Weighing monitoring and early warning method and device for smart scales

Abstract

The present invention discloses a weighing monitoring and early warning method and device for smart scales, which relates to the technical field related to weighing monitoring. The method includes: activating a smart scale detection program to perform weighing record calculations on a target smart scale to determine an initial weighing coefficient; building a weighing scenario test parameter table to monitor and obtain a real-time weight AD value set, and record the actual weight information of the object; performing partition regression fitting to generate a weighing nonlinear model; comparing and calculating the dynamic weighing coefficient set with the initial weighing coefficient in sequence to determine a weighing coefficient offset slope set; constructing a weighing dynamic early warning mechanism to determine a weighing early warning strategy, and performing weighing early warning control on the target smart scale. The method solves the technical problem in the prior art that the weighing of smart scales is easily affected by factors such as the environment and long-term use, resulting in the inability to timely and accurately monitor weighing deviations and early warning lags, thereby achieving the technical effect of improving the accuracy of smart scale weighing and the timeliness of monitoring and early warning.

Description

Technical Field

[0001] The present application relates to the technical field related to weighing monitoring, and specifically to a weighing monitoring and early warning method and device for a smart scale. Background Art

[0002] Smart scales are increasingly being used in various fields. They are essential equipment in areas such as smart homes, health management, and logistics warehousing. Their accuracy and reliability directly impact user experience and business efficiency. From material weighing in industrial production to commodity pricing in retail, to weight monitoring in healthcare, smart scales are essential for accurate services. Traditional weighing equipment typically relies on fixed calibration parameters and linear models for weight measurement. However, in actual use, smart scales face numerous challenges in achieving accurate weighing results. Environmental factors such as temperature and humidity fluctuations, component wear caused by long-term use, and complex weighing scenarios can all lead to deviations in smart scale weighing results. Traditional weighing monitoring methods often fail to detect these deviations promptly and accurately, resulting in inaccurate weighing data and unable to effectively address errors caused by sensor nonlinearity. This can impact production processes, transaction fairness, and the reliability of health monitoring. For example, this can lead to material mismatching errors in industrial production, affecting product quality; cause disputes between consumers and businesses in commercial transactions; and even mislead patients' health assessments in the medical field.

[0003] Therefore, in the current relevant technologies, there is a technical problem that the weighing of smart scales is easily affected by factors such as the environment and long-term use, resulting in the inability to timely and accurately monitor weighing deviations and early warning lags. Summary of the Invention

[0004] This application solves the technical problems in the prior art that the weighing of smart scales is easily affected by factors such as the environment and long-term use, resulting in the inability to timely and accurately monitor weighing deviations and early warning lags, by providing a weighing monitoring and early warning method and device for smart scales. This achieves the technical effect of improving the accuracy of smart scale weighing and the timeliness of monitoring and early warning.

[0005] The present application provides a weighing monitoring and early warning method for a smart scale, the method comprising: placing a target smart scale on a detection platform, activating a smart scale detection program to perform weighing record calculations on the target smart scale, determining an initial weighing coefficient, and performing a dual backup of the initial weighing coefficient, the dual backup including local storage and cloud backup; establishing a weighing scenario test parameter table, acquiring a real-time weight AD value set through multiple monitoring of the target smart scale, and recording actual object weight information; performing partitioned regression fitting based on the actual object weight information and the real-time weight AD value set to generate a weighing nonlinear model, and obtaining a dynamic weighing coefficient set based on the weighing nonlinear model; sequentially comparing and calculating the dynamic weighing coefficient set with the initial weighing coefficient to determine a weighing coefficient offset slope set; constructing a weighing dynamic early warning mechanism, triggering the weighing dynamic early warning mechanism to perform classification matching based on the weighing coefficient offset slope set, determining a weighing early warning strategy, and performing weighing early warning control on the target smart scale through the weighing early warning strategy.

[0006] In a possible implementation, the weighing monitoring and early warning method for a smart scale further performs the following processing: based on the smart scale detection program, starting a smart scale detection module set, the smart scale detection module set including a calibration object placement module, a weight recording module and a coefficient calculation module; numbering the standard weight set to obtain calibration number information, and placing the standard weight set on the target smart scale in sequence according to the calibration number information through the calibration object placement module; based on the weight recording module, recording in sequence the actual weight set and the displayed weight AD value set of the standard weight set; performing regression fitting calculation on the actual weight set and the displayed weight AD value set through the coefficient calculation module to determine the initial weighing coefficient.

[0007] In a possible implementation, the weighing monitoring and early warning method for a smart scale further performs the following processing: obtaining the smart scale application target, setting the test object weight threshold and test environment information according to the smart scale application target and the specification attributes of the target smart scale; determining the weight interval density according to the weighing accuracy requirements, performing multiple selections within the object weight threshold according to the weight interval density, and determining the object test weight set; extracting factors from the test environment information to obtain a test environment associated factor set; performing parameter design on the test environment associated factor set based on the smart scale application target to obtain a test environment factor parameter set; and integrating the object test weight set and the test environment factor parameter set to build the weighing scenario test parameter table.

[0008] In a possible implementation, the weighing monitoring and early warning method for a smart scale further performs the following processing: calculating and obtaining the measured weight difference information of the actual weight information of the object and the real-time weight AD value set; performing threshold partitioning on each scene parameter in the weighing scene test parameter table based on the measured weight difference information to obtain a weighing scene parameter partition threshold; performing partition identification on the actual weight information of the object and the real-time weight AD value set according to the weighing scene parameter partition threshold to obtain a partitioned measured weight data set; performing segmented regression fitting and combined verification and tuning on the partitioned measured weight data set in turn to generate a weighing nonlinear model.

[0009] In a possible implementation, the weighing monitoring and early warning method for a smart scale further performs the following processing: calculating the mean and standard deviation of the measured weight difference information, and statistically analyzing the calculation results to obtain difference data distribution information; selecting a target clustering algorithm based on the difference data distribution information, and using the target clustering algorithm to perform cluster analysis on the measured weight difference information to obtain difference data clustering results; determining a weight difference data cluster based on the difference data clustering results; and performing threshold matching partitioning on each scene parameter in the weighing scene test parameter table based on the weight difference data cluster to obtain the weighing scene parameter partition threshold.

[0010] In a possible implementation, the weighing monitoring and early warning method for a smart scale also performs the following processing: linear regression fitting is performed on the partitioned measured weight data set in sequence to obtain a partitioned weighing coefficient regression model set; the partitioned weighing coefficient regression model set is sequentially combined according to the partition threshold of the weighing scenario parameter to obtain an initial weighing coefficient model; the initial weighing coefficient model is verified and evaluated to obtain a model partition determination coefficient; the partition threshold of the initial weighing coefficient model is optimized based on the model partition determination coefficient to generate the weighing nonlinear model.

