prior art.
[0027] S500: Use the weighing threshold of each weight object to judge whether there is an abnormal risk in the weight of the weighing object and issue an early warning.
[0028] The specific steps to determine whether there is an abnormal risk in the weight of an object and issue an early warning are: S501, after judging the range required for the weight of the object to be weighed in real time by the method described in S400, the range required for the weight of the object to be weighed in real time is obtained. ; According to the range required for real-time weighing of the object weight, select the corresponding abnormal threshold value for weighing , ; S502, extract the weight W_s and volume V_s of the real-time weighing object, and calculate the real-time abnormality index, the formula is: ,when When the weight of the object being weighed in real time is judged to be abnormal, an abnormal warning is issued; When the weight of the object being weighed in real time is judged to have no abnormal risk.
[0029] A multifunctional integrated intelligent weighing system based on artificial intelligence, which includes a data collection module, an object weight classification module, a weighing range division module, an abnormal threshold calculation module, a real-time range judgment module and a real-time abnormal judgment module; The data collection module is used to collect weight data of objects weighed in history; The object weight classification module is used to collect the weights of objects weighed in history, draw a curve graph, calculate the derivative curve of the curve graph, and classify the weights of objects according to the increase in the weight of the objects in history; The weighing range division module is used to analyze the weight data in each object weight set after classification, calculate the limit value of the range division, and generalize the object weight to obtain different ranges when weighing the object; The abnormal threshold calculation module is used to calculate the abnormal threshold in each weighing range by using the data of weight abnormalities that occurred when weighing objects in the history; The real-time range determination module is used to extract the weight of the object being weighed in real time, and to determine the range to be used for weighing the object in real time by using the division limits of different ranges; The real-time abnormality judgment module is used to extract the corresponding abnormal threshold according to the range required for the real-time weighing object to be weighed, and use the abnormal threshold to judge whether there is an abnormal risk for the weight of the real-time weighing object and issue an early warning.
[0030] The object weight classification module includes an object weight curve unit and a derivation unit; The object weight curve unit is used to collect the object weights when the objects are weighed in the history, sort the weights of each object collected from small to large, use the sorted object weights to draw points on the coordinate axis, customize the interval m of the points, each m horizontal coordinates depicts an object weight point, use the weight of the object as a variable, and the interval m of the points is the horizontal coordinate; draw a curve graph of the object weight; The derivation unit is used to perform derivation calculation between every two points in the object weight curve diagram, calculate the slope between every two points, judge all the calculated slopes, and use the slope greater than the average value as the standard for classifying the object weight.
[0031] The weighing range division module includes a limit value calculation unit and a range division unit; The limit value calculation unit is used to calculate the average value of the amplitude change of each object weight, extract the maximum value of each object weight, generalize the maximum value and take an integer multiple of 10, and calculate different limit values for dividing the range; The range division unit is used to divide different ranges of weighing according to the calculated division limit to obtain k+1 different ranges.
[0032] The abnormal judgment module is used to determine the range required for real-time weighing of the object, select the abnormal threshold corresponding to the range, and use the abnormal threshold to determine whether there is an abnormal risk in the weight of the real-time weighing object and issue an early warning.
[0033] Example 1: The weight data of objects weighed in the history are collected as {24, 43, 132, 145, 511, 521}, and the value of m is set to draw a curve graph as shown in Figure 3 ; Calculate the slope between every two points and draw the derivative curve. The formula is: ; The slopes obtained by calculation are {19, 89, 13, 366, 10}. After judgment, the standard for classifying the weight of objects is the weight of two objects corresponding to 89 and 366. Finally, the weights of the collected objects are divided into three types: (24, 43), (132, 145), and (511, 521). Using the formula The average difference of the weight of each object is calculated as follows: 19, 13, and 10. Then the maximum value is extracted and the weight of the object is generalized using the formula to calculate the limit values of different ranges. The formula is: ; The calculation results show that L_min=0, L1=70, L2=160, L3=540; the different ranges when weighing objects are obtained by using the limit value are (0, 70), (70, 160), (160, 540); the data of weight anomalies in each range in history are collected, and the abnormal threshold of each range is calculated. The formula is: ; The calculated abnormal thresholds of the three ranges are 10, 20, and 25 respectively; There is an object being weighed in real time. The weight of the object is 68 and the volume is 5. After judgment, it is found that the range to be used is (70, 160). The calculated abnormal index is 13.6. It is judged that there is a risk of abnormal weight of the object in real time, and an early warning is issued.
