IoT-based Smart Bucket Status Data Monitoring and Management System and Method

Abstract

This invention discloses a smart storage container status data monitoring and management system and method based on the Internet of Things (IoT), relating to the field of staged status data analysis technology for smart storage containers. The method includes: integrating a multi-sensor monitoring device into the smart storage container body; real-time acquisition of physical state data of the substances inside the container; dividing the smart storage container into several working stages to obtain characteristic state parameters for different working stages; analyzing environmental factors affecting the changes in the characteristic state parameters; fitting the relationship between the characteristic state parameters and environmental factors based on several historical observation samples; determining the best fitting expression through a goodness-of-fit index; calculating the predicted value of the characteristic state parameters under the current environment using the best fitting expression; calculating the characteristic error degree based on a preset characteristic state parameter benchmark value; and issuing a real-time warning when the error exceeds a preset threshold. This method is applicable to various application scenarios such as liquid storage, significantly improving the system's intelligent monitoring level and security capabilities.

Description

Technical Field

[0001] This invention relates to the field of phased status data analysis technology for smart buckets, specifically to a smart bucket status data monitoring and management system and method based on the Internet of Things. Background Technology

[0002] With the development of the Internet of Things, artificial intelligence, and intelligent sensing technologies, industrial and environmental monitoring equipment is becoming increasingly intelligent and automated. In high-requirement applications involving hazardous chemicals, liquid raw materials, and food concentrates, real-time monitoring and safety early warning of the state of substances within containers are particularly crucial. Currently, common condition monitoring methods mainly involve installing single or a small number of sensors to collect one or a few physical quantities (such as temperature, pressure, and liquid level). However, these methods are insufficient to meet the demands for refined analysis and safety prediction under complex operating conditions involving multiple parameters, high dimensions, and multiple stages.

[0003] Traditional technologies for assessing the condition of containerized equipment often employ fixed-cycle monitoring methods. These methods lack precise tracking and modeling analysis of material state changes at different stages of operation, and are also insufficient for dynamically adjusting monitoring strategies and early warning thresholds in response to changes in the external environment. Furthermore, the material state within the container can fluctuate significantly under the influence of various environmental factors. Without scientifically effective data fusion and modeling methods, this can easily lead to misjudgments or missed reports, reducing the reliability and practicality of the monitoring system.

[0004] Some systems have introduced multi-sensor acquisition and data normalization processing methods, but these are usually limited to real-time status display or simple threshold alarms, failing to combine historical data for correlation analysis, feature parameter extraction, and multi-stage error tracing. Especially in different operating stages of container use, due to the inconsistent evolution patterns of physical state parameters, it is often difficult to establish effective prediction mechanisms and accurate early warning models without stage-by-stage modeling and feature correlation factor extraction.

[0005] Therefore, there is an urgent need for a full-cycle intelligent monitoring method that integrates functions such as multi-sensor acquisition, phased processing of state data, feature parameter extraction and environmental factor correlation modeling, and dynamic early warning threshold adjustment, so as to achieve efficient, accurate, and interpretable monitoring and early warning of anomalies in the state of materials in containers. Summary of the Invention

[0006] The purpose of this invention is to provide an IoT-based smart bucket status data monitoring and management system and method to solve the problems raised in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring and managing the status data of a smart bucket based on the Internet of Things, specifically including the following steps:

[0008] The smart bucket integrates a multi-sensor monitoring device to collect real-time physical state data of the substances inside the bucket; the multi-sensor monitoring device transmits the physical state data of the substances inside the bucket through the Internet of Things.

[0009] The smart chip bucket is divided into several working stages, and the characteristic state parameters of different working stages are obtained through phased data analysis.

[0010] The environmental factors affecting the changes in the characteristic state parameters are analyzed. The relationship between the characteristic state parameters and the environmental factors is fitted based on several historical observation samples, with the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables. The best fit expression is determined by the goodness-of-fit index.

[0011] Real-time acquisition of current environmental factors, calculation of predicted values ​​of characteristic state parameters under the current environment through the best fit expression, calculation of characteristic error degree based on preset characteristic state parameter benchmark values, and real-time warning when exceeding preset threshold.