[0011] In a possible implementation, the weighing monitoring and early warning method for a smart scale also performs the following processing: performing partition early warning analysis on the weighing coefficient offset slope set according to the partition threshold of the weighing scene parameter, and setting the partition offset slope early warning level; determining the partition weighing early warning strategy based on the partition offset slope early warning level; constructing the weighing dynamic early warning mechanism based on the partition offset slope early warning level and the partition weighing early warning strategy.

[0012] In a possible implementation, the weighing monitoring and early warning method for a smart scale also performs the following processing: setting an abnormal number warning threshold and an abnormal weight warning threshold according to the false alarm processing requirements; constructing a weighing false alarm processing mechanism based on the abnormal number warning threshold and the abnormal weight warning threshold, and supplementing the weighing dynamic early warning mechanism through the weighing false alarm processing mechanism.

[0013] In a possible implementation, the weighing monitoring and early warning method for a smart scale further performs the following processing: when there is an abnormality in the initial weighing coefficient, the initial weighing coefficient is issued using a background authorization mechanism.

[0014] The present application also provides a weighing monitoring and early warning device for a smart scale, including: an initial weighing coefficient determination unit, used to place a target smart scale on a detection platform, activate a smart scale detection program to perform weighing record calculation on the target smart scale, determine an initial weighing coefficient, and perform a double backup of the initial weighing coefficient, the double backup including local storage and cloud backup; a weighing test parameter table construction unit, used to construct a weighing scenario test parameter table, obtain a real-time weight AD value set through multiple monitoring of the target smart scale, and record the actual weight information of the object; a partition regression fitting unit, used to calculate the actual weight of the object based on the actual weight of the object; The method comprises the following steps: performing partition regression fitting on the weight information and the real-time weight AD value set to generate a weighing nonlinear model, and obtaining a dynamic weighing coefficient set according to the weighing nonlinear model; a weighing coefficient offset slope determination unit is used to compare and calculate the dynamic weighing coefficient set with the initial weighing coefficient in sequence to determine a weighing coefficient offset slope set; and a weighing early warning strategy determination unit is used to construct a weighing dynamic early warning mechanism, triggering the weighing dynamic early warning mechanism based on the weighing coefficient offset slope set to perform classification matching, determine a weighing early warning strategy, and perform weighing early warning control on the target smart scale through the weighing early warning strategy.

[0015] The weighing monitoring and early warning method and device for smart scales proposed in this application is intended to activate the smart scale detection program to perform weighing record calculations on the target smart scale and determine the initial weighing coefficient; build a weighing scenario test parameter table to monitor and obtain the real-time weight AD value set and record the actual weight information of the object; perform partition regression fitting to generate a weighing nonlinear model; compare and calculate the dynamic weighing coefficient set with the initial weighing coefficient in sequence to determine the weighing coefficient offset slope set; build a weighing dynamic early warning mechanism, determine the weighing early warning strategy, and perform weighing early warning control on the target smart scale. This solves the technical problem in the prior art that smart scale weighing is easily affected by factors such as the environment and long-term use, resulting in the inability to timely and accurately monitor weighing deviations and early warning lags, achieving the technical effect of improving the accuracy of smart scale weighing and the timeliness of monitoring and early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A flow chart of a weighing monitoring and early warning method for a smart scale provided in an embodiment of the present application.

[0018] Figure 2 A schematic structural diagram of a weighing monitoring and early warning device for a smart scale provided in an embodiment of the present application.

[0019] Description of the accompanying drawings: initial weighing coefficient determining unit 10, weighing test parameter table building unit 20, partition regression fitting unit 30, weighing coefficient offset slope determining unit 40, weighing early warning strategy determining unit 50. DETAILED DESCRIPTION

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0021] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0022] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms “first\second” involved are merely to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0023] The present application embodiment provides a weighing monitoring and early warning method for a smart scale, such as Figure 1 As shown, the method includes:

[0024] In step S100, a target smart scale is placed on a detection platform, and a smart scale detection program is activated to perform weighing record calculations on the target smart scale, determine an initial weighing coefficient, and perform a dual backup of the initial weighing coefficient, wherein the dual backup includes local storage and cloud backup.

[0025] Preferably, the smart scale that needs to be tested and calibrated is placed on a testing platform. The testing platform usually has certain stability and accuracy requirements to ensure that it will not be interfered with by external factors during the weighing process and to ensure the reliability of the weighing results. The smart scale detection program is activated to perform weighing record calculations on the target smart scale. Specifically, the smart scale detection program controls the smart scale to perform multiple operations, which may include checking its own hardware, initializing the sensor, and weighing record calculations. Among them, the smart scale will weigh the standard weight object placed on it and record the data obtained from each weighing. By analyzing and calculating these weighing data, the initial weighing coefficient of the smart scale is determined, which reflects the relationship between the output signal of the smart scale sensor and the actual weight, and serves as a benchmark for subsequent judgment of whether the smart scale weighing is accurate. For example, the standard The weight is fitted with the electrical signal value output by the sensor to obtain a relational expression, in which the relevant parameter is the initial weighing coefficient; the initial weighing coefficient is then doubly backed up, including local storage and cloud backup. Among them, local storage is to store the calculated initial weighing coefficient in the storage device of the smart scale itself, such as built-in flash memory or hard disk. Even if the smart scale encounters power outages or other failures during subsequent use, the initial weighing coefficient can still be read from the local storage for comparison and analysis; at the same time, the initial weighing coefficient is transmitted to the cloud server via the network for backup, realizing remote storage and sharing of data, making it convenient for managers to access and view these data through the network anytime and anywhere. Even if the local storage device of the smart scale itself is damaged, it will not lead to the loss of the initial weighing coefficient, thereby improving the security and reliability of the data.

[0026] Furthermore, step S100 also includes step S110, based on the smart scale detection program, starting the smart scale detection module set, the smart scale detection module set including a calibration object placement module, a weight recording module and a coefficient calculation module; step S120, numbering the standard weight set to obtain calibration number information, and placing the standard weight set on the target smart scale in sequence according to the calibration number information through the calibration object placement module; step S130, based on the weight recording module, recording in sequence the actual weight set and the displayed weight AD value set of the standard weight set; step S140, performing regression fitting calculation on the actual weight set and the displayed weight AD value set through the coefficient calculation module to determine the initial weighing coefficient.