[0034] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0035] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A multifunctional integrated intelligent weighing method based on artificial intelligence, characterized by: The method comprises the following steps: S100, collecting weight data of objects weighed in history, analyzing all collected weight data, and classifying the weighed objects according to their weights; S200, according to the weight data of each weighing object obtained by classification, analyzing the weight difference amplitude of each object to generalize the weight data, calculating the range division limit when weighing the object, and obtaining different ranges for weighing the object; S300, collecting data on abnormal weights of objects when weighing objects in history, and calculating an abnormal threshold value for each weight of the objects; S400, when weighing an object in real time, using the division limits of different ranges to judge the weight of the object being weighed in real time and select a weighing range; S500: Use the weighing threshold of each weight object to judge whether there is an abnormal risk in the weight of the weighing object and issue an early warning.
2. The multifunctional integrated intelligent weighing method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of classifying weighable objects according to weight in S100 are: S101. Collect the weight data of objects in the history when they are weighed. Suppose the weight of each object in the history collected is Indicates collecting the weight of the 1st, 2nd, 3rd, ..., nth object, where n is a positive integer; sorting the weight of each collected object from small to large, using the weight of the sorted objects to plot points on the coordinate axis, and defining a custom plot interval m. Each of the m horizontal coordinates depicts an object weight point, and the weight of the object is used as a variable, and the plot interval m is used as the horizontal coordinate; Plot a graph of the weight of an object; S102, calculating the derivative curve of the drawn object weight curve, the formula is: ； In the formula, sl represents the derivative of each segment of the object weight curve, and Lim represents the limit m is infinitely close to 0. Indicates the weight of the next point in the weight curve of the object. represents the weight of the previous point in the object weight curve; m represents the weight of the previous point in the object weight curve and Corresponding to the change in the value of the independent variable, ; S103, the slope of each two adjacent points in the object weight curve is calculated, and the slope between all two points in the object weight curve is obtained. Indicates that the slopes between the 1st, 2nd, 3rd, ..., n-1th two points in the object weight curve are calculated; and the object weight derivative curve is drawn using the calculated n-1 slopes; S104, judging the derivative curve graph consisting of the calculated n-1 slopes, when When the weight of the object increases sharply, the slope The corresponding two points are used as the weight classification standard of the object. When the weight of the object increases slightly, The slope corresponds to two points with the same type of weight; After judging, the slope with a sharp increase is It represents the 1st, 2nd, 3rd, ..., kth slopes with sharp increases obtained after judgment, and the object weights are classified into k+1 types using the k slopes with sharp increases, and the objects in each object weight are set together to obtain k+1 object sets.
3. The multifunctional integrated intelligent weighing method based on artificial intelligence according to claim 2 is characterized in that: The specific steps of obtaining different ranges of weighing the object in S200 are: S201, extract the weight of the objects in each category set after classifying the weight of the objects in S104 Represents the weight of the 1st, 2nd, 3rd, ..., hth object in each object type set, where h is a positive integer; calculates the average difference in the weight of each object in the set, using the formula: ； In the formula, Am represents the average difference in the weights of h objects in the set. Indicates the weight of the last object in each object weight set. represents the weight of the previous object in each object weight set; h-1 represents the number of calculated object weight differences; S202, taking the calculated average difference in the weight of each set of objects as the change range of the weight of the objects, and calculating the change ranges of the weight of the objects in the k+1 sets to obtain k+1 change ranges of represents the calculated change in the weight of the object in the 1st, 2nd, 3rd, ..., k+1th object weight set; S203: Find the maximum value of the weight of each object in the weight set in advance , according to the calculated weight variation of each object, the weight data is generalized and the range division limit is calculated when the object is weighed. The formula is: ； In the formula, L represents the range division limit when the object is weighed. Indicates the maximum value of each object weight set extracted, Am indicates the variation range of each object weight set, and bit indicates The unit digit of the calculated value; the minimum value of the object weight in the collected history is , calculate the minimum boundary of the range when weighing an object, the formula is: ； In the formula, Indicates the minimum range boundary when weighing the object being calculated. Indicates the minimum object weight in the collected history, Am indicates the change in the weight set of each object, and bit indicates The unit digit of the calculated value; S204, calculating the maximum weight of the objects in the extracted k+1 sets, and obtaining the range division limit values from small to large when weighing the objects. Indicates the calculated 1st, 2nd, 3rd, ..., k+1th range division limits; according to the minimum range boundary and the division limit, the different ranges for weighing the object are obtained from small to large. Indicates the calculated 1st, 2nd, 3rd, ..., k+1th range when weighing an object.
