Efficient weighing device and method for avoiding sample weighing loss

Abstract

The invention discloses an efficient weighing device and method capable of avoiding sample weighing loss, and particularly relates to the technical field of sample weighing. The device comprises a sample weighing data acquisition module, a data processing module, a weight calculation module, a judgment decision module, a control execution module and a man-machine interaction module. Environment parameters and initial weight are obtained through an environment data acquisition unit and a weight data acquisition unit, data are input into a volatile or hygroscopic sample model after data preprocessing and abnormal value detection and correction, and a real-time weight evaluation value is calculated by adopting a fourth-order Runge-Kutta method. And judging a risk level according to a relative error between the actually measured weight and the evaluated value to execute a control strategy. And meanwhile, sample surface state parameters are obtained by using an image recognition technology, and whole-process tracing and model optimization are realized by combining historical data. According to the scheme, through multi-model dynamic modeling, environment parameter deep fusion and intelligent closed-loop control, the weighing precision is effectively improved, and sample volatilization or moisture absorption loss is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of sample weighing, and more particularly to an efficient weighing device and method for avoiding sample weighing loss. Background Art

[0002] In the field of sample weighing technology, existing technologies have made certain progress in automated data collection and basic environmental compensation. For example, some weighing systems can collect temperature and humidity data in real time through sensors and perform preliminary filtering on weight data, thereby improving the stability of weighing data. At the same time, some advanced electronic balances have achieved static weighing with milligram-level accuracy through electromagnetic force compensation technology, meeting the high-precision measurement requirements of routine laboratory samples. In addition, some systems have introduced human-computer interaction interfaces, supporting historical data storage and simple queries, providing basic tools for the traceability of the weighing process.

[0003] However, it still has some shortcomings in actual use, such as the failure to monitor the impact of environmental parameters such as temperature, humidity, and air pressure on sample weight in real time. It is easy to cause weighing errors due to environmental fluctuations, such as temperature increase leading to accelerated volatilization and humidity increase causing moisture absorption. Especially in complex environments, weighing accuracy cannot be guaranteed. No sophisticated mathematical model has been established for volatile and hygroscopic samples, and the volatilization or moisture absorption rate cannot be accurately calculated, resulting in the inability to provide early warning of weighing loss risks and difficulty in achieving differentiated and precise control of different samples. There is a lack of data preprocessing, outlier correction and dynamic parameter calibration mechanisms, and there is a high degree of dependence on manual operation. Risk judgment is based only on a single weight error, and the cause of loss is not analyzed in combination with environmental characteristics and model parameter changes. The control strategy lacks specificity. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an efficient weighing device and method for avoiding sample weighing loss, and solve the problems raised in the above-mentioned background technology through the following scheme.

[0005] To achieve the above object, the present invention provides the following technical solution: an efficient weighing device for avoiding sample weighing loss, comprising a sample weighing data acquisition module, a data processing module, a weight calculation module, a judgment and decision module, a control execution module, and a human-computer interaction module;

[0006] The sample weighing data acquisition module includes an environment data acquisition unit and a weight data acquisition unit, and is used to acquire the first sample feature text and output the first sample feature text to the data processing module;

[0007] The data processing module includes a data preprocessing unit, an outlier detection unit, and a data feature extraction unit, and is configured to process the first sample feature text to obtain the second sample feature text;

[0008] The weight calculation module is used to input the second sample characteristic text into the volatile sample model or the hygroscopic sample model to calculate the rate of change of the sample weight over time, and obtain a weight evaluation value based on the integral of the rate of change of the sample weight over time;

[0009] The judgment and decision module is used to calculate the relative error based on the weight assessment value and the measured weight, determine the risk level based on the relative error, and analyze the cause of the loss;

[0010] The control execution module is used to execute the preset control strategy according to the risk level and display specific processing suggestions on the human-computer interaction interface based on the loss cause analysis results;

[0011] The human-computer interaction module is used to receive information input by the user and provide data query and statistical analysis functions, support querying historical data by time, sample type, and risk level conditions, and generate statistical reports and trend analysis charts.