[0027] Preferably, the smart scale detection module set is further activated according to the smart scale detection program, wherein the smart scale detection module set includes a calibration object placement module, a weight recording module and a coefficient calculation module. Each module is responsible for a different detection task, and they work together to complete the detection of the smart scale and the determination of the initial weighing coefficient. Specifically, the standard weight set is a set of objects with known accurate weights, which are used to calibrate the smart scale. These standard weights are numbered to obtain calibration number information to clarify the identity and placement order of each standard weight. Then, the calibration object placement module places the standard weights on the target in order according to the calibration number information. On the smart scale, it ensures that each time a standard weight is placed, it can be accurately identified and recorded, and the placement order is fixed and traceable; after the standard weight is placed on the smart scale, the weight recording module will perform corresponding recording operations, including recording the actual weight of each object in the standard weight set to form an actual weight set. At the same time, the weight recording module will also record the weight AD value corresponding to each standard weight displayed on the smart scale. Among them, the weight AD value is the digital signal value obtained after the analog signal output by the smart scale sensor is converted from analog to digital, reflecting the current measurement of the object weight by the smart scale and forming a displayed weight AD value set.

[0028] Preferably, the final coefficient calculation module performs a regression fitting calculation on the actual weight set and the displayed weight AD value set recorded previously. That is, by analyzing the relationship between the two sets of data, an optimal mathematical model is found to describe the corresponding relationship between them. For example, a linear regression method may be used to find a linear equation, such as y=kx+b, where y is the actual weight, x is the displayed weight AD value, k is the slope to be determined, and b is a constant, to fit the two sets of data. The initial weighing coefficient that can achieve the best match between the displayed weight AD value and the actual weight of the smart scale is determined by calculation. For example, Table 1 provides a set of data showing different displayed weight AD values ​​and the corresponding actual weights, as well as the initial weighing coefficients (slope k and intercept b) calculated by linear regression, and verifies the accuracy of the linear equation.

[0029] Table 1 Example of linear regression fitting of smart scale weighing data

[0030]

[0031] The equation y=0.97x+1 obtained by linear regression can better fit the relationship between the displayed weight AD value and the actual weight. Here, k=0.97 and b=1 are the determined initial weighing coefficients. The initial weighing coefficients are then used to judge the accuracy of the smart scale. For example, if a newly measured displayed weight AD value is 600, the theoretical actual weight y=0.97×600+1=583 is substituted into the equation. If the actual weighed weight deviates greatly from 583, it means that the smart scale may have an inaccurate weighing problem.

[0032] Furthermore, step S100 also includes, when there is an abnormality in the initial weighing coefficient, using a background authorization mechanism to issue the initial weighing coefficient.

[0033] Preferably, if there is an abnormality in the initial weighing coefficient, the initial weighing coefficient is issued based on the background authorization mechanism. For example, the weighing equipment is calibrated by using calibration tools such as standard weights, and the correct value of the initial weighing coefficient is determined based on the calibration results. For example, when using high-precision standard weights to calibrate the electronic scale, the appropriate initial weighing coefficient is calculated and set based on the actual weight of the weights and the display value of the electronic scale to ensure the accuracy of the weighing; refer to the historical weighing data during normal operation in the past and the corresponding accurate initial weighing coefficient to determine the initial weighing coefficient that should be issued currently, such as comparing the initial weighing coefficient of the same model equipment or the same environment during normal operation in the past, and use it as the new issued value.

[0034] Step S200: Build a weighing scenario test parameter table, obtain a real-time weight AD value set through multiple monitoring of the target smart scale, and record the actual weight information of the object.

[0035] Preferably, the purpose of clearly building a weighing scenario test parameter table is to comprehensively and systematically test the weighing performance of the smart scale in different scenarios, such as weighing different materials in industrial production, pricing of different commodities in commercial retail, or measuring the weight of different groups of people in health care. Specifically, the test parameters are defined (including environmental parameters, characteristic parameters of the weighing items, and weighing frequency). Considering that environmental factors such as temperature, humidity, and air pressure may affect the sensor performance of the smart scale, these parameters are included in the table. For example, the temperature range is set to -10°C to 50°C, and the humidity range is set to 20%RH to 80%. RH, records the weighing performance of the smart scale under different environmental conditions; the characteristic parameters of the weighed items include the material (such as metal, plastic, wood, etc.), shape (regular or irregular), and weight range (from lightweight items such as a few grams of tablets to heavy items such as several tons of cargo). Items of different materials and shapes may produce different pressure distributions on the sensor, thereby affecting weighing accuracy; determine the number of times the same item is weighed within a period of time to test the stability of the smart scale under continuous use. For example, set the scale to weigh 10 times per hour and observe the changes in the weighing data of the smart scale within 8 hours of continuous operation.

[0036] Preferably, according to the test parameter table, various objects with different characteristics are weighed multiple times under different environmental conditions. For example, a 1 kg regular metal block is weighed 10 times in a set temperature of 25°C and humidity of 50%. The corresponding real-time weight AD value is recorded after each weighing. As the test progresses, a large amount of AD value data from different scenarios is accumulated, forming a real-time weight AD value set, reflecting the output of the smart scale's sensor under various complex conditions. At the same time, the actual weight information of the object is determined. For each object used in the test, its accurate actual weight must be determined. For objects with known weights, such as standard weights, the nominal weight can be used directly. For other items, a high-precision reference scale can be used to obtain an accurate actual weight value. Each time a real-time weight AD value is obtained, the corresponding actual weight information is recorded in detail. Using this recording method, the weight AD value displayed by the smart scale is compared and analyzed with the actual weight of the object to evaluate the weighing accuracy and performance of the smart scale.

[0037] Furthermore, step S200 also includes step S210, obtaining the application target of the smart scale, and setting the test object weight threshold and test environment information according to the smart scale application target and the specification attributes of the target smart scale; step S220, determining the weight interval density according to the weighing accuracy requirement, and performing multiple selections within the object weight threshold according to the weight interval density to determine the object test weight set; step S230, extracting factors from the test environment information to obtain a test environment related factor set; step S240, performing parameter design on the test environment related factor set based on the smart scale application target to obtain a test environment factor parameter set; step S250, combining and integrating the object test weight set and the test environment factor parameter set to build the weighing scenario test parameter table.