[0012] The physical state data of the substances inside the smart chip bucket are collected throughout its entire working cycle; the smart chip bucket is divided into several working stages [s1, s2, ..., sn]; where s1, s2, ..., sn represent the 1st, 2nd, ..., nth working stages respectively; the physical state data of the substances inside the bucket in each stage are extracted respectively, and the normalization method is used to unify the dimensions between different physical state data.

[0013] For the i-th working stage, the real-time physical state data of the substances inside the container is represented as [k i1 k i2 ..., k im ]; where k i1 k i2 ..., k im Let represent the 1st, 2nd, ..., mth real-time physical state data in the i-th working stage, where m is a positive integer;

[0014] Based on the preset expected physical state data [K] for the i-th working stage i1 K i2 ... K im ], where K i1 K i2 ... K im These represent the 1st, 2nd, ..., mth desired physical state data in the preset i-th working stage; the actual physical state error is calculated and represented as: W_i = Σ u=1 u=m |k iu -K iu | / m; where W_i represents the actual physical state error of the i-th working stage; u represents the identifier of the working stage;

[0015] The system acquires real-time physical state data and actual physical state error of the X-wheel smart chip bucket when loading corresponding materials in historical data, and analyzes the characteristic state parameters affecting different working stages, specifically:

[0016] ; where y v represents the correlation coefficient between the v-th physical state data and the actual physical state error in different working stages, where v is a positive integer and v∈[1,m]; This indicates the acquisition of the v-th real-time physical state data during the i-th working stage when the X-wheel smart chip bucket was loaded with the corresponding material in the history; This indicates the acquisition of the v-th desired physical state data in the i-th working stage when the X-wheel Smart Chip Bucket is loaded with the corresponding material in the history; This represents the average value of the actual physical state error when the X-wheel Smart Chip Bucket was loaded with the corresponding material in the historical data.

[0017] Based on the above method, the correlation coefficients between several physical state data and the actual physical state error in different working stages are calculated sequentially. Physical state data exceeding a preset correlation coefficient threshold are identified as characteristic state parameters. Characteristic state parameters for different working stages are then obtained. The characteristic state parameter for the i-th working stage is denoted as [G]. i1 G i2 ... G iL ]; G i1 G i2 ... G iL Let m and m represent the 1st, 2nd, ..., Lth characteristic state parameters, respectively, where L is a positive integer and L≤m;

[0018] Furthermore, based on the n working stages, several environmental factors are obtained when the characteristic state parameters of the smart chip bucket change during historical operation. By using the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables, the relationship between the characteristic state parameters and environmental factors is fitted based on x historical observation samples.

[0019] Among them, several environmental factors that cause the characteristic state parameters of the smart chip bucket to change during the i-th working stage are denoted as [h]. i1 h i2 ... h il ]; h i1 h i2 ... h il Let l represent the 1st, 2nd, ..., lth environmental factors, where l is a positive integer;

[0020] The fitting formula is characterized as: G ij =α0+Σ s α s *hxs +Σ s,e α se *h xs *h xe Among them, G ij Let αj represent the observed value of the j-th feature state parameter in the i-th working stage of the x-th sample, where j is the identifier of the feature state parameter, j is a positive integer, and j∈[1,L]; α0 represents the fitting constant term; αj s α represents the regression coefficient of the s-th first-order term; se h represents the coefficient of the s-th and e-th second-order cross terms; s and e are positive integers, s and e ∈ [1, l]; xs h represents the s-th environmental factor in the x-th sample; xe This represents the e-th environmental factor in the x-th sample;

[0021] Furthermore, it also includes predicting characteristic state parameters based on environmental factors obtained from the fitting formula, with the predicted value denoted as G. ij ’ ;

[0022] By adjusting the observed samples x that fit the relationship between the characteristic state parameters and environmental factors based on the difference between the observed and predicted values ​​of the characteristic state parameters, the best fitting expression is determined, specifically:

[0023] R 2 =1-[Σ x (G ij ’ -G ij ) 2 ] / [Σ x (G ij -G ij ’’ ) 2 ]; where R 2 R represents the goodness-of-fit index. 2 ∈[0,1];G ij ’’ This represents the historical average value of the characteristic state parameters;

[0024] Different historical observation samples x were selected for fitting, and the goodness-of-fit index R was calculated based on the magnitude of the goodness-of-fit index. 2 The largest fitted expression is determined as the best fitted expression;