[0012] Preferably, an efficient weighing method for avoiding sample weighing loss comprises:

[0013] S1: Data collection: Obtain a first sample feature text through the environmental data collection unit and the weight data collection unit;

[0014] S2: Data processing: performing data preprocessing, outlier detection and correction on the first sample feature text obtained in step S1 to obtain a second sample feature text;

[0015] S3: Calculation of weight assessment value: The operator inputs the sample type through the human-computer interaction module, and the system selects the corresponding mathematical model based on the input. The system obtains calculation parameters based on the sample information input by the operator and the typical parameter values ​​of the sample pre-stored in the database. The second sample feature text obtained in step S2 is substituted into the calculation model to obtain the rate of change of the sample weight over time. The weight assessment value is obtained based on the integration of the rate of change of the sample weight over time;

[0016] S4: Judgment and decision-making: Calculate the relative error based on the measured weight and the weight assessment value obtained in step S3, and determine the risk level based on the relative error, while analyzing the cause of the loss;

[0017] S5: Control execution: Execute the preset control strategy based on the risk level obtained in step S4, and display specific processing suggestions on the human-computer interaction interface based on the loss cause analysis results.

[0018] Technical effects and advantages of the present invention:

[0019] 1. The present invention collects environmental parameters in real time through temperature, humidity, and air pressure sensors, and deeply integrates them into the volatile sample model and the hygroscopic sample model. At the same time, the fourth-order Runge-Kutta method is used to numerically integrate the differential equations, and the weight evaluation value is calculated in real time with a 1-second time step to ensure the timeliness and accuracy of the model output;

[0020] 2. The present invention establishes dedicated models for volatile and hygroscopic samples, dynamically adjusting model parameters through feature extraction to achieve refined simulation of the volatilization or hygroscopic behavior of different samples. Furthermore, the present invention comprehensively determines the risk level and cause of loss based on relative error and changes in model parameters. For example, the cause of accelerated volatilization is determined by linking α with the temperature change rate, thereby improving the scientific nature of the judgment.

[0021] 3. Realize full process automation of data collection, processing, calculation, decision-making and execution. For example, when the risk is medium, the temperature and humidity are automatically adjusted to the set value ±0.5℃ / ±3%RH. When the risk is high, the weighing is immediately suspended and the inert gas protection or sealing device is activated to reduce human intervention errors. At the same time, the present invention realizes historical data query and statistical analysis through the human-computer interaction module, optimizes model parameters through data tracing, and forms an intelligent closed loop of "data-model-control". BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0023] Figure 2 Schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0025] As attached Figure 1 An efficient weighing device for avoiding sample weighing loss is shown, comprising a sample weighing data acquisition module, a data processing module, a weight calculation module, a judgment and decision module, a control execution module, and a human-computer interaction module:

[0026] The sample weighing data acquisition module includes an environment data acquisition unit and a weight data acquisition unit, and is used to acquire the first sample feature text and output the first sample feature text to the data processing module;

[0027] It should be further explained that the environmental data acquisition unit includes a temperature sensor, a humidity sensor and an air pressure sensor;

[0028] The first sample feature text includes current ambient temperature, current ambient humidity, current ambient air pressure and initial weight;

[0029] The data processing module includes a data preprocessing unit, an outlier detection unit, and a data feature extraction unit, and is configured to process the first sample feature text to obtain the second sample feature text;

[0030] It should be specifically noted that the method for processing the first sample feature text includes data preprocessing and outlier detection and correction;

[0031] The data preprocessing refers to normalizing the temperature, humidity, and air pressure data, mapping them to the interval [0, 1], and smoothing the weight data using a sliding average method with a sliding window size of n=5;

[0032] The outlier detection and correction includes using the 3σ principle to detect outliers in environmental parameters and weight data, correcting abnormal environmental parameters using linear interpolation of data before and after, and replacing abnormal weight data with the average of the two weight data before and after if it is a single abnormality;

[0033] The weight calculation module is used to input the second sample characteristic text into the volatile sample model or the hygroscopic sample model to calculate the rate of change of the sample weight over time, and obtain a weight evaluation value based on the integral of the rate of change of the sample weight over time;

[0034] The volatile sample model is as follows:

[0035] in Refers to the rate of change of sample weight over time, that is, the volatilization rate, in grams per second. By combining this value with the time integral, the weight evaluation value at different times can be obtained. α refers to the dynamic correction coefficient, k1 refers to the volatilization coefficient, S is the sample surface area, P vap Refers to the sample vapor pressure, E a refers to the activation energy, R refers to the gas constant, T refers to the current absolute temperature, H refers to the current relative humidity, H sat Refers to the saturated humidity of the sample at the current temperature, P atm Refers to the real-time air pressure, P std Refers to standard atmospheric pressure;

[0036] It should be noted that α is a dimensionless value, which is adaptively adjusted according to the volatility error under the same environmental conditions in the historical weighing data. The initial value is set to 1, and the adjustment formula is: Where W 1，t-1 Refers to the weight evaluation value at time t-1, W 2，t-1Refers to the measured weight at time t-1, β is the preset adjustment step, with a value range of (0, 0.1), which is used to correct the deviation between the model calculation and the actual situation, k1 is obtained through experimental measurement or reference to relevant literature, and characterizes the volatilization speed of the sample under conditions such as unit surface area and unit vapor pressure, S is obtained through geometric measurement, and P vap According to the relevant physical property manual, when the temperature change rate exceeds the threshold or the humidity shows an upward trend, the empirical formula is used. Correction is performed, where γ1 and γ2 are correction coefficients, and δ H It is the humidity change trend indicator, rising is 1, stable is 0, and falling is -1. a It is the energy barrier that the sample molecules need to overcome to transform from condensed state to gaseous state. The larger the value, the more difficult it is to volatilize. It is determined by experimental methods such as thermal analysis.

[0037] The hygroscopic sample model is as follows:

[0038] in Refers to the rate of change of sample weight over time, that is, the moisture absorption rate, k2 refers to the moisture absorption ratio coefficient, A refers to the moisture absorption activity parameter, Refers to the sample surface state parameters, H refers to the current relative humidity, H eq Refers to the sample equilibrium humidity, P atm Refers to the real-time air pressure, P std Refers to standard atmospheric pressure;

[0039] It should be specifically noted that k2 is related to factors such as the affinity between the sample and water vapor. It is a dimensionless value determined by the material properties of the sample and can be obtained by fitting experimental data. It reflects the hygroscopic capacity of the sample under conditions such as unit humidity difference and characterizes the number or activity of active sites participating in the hygroscopic process of the sample. The unit is 1 / second. The larger the value, the stronger the hygroscopic capacity of the sample. It can be determined through experimental measurement or theoretical calculation based on the chemical composition, structure and other characteristics of the sample. When the humidity is rapidly decreasing, the formula A′=A(1-λ·δ H ) to adjust A, where λ is the adjustment coefficient, δ H The humidity change trend is marked as 1 when rising, 0 when stable, and -1 when falling. eq It refers to the relative humidity of a sample when it absorbs or dehumidifies to reach equilibrium under specific environmental conditions. It is related to the properties of the sample itself and the ambient temperature and humidity. It can be determined by experimental methods, that is, by placing the sample in a specific environment for a long enough time and measuring the ambient humidity when the weight no longer changes. The roughness and porosity of the sample surface are obtained through image recognition technology and quantified. The value range is [0,1], where 0 means the surface is smooth, without pores, and almost no moisture absorption, and 1 means the surface is rough, with rich pores and strong moisture absorption capacity;

[0040] described The specific steps to obtain are as follows:

[0041] A1: Image acquisition: Use a 1080P high-definition camera equipped with a ring light source to capture multi-angle high-definition images of the sample surface under fixed light intensity, shooting distance and angle. Multi-angle shooting can fully capture the sample surface information and reduce feature omissions caused by viewing angle problems.

[0042] A2: Image preprocessing: Grayscale the collected image to eliminate color information interference and simplify calculations; then perform filtering operations, such as using Gaussian filtering to remove noise in the image and improve image quality; then perform image enhancement, using methods such as histogram equalization to enhance image contrast and make subtle features on the sample surface clearer.

[0043] A3: Feature extraction: Use digital image processing algorithms to extract key features, adopt edge detection algorithms to identify the boundaries and contours of the sample surface, and obtain surface shape information; use threshold segmentation algorithms to divide the image into different areas and distinguish between pores and non-porous parts of the surface; use texture analysis algorithms to extract texture features related to surface roughness and calculate texture energy, entropy, and contrast parameters.