[0038] Preferably, obtaining the application target of the smart scale means clarifying the ultimate usage scenario and purpose of the smart scale. For example, the smart scale is used for weighing and pricing of goods in commercial retail, accurately weighing materials on industrial production lines, and measuring human body weight in the field of healthcare. Combined with the specification attributes of the smart scale (such as maximum weighing range, accuracy level, etc.), the test object weight threshold and test environment information are set. For example, if the smart scale is used for commercial retail, the maximum weighing range is 100 kilograms, and the accuracy is ±5 grams, then the test object weight threshold may be set to 0 to 100 kilograms; the test environment information may include a temperature range (such as 0°C to 40°C, simulating the temperature of a general commercial environment). changes), humidity range (such as 30%RH to 70%RH), etc.; according to the weighing accuracy requirements of the smart scale, determine the interval density of selecting the test weight within the object weight threshold range. If the accuracy requirement is high, the weight interval density will be smaller, that is, more test weight points will be selected; conversely, if the accuracy requirement is relatively low, the weight interval density can be larger. For example, for industrial weighing smart scales with high accuracy requirements, a test weight point may be set for every 1 kilogram; and for general commercial retail smart scales, a test weight point may be set for every 5 kilograms; according to the determined weight interval density, multiple selections are made within the object weight threshold to obtain the object test weight set.

[0039] Preferably, the established test environment information is analyzed in detail to extract key factors. These factors, in addition to temperature and humidity, may also include air pressure, vibration (if the smart scale is used in a vibrating environment), and electromagnetic interference (e.g., nearby large electrical equipment that may generate electromagnetic interference), and are then organized into a test environment-related factor set. For example, the test environment-related factor set may include temperature, humidity, air pressure, vibration, and electromagnetic interference. Based on the application objectives of the smart scale, specific parameters are designed for each factor in the test environment-related factor set, and the parameters of each factor are organized into a test environment factor parameter set. For example, the test environment factor parameter set may include temperature of 20°C to 25°C; humidity of 40% to 60% RH; air pressure of standard atmospheric pressure ±5%, no significant vibration; and electromagnetic interference, with no large electrical equipment within 1 meter. Finally, the determined object test weight set and the test environment factor parameter set are integrated and associated to generate a weighing scenario test parameter table. During testing, different object test weights can be weighed under different test environment parameter conditions, allowing for systematic testing and evaluation of the smart scale's performance in different scenarios.

[0040] Step S300 , performing partition regression fitting based on the actual weight information of the object and the real-time weight AD value set to generate a weighing nonlinear model, and obtaining a dynamic weighing coefficient set according to the weighing nonlinear model.

[0041] Preferably, the weighing of the smart scale may exhibit different characteristics in different weight ranges, or due to various factors (such as the nonlinear characteristics of the sensor, the influence of the mechanical structure, etc.), the relationship between weight and AD value is not a simple linear relationship. Therefore, the entire weight range is divided into different intervals, and then regression fitting is performed in each interval to more accurately describe the nonlinear relationship. Specifically, according to the actual weight information of the object and the corresponding real-time weight AD value set, the weight range is divided into several small intervals according to certain rules (such as equal weight intervals, according to data distribution characteristics, etc.). For the data in each small interval, a suitable regression method (such as least squares method, etc.) is used to find a function that can best fit the data in the interval. It can be a nonlinear function such as a polynomial function, exponential function, logarithmic function; then the regression fitting functions in each small interval are integrated together to form a weighing nonlinear model for the entire weighing range, which can more accurately describe the complex relationship between the actual weight and real-time weight AD value of the smart scale at different weights, better fit the data, and improve the accuracy of weighing. Assume that the weighing range of 0-100 kg is divided into three intervals: 0-20 kg, 20-50 kg, and 50-100 kg. In the 0-20 kg interval, the fitted function is ; In the range of 20-50 kg, the function is (It may be approximately linear in this range); in the range of 50-100 kg, the function is ; The entire weighing nonlinear model is a piecewise function composed of these three functions in their respective intervals. Among them, 、 、 、 、 、 、 The dynamic weighing coefficients reflect the relationship between the actual weight and the real-time AD value within different weight ranges. These coefficients are fitted based on data from different weight ranges and change with changes in weight. The dynamic weighing coefficients more accurately reflect the actual weighing conditions of the smart scale at different weights, helping to improve weighing accuracy and reliability. They can also be used to more precisely calibrate and adjust the smart scale's weighing results.

[0042] Furthermore, step S300 also includes step S310, calculating and obtaining the measured weight difference information of the actual weight information of the object and the real-time weight AD value set; step S320, performing threshold partitioning on each scene parameter in the weighing scene test parameter table based on the measured weight difference information to obtain the weighing scene parameter partition threshold; step S330, partitioning the actual weight information of the object and the real-time weight AD value set according to the weighing scene parameter partition threshold to obtain a partitioned measured weight data set; step S340, performing segmented regression fitting and combined verification and tuning on the partitioned measured weight data set in turn to generate a weighing nonlinear model.

[0043] Preferably, for each set of corresponding object actual weight information and real-time weight AD value, the real-time weight AD value is converted into the corresponding measured weight value through a certain conversion relationship (because the AD value itself may not be a direct weight unit and needs to be converted according to the characteristics of the smart scale), and the measured weight is subtracted from the actual weight of the object to obtain the measured weight difference information, which reflects the deviation between the measurement result of the smart scale and the actual weight; according to the measured weight difference information, different threshold ranges are determined, and each scene parameter in the weighing scene test parameter table is divided into different intervals, including treating scene parameters with small weight differences within the preset threshold as the same area, and zoning to obtain weight data of different areas with large scene parameter differences. For example, the difference within the range of ±0.1 grams may be divided into one interval, and the difference between 0.1 grams and 0.5 grams may be divided into another interval; the scene parameters are classified according to different error ranges to facilitate more detailed analysis and processing of data, and the boundary values ​​of these different intervals are the weighing scene parameter partition thresholds.

[0044] Preferably, according to the determined weighing scene parameter partition threshold, the actual weight information of the object and the real-time weight AD value set are classified and identified, and the data belonging to the same threshold range are grouped together to form a partitioned measurement weight data set, ensuring that the data in each set has similar measurement error characteristics, which is convenient for subsequent processing and analysis of data in different error ranges; for each partitioned measurement weight data set, a suitable regression method (such as polynomial regression, nonlinear regression, etc.) is used to fit them separately to find a function model that can best describe the data relationship in the set and more accurately capture the relationship between the actual weight and the measured weight; finally, the regression fitting models of each segment are combined together, and the model is verified and optimized through cross-validation, mean square error evaluation, etc. According to the verification results, the parameters or structure of the model are adjusted to improve the accuracy and generalization ability of the model, so that it can better adapt to different weighing scenarios and object weight ranges, and finally generate a weighing nonlinear model.