[0025] Furthermore, specifically:

[0026] For each working stage, baseline environmental conditions are set for the characteristic state parameters, where the baseline environmental factors for the i-th working stage are denoted as [h]. i1,base h i2,base ... h il,base ], hi1,base h i2,base ... h il,base These are the first, second, ..., lth reference environmental factors; the preset reference values ​​G for the characteristic state parameters under these reference environments are... ij，base ;

[0027] Real-time collection of current environmental factors [h i1,curr h i2,curr ... h il,curr ]; where h i1,curr h i2,curr ... h il,curr Let each of the following represent the first, second, ..., lth current environmental factors; substitute them into the best-fit expression to calculate the predicted value G of the characteristic state parameter under the current environment. ij，pred ’ ;

[0028] Based on the preset characteristic state parameter reference value G under the reference environment of the i-th working stage. ij，base and the predicted value G of the characteristic state parameters under the current environment ij，pred ’ Calculate the characteristic error E ij =|G ij，pred ’ -G ij，base | If the feature error exceeds its preset threshold Th ij If so, the characteristic state parameter is marked as abnormal, and a real-time warning is issued;

[0029] Optionally, the preset threshold Th ij One calculation method is as follows:

[0030] Collect normal samples from group F (environmental factors fluctuate within a reasonable range, and material state parameters are normal) from the i-th stage in history; for each group of samples, calculate the environmental factors [h]. i1,F h i2,F ... h il,F The corresponding predicted value G ij，pred，F ’ ;

[0031] Calculate the historical normal error: E ij,F =|G ij，pred,F ’ -G ij，base |;Calculate the mean μ of the normal samples in group F. ij and standard deviation σ ij Preset threshold Th ij =μ ij +bσ ij Where b is the safety factor.

[0032] The IoT-based smart bucket status data monitoring and management system includes a data acquisition module, a feature selection module, a fitting analysis module, and a real-time early warning module.

[0033] The data acquisition module is used to collect physical state data of the substances inside the bucket in real time based on the multi-sensor monitoring device integrated into the smart bucket body; it also includes the collection of environmental data. The multi-sensor monitoring device transmits the physical state data of the substances inside the bucket through the Internet of Things.

[0034] The feature selection module is used to divide the smart chip bucket into several working stages, and obtain the feature state parameters of different working stages through phased data analysis.

[0035] The fitting analysis module is used to analyze the environmental factors that affect the changes in the characteristic state parameters. By using the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables, it fits the relationship between the changes in the characteristic state parameters and the environmental factors based on several historical observation samples; and determines the best fitting expression through the goodness-of-fit index.

[0036] The real-time early warning module is used to collect current environmental factors in real time, calculate the predicted value of the characteristic state parameter under the current environment through the best fitting expression, calculate the characteristic error degree based on the preset characteristic state parameter benchmark value, and issue a real-time early warning when the preset threshold is exceeded.

[0037] Compared with existing technologies, the beneficial effects of this invention are as follows: By integrating a multi-sensor monitoring device into the smart container and combining it with IoT technology, this invention achieves real-time monitoring of the contents of the container across multiple stages and parameters throughout the entire working cycle, demonstrating significant technical advantages and practical effects. First, by employing stage division and normalization processing, it can comprehensively collect and uniformly process data from different physical states, achieving refined monitoring throughout the entire process. Second, by constructing a physical state error index and combining it with correlation analysis methods, it automatically extracts key characteristic parameters closely related to state deviations, improving data processing efficiency and diagnostic accuracy. Furthermore, this invention introduces a dynamic fitting modeling mechanism between environmental factors and characteristic parameters, integrating first-order and second-order cross-terms to establish a highly fitted prediction model, and selecting the optimal expression based on historical samples, effectively improving prediction accuracy. On this basis, by statistically analyzing error fluctuations under historical normal conditions, it adaptively sets early warning thresholds, achieving real-time early warning responses to state anomalies caused by environmental disturbances. The overall solution possesses good robustness, adaptability, and versatility, suitable for various application scenarios such as chemical containers and liquid substance storage, significantly improving the system's intelligent monitoring level and safety assurance capabilities. Attached Figure Description

[0038] Figure 1This is a flowchart illustrating the IoT-based smart bucket status data monitoring and management method of the present invention. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] Example: Figure 1 As shown, the present invention provides a technical solution, a method for monitoring and managing the status data of a smart bucket based on the Internet of Things, which specifically includes the following steps:

[0041] The smart bucket integrates a multi-sensor monitoring device to collect real-time physical state data of the substances inside the bucket; the multi-sensor monitoring device transmits the physical state data of the substances inside the bucket through the Internet of Things.