[0044] A4: Quantitative evaluation: Establish characteristic parameters and The mapping relationship of the values ​​is set, and the rules are set. If the pore area in the image is larger, the edge complexity is higher, and the texture roughness parameter value is larger, then The closer the value is to 1, the smoother the surface and the fewer pores. The closer the value is to 0, for example, the regression model in machine learning can be used to utilize a large number of known value( The value is determined by manual evaluation combined with experimental measurement) and trained with sample data of corresponding image feature parameters to obtain a model that can accurately predict the φ value based on image features.

[0045] A5: Real-time update: During the weighing process, the image is re-collected and processed at 10-second intervals, and updated in real time value, ensuring that the model calculation can timely reflect the impact of changes in the sample surface state on the moisture absorption process.

[0046] It should be noted that when calculating Then, the fourth-order Runge–Kutta method was used for numerical integration to calculate the real-time weight evaluation value with a time step of Δt = 1 second;

[0047] It should be further explained that the fourth-order Runge-Kutta method (RK4) is mainly used to numerically integrate the rate of change of sample weight over time to calculate the real-time weight evaluation value. The specific application process is as follows:

[0048] B1: The environmental parameters and initial sample weight are acquired in real time through the environmental data acquisition unit and the weight data acquisition unit. The operator inputs the sample type, and the system automatically retrieves the sample calculation parameters pre-stored in the database, or prompts for manual input.

[0049] B2: Based on the sample type, the weight change rate is expressed as a differential equation related to environmental parameters and sample characteristics, and the correction coefficient in the differential equation is dynamically adjusted to make the rate calculation more realistic.

[0050] B3: With Δt = 1 second as the interval, the continuous weight change process is discretized into multiple time steps to ensure the real-time calculation. In each time step, the weight change rate estimates at the current moment and the intermediate moment are weighted averaged to obtain the average rate, and then the weight evaluation value at the next moment is calculated.

[0051] This process does not require exposing the specific form of differential equations. It only requires the use of a "black box" function f(t,W) to realize rate calculation, which integrates the combined influence of sample characteristics and environmental parameters.

[0052] It should be further explained that the α, k1, S, P vap 、E a , R, H sa 、P std , k2, A, and H eq It is based on the sample information input by the operator and the typical parameter values ​​of the sample stored in the database. If there is no corresponding data in the database, the system prompts the operator to manually enter the relevant parameters;

[0053] The judgment and decision module is used to calculate the relative error based on the weight assessment value and the measured weight, determine the risk level based on the relative error, and analyze the cause of the loss;

[0054] The relative error is calculated as follows:

[0055] Where ω refers to the relative error W 实际 Refers to the measured weight, W 评估 Refers to the weight assessment value;

[0056] The risk level determination method is as follows:

[0057] When the relative error is less than 2%, it is judged as low risk and the system continues to monitor normally. When 2% ≤ relative error < 6%, it is judged as medium risk and the system issues a yellow warning. When the relative error is ≥ 6%, it is judged as high risk and the system issues a red warning and suspends weighing.

[0058] The analysis method of the loss causes is as follows:

[0059] For volatile samples, if the dynamic correction coefficient α continues to increase and the temperature change rate exceeds the preset threshold, it is judged that the temperature rise causes the volatilization to accelerate. If P vap If corrections are frequent and the humidity trend is increasing, it is judged to be affected by humidity. For hygroscopic samples, if the difference between theoretical and actual weights cannot be narrowed after adjustment A, and the humidity increases, it is judged that the surface state of the sample affects moisture absorption.

[0060] The control execution module is used to execute the preset control strategy according to the risk level and display specific processing suggestions on the human-computer interaction interface based on the loss cause analysis results;

[0061] The control strategy is as follows:

[0062] When the risk level is low, the system continues to operate normally, recording environmental parameters, weight data, and model calculation results every 5 minutes;

[0063] When the risk level is medium, the temperature and humidity controller is activated to adjust the ambient temperature to the set value ±0.5°C and the humidity to the set value ±3%RH. An early warning message is displayed on the human-computer interaction interface to prompt the operator to check the sample status and weighing environment.