[0045] Furthermore, step S320 also includes step S321, calculating the mean and standard deviation of the measured weight difference information, and statistically analyzing the calculation results to obtain difference data distribution information; step S322, selecting a target clustering algorithm based on the difference data distribution information, and using the target clustering algorithm to perform cluster analysis on the measured weight difference information to obtain difference data clustering results; step S323, determining a weight difference data cluster based on the difference data clustering results; step S324, performing threshold matching partitioning on each scene parameter in the weighing scene test parameter table based on the weight difference data cluster to obtain the weighing scene parameter partition threshold.

[0046] Preferably, by calculating the mean and standard deviation of each measured weight difference, wherein the mean reflects the average level of the measured weight difference, and the standard deviation measures the degree of dispersion of the measured weight difference information. The larger the standard deviation, the more dispersed the data is and the worse the stability of the measurement result is; conversely, the more stable the measurement result is; a comprehensive analysis is performed on the mean and standard deviation, and the specific data of the measured weight difference information is combined to determine the distribution of the difference data, for example, to determine whether the data presents a normal distribution, a skewed distribution, etc., as well as the characteristics of the data such as the central tendency and dispersion degree; according to the characteristics of the difference data distribution information, an appropriate A combined clustering algorithm, such as the K-means clustering algorithm, the DBSCAN density clustering algorithm, the hierarchical clustering algorithm, etc., may be used. If the difference data is distributed relatively evenly and the number of clusters is roughly known, the K-means clustering algorithm may be more appropriate; if there are different density areas in the data distribution, the DBSCAN density clustering algorithm may be more able to accurately identify different clusters; then the measured weight difference information is processed, that is, according to the similarity between the data, the measured weight difference information is divided into different categories, so that the data in the same category has a higher similarity, and the data between different categories have a larger difference.

[0047] Preferably, after cluster analysis, the different categories obtained are the difference data clustering results. Each category can be regarded as a weight difference data cluster, representing different groups with similar measurement error characteristics. For example, one data cluster may contain all data with small measurement errors and relatively stable characteristics, while another data cluster may contain data with large measurement errors and large fluctuations. By determining the weight difference data cluster, data with different error characteristics can be analyzed and processed more targetedly; finally, for each weight difference data cluster, its data characteristics and distribution range are analyzed to determine a suitable threshold range. For example, for data clusters with small measurement errors, a smaller threshold range may be set; for data clusters with large measurement errors, a larger threshold range is set. Then, according to the threshold range, each scene parameter in the weighing scene test parameter table is partitioned to ensure that the scene parameters in each partition correspond to weight difference data clusters with similar measurement error characteristics, thereby obtaining the weighing scene parameter partition threshold, thereby ensuring weighing performance and accuracy.

[0048] Furthermore, step S340 also includes step S341, performing linear regression fitting on the partitioned measured weight data set in sequence to obtain a partitioned weighing coefficient regression model set; step S342, sequentially combining the partitioned weighing coefficient regression model set according to the partition threshold of the weighing scene parameter to obtain an initial weighing coefficient model; step S343, verifying and evaluating the initial weighing coefficient model to obtain a model partition determination coefficient; step S344, performing partition threshold tuning on the initial weighing coefficient model based on the model partition determination coefficient to generate the weighing nonlinear model.

[0049] Preferably, for each partitioned measured weight data set, a linear regression method is used to find the linear relationship between the data. Linear regression attempts to find a straight line (a hyperplane in multidimensional space) so that the straight line can best fit the data points, that is, to minimize the sum of the squares of the distances from the data points to the straight line, and then obtain a corresponding linear regression model for each partitioned measured weight data set, describing the linear relationship between the actual weight of the object and the measured weight in the partition, wherein each model has its own specific coefficient, thereby forming a partitioned weighing coefficient regression model set; according to the determined weighing scene parameter partition threshold, the partitioned weighing coefficient regression models are combined in a certain order. For example, if the partitions are divided according to the size of the weight difference, the corresponding models are combined in order from small to large or from large to small, thereby obtaining an overall model containing multiple partition models, that is, the initial weighing coefficient model, which can describe the relationship between the actual weight of the object and the measured weight through different linear relationships within different weight difference intervals, thereby more comprehensively covering the entire weighing range.

[0050] Preferably, the initial weighing coefficient model is verified and evaluated, and the model partition determination coefficient is calculated, wherein the model partition determination coefficient is the degree of fit of the model to the data, and the value range is between 0 and 1. The closer it is to 1, the better the model fits the data; the closer it is to 0, the worse the model fits. For each partition, the model partition determination coefficient is calculated separately to understand the model's fit to the data in different weight difference intervals, and then evaluate the performance of the initial weighing coefficient model in each partition, and find out the areas where there may be poor fit; then, based on the results of the model partition determination coefficient, the initial weighing coefficient model is adjusted and optimized. If the determination coefficient of a partition is low, it means that the model fit of the partition is not good, and it may be necessary to adjust the threshold of the partition, or further improve the model of the partition, such as adding higher-order terms to become a nonlinear model, so that the model can better fit the data, and finally generate a weighing nonlinear model that has a good fit effect in the entire weighing range, thereby improving the accuracy and reliability of the weighing system.

[0051] Step S400 : Compare and calculate the dynamic weighing coefficient set with the initial weighing coefficient in sequence to determine a weighing coefficient offset slope set.

[0052] Preferably, each coefficient in the dynamic weighing coefficient set is compared with the corresponding coefficient in the initial weighing coefficient, and the difference or ratio is calculated to measure the degree of difference between the two. The offset slope of the weighing coefficient is further calculated. The weighing coefficient offset slope set represents the rate of change of the dynamic weighing coefficient relative to the initial weighing coefficient. If the dynamic weighing coefficient and the initial weighing coefficient are regarded as functions of a certain variable (such as the weight of the object or the weighing scenario parameters), then the offset slope set is the set of change rates between these functions. By determining the weighing coefficient offset slope set, the trend of change and the degree of difference between the dynamic weighing coefficient and the initial weighing coefficient can be understood. If the value of the slope set is small and stable, it means that the initial weighing model is closer to the actual dynamic situation in different scenarios and the model accuracy is high. Conversely, if the slope set fluctuates greatly, it indicates that the initial weighing model may have deficiencies and needs further optimization.