4. The multifunctional integrated intelligent weighing method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of calculating the abnormal threshold value of each object weight in S300 are: S301, collecting the data of abnormal weight when weighing each object in the history, assuming that the collected data of abnormal weight when weighing each object is , It means collecting the 1st, 2nd, 3rd, ..., pth data of abnormal weight when weighing each object in the history, where p is a positive integer; the volume of each object with abnormal weight is collected respectively represents the volume of the 1st, 2nd, 3rd, ..., pth objects with abnormal weight collected; S302, using the data of weight anomalies when weighing each object in the collected history, calculate the abnormal threshold of the weight of each object, the formula is: ； In the formula, Z_th represents the calculated abnormal threshold when weighing an object, Z represents the data of abnormal weight when weighing each object in the collected history, Z_st represents the standard deviation of the data of abnormal weight when weighing each object in the collected history, and p represents the number of data of abnormal weight when weighing each object in the collected history; S303: Using the formula in S302, the abnormal threshold of the weight of k+1 objects is: Indicates the calculated abnormal thresholds of the weights of the 1st, 2nd, 3rd, ..., k+1th objects.
5. The multifunctional integrated intelligent weighing method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of determining the weight of the object to be weighed in real time and selecting the weighing range in S400 are: S401, when weighing an object in real time, extract the weight of the object in real time as W_s, and use the division limit values of different ranges when weighing the object in S200 to judge the weight of the object being weighed in real time. When the weight of the object is determined in real time, the range is When the weight of the object is determined in real time, the range is ; Indicates the division limits of different ranges when weighing objects. .
6. The multifunctional integrated intelligent weighing method based on artificial intelligence according to claim 5 is characterized in that: The specific steps of determining whether there is an abnormal risk in the weight of the object and issuing an early warning in S500 are: S501, after judging the range required for the weight of the object to be weighed in real time by the method described in S400, the range required for the weight of the object to be weighed in real time is obtained. ; According to the range required for real-time weighing of the object weight, select the corresponding abnormal threshold value for weighing , ; S502, extract the weight W_s and volume V_s of the real-time weighing object, and calculate the real-time abnormality index, the formula is: ,when When the weight of the object being weighed in real time is judged to be abnormal, an abnormal warning is issued; When the weight of the object being weighed in real time is judged to have no abnormal risk.
7. The multifunctional integrated intelligent weighing system based on artificial intelligence is characterized by: The multifunctional integrated intelligent weighing system includes a data collection module, an object weight classification module, a weighing range division module, an abnormal threshold calculation module, a real-time range judgment module and a real-time abnormal judgment module; The data collection module is used to collect weight data of objects weighed in history; The object weight classification module is used to collect the weights of objects weighed in history, draw a curve graph, calculate the derivative curve of the curve graph, and classify the weights of objects according to the increase in the weight of the objects in history; The weighing range division module is used to analyze the weight data in each object weight set after classification, calculate the limit value of the range division, and generalize the object weight to obtain different ranges when weighing the object; The abnormal threshold calculation module is used to calculate the abnormal threshold in each weighing range by using the data of weight abnormalities that occurred when weighing objects in the history; The real-time range determination module is used to extract the weight of the object being weighed in real time, and to determine the range to be used for weighing the object in real time by using the division limits of different ranges; The real-time abnormality judgment module is used to extract the corresponding abnormal threshold according to the range required for the real-time weighing object to be weighed, and use the abnormal threshold to judge whether there is an abnormal risk for the weight of the real-time weighing object and issue an early warning.
8. The multifunctional integrated intelligent weighing system based on artificial intelligence according to claim 7 is characterized in that: The object weight classification module includes an object weight curve unit and a derivation unit; The object weight curve unit is used to collect the object weights when the objects are weighed in the history, sort the weights of each object collected from small to large, use the sorted object weights to draw points on the coordinate axis, customize the interval m of the points, each m horizontal coordinates depicts an object weight point, and use the weight of the object as a variable, and the interval m of the points is the horizontal coordinate; Plot a graph of the weight of an object; The derivation unit is used to perform derivation calculation between every two points in the object weight curve diagram, calculate the slope between every two points, judge all the calculated slopes, and use the slope greater than the average value as the standard for classifying the object weight.
9. The multifunctional integrated intelligent weighing system based on artificial intelligence according to claim 7 is characterized in that: The weighing range division module includes a limit value calculation unit and a range division unit; The limit value calculation unit is used to calculate the average value of the amplitude change of each object weight, extract the maximum value of each object weight, generalize the maximum value and take an integer multiple of 10, and calculate different limit values for dividing the range; The range division unit is used to divide different ranges of weighing according to the calculated division limit to obtain k+1 different ranges.
10. The multifunctional integrated intelligent weighing system based on artificial intelligence according to claim 7, characterized in that: The abnormality judgment module is used to select an abnormal threshold corresponding to the range after judging the range required for real-time weighing of the object, and use the abnormal threshold to judge the weight of the real-time weighing object to determine whether there is an abnormal risk in the weight of the real-time weighing object and issue an early warning.