[0064] When the risk level is high, immediately stop the current weighing operation, lock the weighing platform to prevent operator misoperation, start the air purifier to reduce the concentration of particulate matter in the environment, adjust the anti-vibration platform to reduce the impact of external vibration, and start the inert gas protection device for volatile samples; start the sealing device for hygroscopic samples;

[0065] The human-computer interaction module is used to receive information input by the user and provide data query and statistical analysis functions, support querying historical data by time, sample type, and risk level conditions, and generate statistical reports and trend analysis charts;

[0066] As attached Figure 2 An efficient weighing method to avoid sample weighing loss is shown, including:

[0067] S1: Data collection: Obtain a first sample feature text through the environmental data collection unit and the weight data collection unit;

[0068] S2: Data processing: performing data preprocessing, outlier detection and correction on the first sample feature text obtained in step S1 to obtain a second sample feature text;

[0069] S3: Calculation of weight assessment value: The operator inputs the sample type through the human-computer interaction module, and the system selects the corresponding mathematical model based on the input. The system obtains calculation parameters based on the sample information input by the operator and the typical parameter values ​​of the sample pre-stored in the database. The second sample feature text obtained in step S2 is substituted into the calculation model to obtain the rate of change of the sample weight over time. The weight assessment value is obtained based on the integration of the rate of change of the sample weight over time;

[0070] It should be noted that the calculation parameters are α, k1, S, P vap 、E a , R, H sa 、P std , k2, A, and H eq ;

[0071] S4: Judgment and decision-making: Calculate the relative error based on the measured weight and the weight assessment value obtained in step S3, and determine the risk level based on the relative error, while analyzing the cause of the loss;

[0072] S5: Control execution: Execute the preset control strategy based on the risk level obtained in step S4, and display specific processing suggestions on the human-computer interaction interface based on the loss cause analysis results.

[0073] Secondly: The drawings of the embodiments disclosed in the present invention only involve structures related to the embodiments disclosed in the present invention. Other structures may refer to conventional designs. The same embodiment and different embodiments of the present invention may be combined with each other without conflict.

[0074] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An efficient weighing device for avoiding sample weighing loss, characterized in that: include: Sample weighing data acquisition module, data processing module, weight calculation module, judgment and decision module, control execution module and human-computer interaction module; The sample weighing data acquisition module includes an environment data acquisition unit and a weight data acquisition unit, and is used to acquire the first sample feature text and output the first sample feature text to the data processing module; The data processing module includes a data preprocessing unit, an outlier detection unit, and a data feature extraction unit, and is configured to process the first sample feature text to obtain the second sample feature text; The weight calculation module is used to input the second sample characteristic text into the volatile sample model or the hygroscopic sample model to calculate the rate of change of the sample weight over time, and obtain a weight evaluation value based on the integral of the rate of change of the sample weight over time; The judgment and decision module is used to calculate the relative error based on the weight assessment value and the measured weight, determine the risk level based on the relative error, and analyze the cause of the loss; The control execution module is used to execute the preset control strategy according to the risk level and display specific processing suggestions on the human-computer interaction interface based on the loss cause analysis results; The human-computer interaction module is used to receive information input by the user and provide data query and statistical analysis functions, support querying historical data by time, sample type, and risk level conditions, and generate statistical reports and trend analysis charts.

2. The efficient weighing device for avoiding sample weighing loss according to claim 1, characterized in that: The environmental data acquisition unit includes a temperature sensor, a humidity sensor and an air pressure sensor; The first sample characteristic text includes current ambient temperature, current ambient humidity, current ambient air pressure and initial weight.

3. The efficient weighing device for avoiding sample weighing loss according to claim 1, characterized in that: The processing method of the first sample feature text includes data preprocessing and outlier detection and correction; The data preprocessing refers to normalizing the temperature, humidity, and air pressure data, mapping them to the interval [0, 1], and smoothing the weight data using a sliding average method with a sliding window size of n=5; The outlier detection and correction includes using the 3σ principle to detect outliers in environmental parameters and weight data, correcting abnormal environmental parameters using linear interpolation of data before and after, and replacing abnormal weight data with the average of the two weight data before and after if it is a single abnormality.