[0053] Step S500: construct a weighing dynamic warning mechanism, trigger the weighing dynamic warning mechanism based on the weighing coefficient offset slope set to perform classification matching, determine a weighing warning strategy, and perform weighing warning control on the target smart scale through the weighing warning strategy.

[0054] Step S500 further includes step S510, performing partition warning analysis on the weighing coefficient offset slope set according to the partition threshold of the weighing scene parameter, and setting the partition offset slope warning level; step S520, determining the partition weighing warning strategy according to the partition offset slope warning level; step S530, constructing the weighing dynamic warning mechanism based on the partition offset slope warning level and the partition weighing warning strategy.

[0055] Preferably, various types of data are continuously collected during the operation of the smart scale, including real-time weight AD value, actual weight information, weighing scene parameters (such as ambient temperature, humidity, item material, etc.), and the dynamic weighing coefficient set and initial weighing coefficient calculated therefrom, etc., to determine the key indicators for triggering the early warning, that is, the weighing coefficient offset slope set, and set different slope threshold ranges by analyzing historical data and actual application needs; when the calculated weighing coefficient offset slope exceeds the pre-set normal range, the weighing dynamic early warning mechanism is triggered. For example, if the normal weighing coefficient offset slope range is set between -0.05 and 0.05, when a certain slope value is calculated to be 0.08, the early warning mechanism is activated, and the early warning situations are divided into different categories according to different slope ranges. Common classification methods include mild warning, moderate warning and high warning. The mild warning slope is between 0.05 and 0.1 (or a similar relatively small deviation range), indicating that the weighing system may begin to have some small deviations, but has not yet seriously affected the weighing accuracy; the moderate warning slope is between 0.1 and 0.2, indicating that the weighing deviation is more obvious, which may affect some applications with high weight accuracy requirements. The reasons may include the gradual decline in sensor performance, slight wear of equipment components, etc.; the high warning slope is higher than 0.2 (or a larger deviation value), and there is a serious problem with the weighing system, which is very likely to cause a large inaccuracy in the weighing results. Immediate measures need to be taken. Possible reasons include sensor failure, serious external interference to the equipment, or damage to key components.

[0056] Preferably, the weighing warning strategy, i.e., the response measures under different warnings, is determined based on the analysis of the warning classification results, and according to the determined warning strategy, the corresponding control operations are automatically executed to perform weighing warning control, including sending instructions to the smart scale to make it enter a specific maintenance mode, such as stopping data collection, displaying warning information, etc.; sending notifications to relevant personnel to ensure that they understand the situation in a timely manner and take action; starting backup equipment or processes, etc. to ensure the accuracy of weighing data. Specifically, combining the weighing coefficient offset slope set and the weighing scene parameter partition threshold, for each weighing scene parameter partition, analyze the change of the slope in the interval. For example, in the low temperature environment interval, study the distribution and fluctuation of the weighing coefficient offset slope; based on the analysis results, set different warning levels for each partition, such as mild, moderate, and high warnings; then determine the partition weighing warning strategy based on the partition offset slope warning level, which may include taking strategies such as closely monitoring changes in weighing data and recording relevant parameters when the warning level of a partition is mild; for partitions with moderate warnings, arrange for professionals to check whether the sensors of the weighing equipment are normal, whether the equipment connections are stable, etc., and stop some weighing operations with extremely high precision requirements; if the warning level is high, the weighing equipment in the area must be immediately deactivated, and the cause of the fault must be fully investigated. At the same time, the accuracy of the recent weighing data in the area must be evaluated to see if re-weighing or data correction is required. Finally, the determined zone offset slope warning level and the corresponding zone weighing warning strategy are integrated to form a weighing dynamic warning mechanism, which automatically determines which zone the user is currently in and triggers the corresponding warning strategy based on the zone's corresponding warning level, thus achieving dynamic warning and management of the smart scale weighing process. As shown in Table 2, an example is given of the relationship between the weighing coefficient offset slope, warning level, and recommended measures:

[0057] Table 2 Comparison table of smart scale weighing warnings and measures (according to weighing coefficient offset slope)

[0058]

[0059] Furthermore, step S530 also includes step S531, setting an abnormal number warning threshold and an abnormal weight warning threshold according to the false alarm processing requirements; step S532, constructing a weighing false alarm processing mechanism based on the abnormal number warning threshold and the abnormal weight warning threshold, and supplementing the weighing dynamic warning mechanism through the weighing false alarm processing mechanism.

[0060] Preferably, based on the actual false alarm processing requirements, the upper limit of the number of times the weighing system may experience abnormal situations (such as triggering an alarm but the actual weighing is normal) within a certain time range is determined. For example, it is set that the number of times the weighing system triggers an abnormal alarm within 1 hour cannot exceed 3 times. This "3 times" is the abnormal number warning threshold, which can be set based on the frequency of false alarms in historical data and the degree of false alarms acceptable to the business. Also based on the false alarm processing requirements, a weight value standard is set. When the abnormal weight displayed by the weighing system exceeds this standard, it is regarded as an abnormal situation. For example, for general item weighing scenarios, the abnormal weight warning threshold is set to ±5 kilograms, that is, when the deviation of the weighing result from the normal expected weight exceeds plus or minus 5 kilograms, it is regarded as an abnormal weight situation.

[0061] Preferably, the weighing system continuously monitors the number of abnormalities and abnormal weight during the weighing process. When the number of abnormalities reaches or exceeds the abnormal number warning threshold, or the abnormal weight exceeds the abnormal weight warning threshold, the corresponding processing flow is triggered. Once triggered, corresponding processing measures are taken. For example, if the number of abnormalities reaches the threshold, the automatic warning function of the weighing system may be suspended and switched to manual review mode. The staff will check and judge the subsequent weighing data to avoid continuous false alarms affecting the business; if the abnormal weight exceeds the threshold, professional personnel can be arranged to calibrate and inspect the weighing equipment to ensure the accurate operation of the equipment and review the previously affected weighing data.

[0062] Preferably, the constructed weighing false alarm processing mechanism is used to supplement the weighing dynamic early warning mechanism to optimize the accuracy and reliability of the early warning. After the weighing dynamic early warning mechanism triggers the early warning, the weighing false alarm processing mechanism will further determine whether these warnings are false alarms. If they are false alarms, they will be responded to according to the set processing measures to reduce the interference of false alarms on the business; if they are not false alarms, they will continue to be processed according to the original process of the weighing dynamic early warning mechanism, thereby improving the early warning and processing system of the entire weighing system; thereby further ensuring the accuracy of the weighing monitoring and early warning of the smart scale.