4. The efficient weighing device for avoiding sample weighing loss according to claim 1, characterized in that: The volatile sample model is as follows: in Refers to the rate of change of sample weight over time, that is, the volatilization rate, in grams per second. By combining this value with the time integral, the weight evaluation value at different times can be obtained. α refers to the dynamic correction coefficient, k1 refers to the volatilization coefficient, S is the sample surface area, P vap Refers to the sample vapor pressure, E a refers to the activation energy, R refers to the gas constant, T refers to the current absolute temperature, H refers to the current relative humidity, H sat Refers to the saturated humidity of the sample at the current temperature, P atm Refers to the real-time air pressure, P std Refers to standard atmospheric pressure.

5. The efficient weighing device for avoiding sample weighing loss according to claim 1, characterized in that: The hygroscopic sample model is as follows: in Refers to the rate of change of sample weight over time, that is, the moisture absorption rate, k2 refers to the moisture absorption ratio coefficient, A refers to the moisture absorption activity parameter, Refers to the sample surface state parameters, H refers to the current relative humidity, H eq Refers to the sample equilibrium humidity, P atm Refers to the real-time air pressure, P std Refers to standard atmospheric pressure.

6. The efficient weighing device for avoiding sample weighing loss according to claim 1, characterized in that: The relative error is calculated as follows: Where ω refers to the relative error W 实际 Refers to the measured weight, W 评估 Refers to the weight assessment value.

7. The efficient weighing device for avoiding sample weighing loss according to claim 1, characterized in that: The risk level determination method is as follows: When the relative error is less than 2%, it is judged as low risk and the system continues to monitor normally. When 2% ≤ relative error < 6%, it is judged as medium risk and the system issues a yellow warning. When the relative error is ≥ 6%, it is judged as high risk and the system issues a red warning and suspends weighing. The analysis method of the loss causes is as follows: For volatile samples, if the dynamic correction coefficient α continues to increase and the temperature change rate exceeds the preset threshold, it is judged that the temperature rise causes the volatilization to accelerate. If P vap If corrections are frequent and the humidity trend is increasing, it is judged to be affected by humidity. For hygroscopic samples, if the difference between theoretical and actual weights cannot be narrowed after adjustment A, and the humidity increases, it is judged that the surface state of the sample affects moisture absorption.

8. The efficient weighing device for avoiding sample weighing loss according to claim 1, characterized in that: The control strategy is as follows: When the risk level is low, the system continues to operate normally, recording environmental parameters, weight data, and model calculation results every 5 minutes; When the risk level is medium, the temperature and humidity controller is activated to adjust the ambient temperature to the set value ±0.5°C and the humidity to the set value ±3%RH. An early warning message is displayed on the human-computer interaction interface to prompt the operator to check the sample status and weighing environment. When the risk level is high, stop the current weighing operation immediately, lock the weighing platform to prevent the operator from operating it incorrectly, start the air purifier to reduce the concentration of particulate matter in the environment, adjust the anti-vibration platform to reduce the impact of external vibration, and for volatile samples, start the inert gas protection device; for hygroscopic samples, start the sealing device.

9. An efficient weighing method for avoiding sample weighing loss, used to implement the efficient weighing device for avoiding sample weighing loss as described in any one of claims 1 to 8, characterized in that: include: S1: Data collection: Obtain a first sample feature text through the environmental data collection unit and the weight data collection unit; S2: Data processing: performing data preprocessing, outlier detection and correction on the first sample feature text obtained in step S1 to obtain a second sample feature text; S3: Calculation of weight assessment value: The operator inputs the sample type through the human-computer interaction module, and the system selects the corresponding mathematical model based on the input. The system obtains calculation parameters based on the sample information input by the operator and the typical parameter values ​​of the sample pre-stored in the database. The second sample feature text obtained in step S2 is substituted into the calculation model to obtain the rate of change of the sample weight over time. The weight assessment value is obtained based on the integral of the rate of change of the sample weight over time; S4: Judgment and decision-making: Calculate the relative error based on the measured weight and the weight assessment value obtained in step S3, and determine the risk level based on the relative error, while analyzing the cause of the loss; S5: Control execution: Execute the preset control strategy based on the risk level obtained in step S4, and display specific processing suggestions on the human-computer interaction interface based on the loss cause analysis results.

Images