[0063] In the above, refer to Figure 1 The weighing monitoring and early warning method for a smart scale according to an embodiment of the present invention is described in detail. Figure 2 A weighing monitoring and early warning device for a smart scale according to an embodiment of the present invention is described.

[0064] The weighing monitoring and early warning device for smart scales according to the embodiment of the present invention is used to solve the technical problems in the prior art that the weighing of smart scales is easily affected by factors such as the environment and long-term use, resulting in the inability to timely and accurately monitor weighing deviations and early warning lags, thereby achieving the technical effect of improving the accuracy of smart scale weighing and the timeliness of monitoring and early warning. Figure 2As shown, the weighing monitoring and early warning device for a smart scale includes: an initial weighing coefficient determination unit 10, a weighing test parameter table construction unit 20, a partition regression fitting unit 30, a weighing coefficient offset slope determination unit 40, and a weighing early warning strategy determination unit 50.

[0065] The initial weighing coefficient determination unit 10 is used to place the target smart scale on the detection platform, activate the smart scale detection program to perform weighing record calculation on the target smart scale, determine the initial weighing coefficient, and perform dual backup of the initial weighing coefficient, the dual backup includes local storage and cloud backup; the weighing test parameter table construction unit 20 is used to build a weighing scenario test parameter table, obtain a real-time weight AD value set through multiple monitoring of the target smart scale, and record the actual weight information of the object; the partition regression fitting unit 30 is used to calculate the actual weight information of the object based on the actual weight information of the object and the real-time weight AD value set. The D value set is subjected to partition regression fitting to generate a weighing nonlinear model, and a dynamic weighing coefficient set is obtained based on the weighing nonlinear model; a weighing coefficient offset slope determination unit 40 is used to compare and calculate the dynamic weighing coefficient set with the initial weighing coefficient in sequence to determine the weighing coefficient offset slope set; a weighing early warning strategy determination unit 50 is used to construct a weighing dynamic early warning mechanism, trigger the weighing dynamic early warning mechanism based on the weighing coefficient offset slope set to perform classification matching, determine the weighing early warning strategy, and perform weighing early warning control on the target smart scale through the weighing early warning strategy.

[0066] The specific configuration of the initial weighing coefficient determination unit 10 will be described in detail below. The initial weighing coefficient determination unit 10 further includes: based on the smart scale detection program, starting the smart scale detection module set, the smart scale detection module set including a calibration object placement module, a weight recording module, and a coefficient calculation module; numbering the standard weight set to obtain calibration number information, and sequentially placing the standard weight set on the target smart scale according to the calibration number information via the calibration object placement module; sequentially recording the actual weight set and the displayed weight AD value set of the standard weight set based on the weight recording module; and performing regression fitting calculation on the actual weight set and the displayed weight AD value set via the coefficient calculation module to determine the initial weighing coefficient.

[0067] The specific configuration of the weighing test parameter table building unit 20 will be described in detail below. The weighing test parameter table building unit 20 further includes: obtaining the smart scale application target, setting the test object weight threshold and test environment information according to the smart scale application target and the specification attributes of the target smart scale; determining the weight interval density according to the weighing accuracy requirement, performing multiple selections within the object weight threshold according to the weight interval density, and determining the object test weight set; extracting factors from the test environment information to obtain a test environment related factor set; performing parameter design on the test environment related factor set based on the smart scale application target to obtain a test environment factor parameter set; and combining the object test weight set and the test environment factor parameter set to build the weighing scenario test parameter table.

[0068] The specific configuration of the partitioned regression fitting unit 30 will be described in detail below. The partitioned regression fitting unit 30 further includes: calculating and obtaining the measured weight difference information between the actual weight information of the object and the real-time weight AD value set; performing threshold partitioning on each scenario parameter in the weighing scenario test parameter table based on the measured weight difference information to obtain a weighing scenario parameter partition threshold; performing partition identification on the actual weight information of the object and the real-time weight AD value set according to the weighing scenario parameter partition threshold to obtain a partitioned measured weight data set; and sequentially performing segmented regression fitting and combined verification and optimization on the partitioned measured weight data set to generate a weighing nonlinear model.

[0069] The specific configuration of the partitioned regression fitting unit 30 will be described in detail below. The partitioned regression fitting unit 30 further includes: calculating the mean and standard deviation of the measured weight difference information, and statistically analyzing the calculation results to obtain difference data distribution information; selecting a target clustering algorithm based on the difference data distribution information, and using the target clustering algorithm to perform cluster analysis on the measured weight difference information to obtain difference data clustering results; determining weight difference data clusters based on the difference data clustering results; and performing threshold matching partitioning on each scenario parameter in the weighing scenario test parameter table based on the weight difference data clusters to obtain the weighing scenario parameter partition threshold.

[0070] The specific configuration of the partition regression fitting unit 30 will be described in detail below. The partition regression fitting unit 30 further includes: sequentially performing linear regression fitting on the partition measurement weight data set to obtain a partition weighing coefficient regression model set; sequentially combining the partition weighing coefficient regression model set according to the partition threshold of the weighing scenario parameter to obtain an initial weighing coefficient model; verifying and evaluating the initial weighing coefficient model to obtain a model partition determination coefficient; and performing partition threshold tuning on the initial weighing coefficient model based on the model partition determination coefficient to generate the weighing nonlinear model.

[0071] The specific configuration of the weighing warning strategy determination unit 50 will be described in detail below. The weighing warning strategy determination unit 50 further includes: performing a zone warning analysis on the weighing coefficient offset slope set according to the zone threshold of the weighing scenario parameter, setting a zone offset slope warning level; determining a zone weighing warning strategy based on the zone offset slope warning level; and constructing the weighing dynamic warning mechanism based on the zone offset slope warning level and the zone weighing warning strategy.

[0072] The specific configuration of the weighing warning strategy determination unit 50 will be described in detail below. The weighing warning strategy determination unit 50 further includes: setting an abnormal number warning threshold and an abnormal weight warning threshold based on false alarm processing requirements; constructing a weighing false alarm processing mechanism based on the abnormal number warning threshold and the abnormal weight warning threshold, and supplementing the weighing dynamic warning mechanism with the weighing false alarm processing mechanism.

[0073] The following will further describe in detail the specific configuration of the initial weighing coefficient determination unit 10. The initial weighing coefficient determination unit 10 further includes: when the initial weighing coefficient is abnormal, using a background authorization mechanism to issue the initial weighing coefficient.

[0074] The weighing monitoring and early warning device for a smart scale provided in an embodiment of the present invention can execute the weighing monitoring and early warning method for a smart scale provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0075] Although the present application makes various references to certain modules in the apparatus according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0076] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A weighing monitoring and early warning method for a smart scale, characterized in that: The method comprises: Placing the target smart scale on the testing platform, activating the smart scale testing program to perform weighing record calculations on the target smart scale, determining an initial weighing coefficient, and performing a dual backup of the initial weighing coefficient, the dual backup including local storage and cloud backup; Build a weighing scenario test parameter table, obtain a real-time weight AD value set through multiple monitoring of the target smart scale, and record the actual weight information of the object; Performing partition regression fitting based on the actual weight information of the object and the real-time weight AD value set to generate a weighing nonlinear model, and obtaining a dynamic weighing coefficient set according to the weighing nonlinear model; Compare and calculate the dynamic weighing coefficient set with the initial weighing coefficient in sequence to determine a weighing coefficient offset slope set; Constructing a weighing dynamic early warning mechanism, triggering the weighing dynamic early warning mechanism to perform classification matching based on the weighing coefficient offset slope set, determining a weighing early warning strategy, and performing weighing early warning control on the target smart scale through the weighing early warning strategy; The generating of the weighing nonlinear model comprises: Calculating and obtaining the actual weight information of the object and the measured weight difference information of the real-time weight AD value set; Performing threshold partitioning on each scene parameter in the weighing scene test parameter table based on the measured weight difference information to obtain a weighing scene parameter partition threshold; Partitioning the actual weight information of the object and the real-time weight AD value set according to the weighing scene parameter partition threshold to obtain a partitioned measured weight data set; Performing segmented regression fitting and combined verification and optimization on the partitioned measurement weight data set in sequence to generate a weighing nonlinear model; The generating of the weighing nonlinear model comprises: Performing linear regression fitting on the partitioned weight measurement data set in sequence to obtain a partitioned weighing coefficient regression model set; Sequentially combining the partitioned weighing coefficient regression model set according to the partitioned threshold value of the weighing scenario parameter to obtain an initial weighing coefficient model; Verifying and evaluating the initial weighing coefficient model to obtain the model partition determination coefficient; The initial weighing coefficient model is partitioned and threshold value optimized based on the model partition determination coefficient to generate the weighing nonlinear model.

2. The weighing monitoring and early warning method for a smart scale according to claim 1, characterized in that: Determining the initial weighing coefficient includes: Based on the smart scale detection program, starting a smart scale detection module set, the smart scale detection module set including a calibration object placement module, a weight recording module and a coefficient calculation module; Numbering the set of standard weights to obtain calibration number information, and placing the set of standard weights on the target smart scale in sequence according to the calibration number information by the calibration object placement module; The actual weight set and the displayed weight AD value set of the standard weight set are recorded in sequence based on the weight recording module; The coefficient calculation module performs regression fitting calculation on the actual weight set and the displayed weight AD value set to determine the initial weighing coefficient.

3. The weighing monitoring and early warning method for a smart scale according to claim 1, characterized in that: The weighing scenario test parameter table includes: Obtaining a smart scale application target, and setting a test object weight threshold and test environment information according to the smart scale application target and the specification attributes of the target smart scale; Determine the weight interval density according to the weighing accuracy requirement, and perform multiple selections within the object weight threshold according to the weight interval density to determine the object test weight set; Extracting factors from the test environment information to obtain a test environment-related factor set; Performing parameter design on the test environment associated factor set based on the application target of the smart scale to obtain a test environment factor parameter set; The object test weight set and the test environment factor parameter set are combined and integrated to build the weighing scenario test parameter table.

4. The weighing monitoring and early warning method for a smart scale according to claim 1, characterized in that: Obtaining the weighing scene parameter partition threshold includes: Calculating the mean and standard deviation of the measured weight difference information, and statistically analyzing the calculation results to obtain difference data distribution information; Selecting a target clustering algorithm according to the difference data distribution information, and performing cluster analysis on the measured weight difference information using the target clustering algorithm to obtain a difference data clustering result; Determining a weight difference data cluster according to the difference data clustering result; Based on the weight difference data cluster, each scene parameter in the weighing scene test parameter table is subjected to threshold matching partitioning to obtain the weighing scene parameter partition threshold.

5. The weighing monitoring and early warning method for a smart scale according to claim 1, characterized in that: The construction of the weighing dynamic early warning mechanism includes: Performing a partition warning analysis on the weighing coefficient offset slope set according to the partition threshold of the weighing scenario parameter, and setting a partition offset slope warning level; Determine a zone weighing warning strategy according to the zone offset slope warning level; Based on the partition offset slope warning level and the partition weighing warning strategy, the weighing dynamic warning mechanism is constructed.

6. The weighing monitoring and early warning method for a smart scale according to claim 5, characterized in that: The method further comprises: According to the false alarm processing requirements, set the abnormal number warning threshold and abnormal weight warning threshold; Based on the abnormal number warning threshold and the abnormal weight warning threshold, a weighing false alarm processing mechanism is constructed, and the weighing dynamic warning mechanism is supplemented by the weighing false alarm processing mechanism.

7. The weighing monitoring and early warning method for a smart scale according to claim 1, characterized in that: The method further comprises: When the initial weighing coefficient is abnormal, the initial weighing coefficient is issued using a background authorization mechanism.

8. A weighing monitoring and early warning device for a smart scale, characterized in that: The device is used to implement the weighing monitoring and early warning method for a smart scale according to any one of claims 1 to 7, and the device includes: an initial weighing coefficient determination unit, configured to place a target smart scale on a detection platform, activate a smart scale detection program to perform weighing record calculations on the target smart scale, determine an initial weighing coefficient, and perform a dual backup of the initial weighing coefficient, the dual backup including local storage and cloud backup; A weighing test parameter table building unit is used to build a weighing scenario test parameter table, obtain a real-time weight AD value set through multiple monitoring of the target smart scale, and record the actual weight information of the object; A partition regression fitting unit is used to perform partition regression fitting based on the actual weight information of the object and the real-time weight AD value set to generate a weighing nonlinear model, and obtain a dynamic weighing coefficient set according to the weighing nonlinear model; a weighing coefficient offset slope determining unit, configured to compare and calculate the dynamic weighing coefficient set with the initial weighing coefficient in sequence to determine a weighing coefficient offset slope set; A weighing warning strategy determination unit is used to construct a weighing dynamic warning mechanism, trigger the weighing dynamic warning mechanism based on the weighing coefficient offset slope set to perform classification matching, determine the weighing warning strategy, and perform weighing warning control on the target smart scale through the weighing warning strategy.

Images