[0042] By integrating multiple sensing devices such as temperature sensors, pressure sensors, and concentration detection devices into the smart tank, the physical state data of the substances inside the tank (such as temperature, pressure, and liquid level changes) are collected in real time. The data is then remotely transmitted to the management terminal via IoT technology, enabling dynamic perception of the state inside the tank and real-time data exchange.

[0043] The smart chip bucket is divided into several working stages, and the characteristic state parameters of different working stages are obtained through phased data analysis.

[0044] The entire lifecycle of the smart chip bucket is divided into different stages. Based on the characteristics of the material in each stage, correlation analysis is used to screen out the characteristic parameters that have the greatest impact on the state of the material, so as to achieve precise monitoring in stages and avoid indiscriminate monitoring of all parameters throughout the entire lifecycle.

[0045] Focus on the core risk points at each stage, reduce data processing costs, and improve monitoring efficiency;

[0046] Specifically:

[0047] The physical state data of the substances inside the smart chip bucket are collected throughout its entire working cycle; the smart chip bucket is divided into several working stages [s1, s2, ..., sn]; where s1, s2, ..., sn represent the 1st, 2nd, ..., nth working stages respectively; the physical state data of the substances inside the bucket in each stage are extracted respectively, and the normalization method is used to unify the dimensions between different physical state data.

[0048] For the i-th working stage, the real-time physical state data of the substances inside the container is represented as [k i1 ki2 ..., k im ]; where k i1 k i2 ..., k im Let represent the 1st, 2nd, ..., mth real-time physical state data in the i-th working stage, where m is a positive integer;

[0049] Based on the preset expected physical state data [K] for the i-th working stage i1 K i2 ... K im ], where K i1 K i2 ... K im These represent the 1st, 2nd, ..., mth desired physical state data in the preset i-th working stage; the actual physical state error is calculated and represented as: W_i = Σ u=1 u=m |k iu -K iu | / m; where W_i represents the actual physical state error of the i-th working stage; u represents the identifier of the working stage;

[0050] It should be noted that the container is equipped with a chip that has pressure and vibration sensing capabilities. The chip can display the remaining amount on the external screen of the container. The chip has a positioning function that can track the location of the container. It can detect violent collisions and issue an alarm in time. The container can be recycled and its historical liquid records can be tracked. The container can be easily repackaged using a ton container.

[0051] In this embodiment, the smart chip bucket is divided into several working stages, including the loading stage, storage stage, fermentation stage and transportation stage.

[0052] Physical state data includes the temperature of the loaded material during loading, the inner wall temperature of the smart tank, the concentration of the loaded material, and the pressure of liquid level changes.

[0053] The above data is based on the multi-sensor monitoring device integrated into the smart bucket body. The multi-sensor monitoring device includes, but is not limited to, temperature and humidity sensors, pressure sensors, and concentration detection devices.

[0054] The system acquires real-time physical state data and actual physical state error of the X-wheel smart chip bucket when loading corresponding materials in historical data, and analyzes the characteristic state parameters affecting different working stages, specifically:

[0055] ; where y v represents the correlation coefficient between the v-th physical state data and the actual physical state error in different working stages, where v is a positive integer and v∈[1,m]; This indicates the acquisition of the v-th real-time physical state data during the i-th working stage when the X-wheel smart chip bucket was loaded with the corresponding material in the history; This indicates the acquisition of the v-th desired physical state data in the i-th working stage when the X-wheel Smart Chip Bucket is loaded with the corresponding material in the history; This represents the average value of the actual physical state error when the X-wheel Smart Chip Bucket was loaded with the corresponding material in the historical data.

[0056] Based on the above method, the correlation coefficients between several physical state data and the actual physical state error in different working stages are calculated sequentially. Physical state data exceeding a preset correlation coefficient threshold are identified as characteristic state parameters. Characteristic state parameters for different working stages are then obtained. The characteristic state parameter for the i-th working stage is denoted as [G]. i1 G i2 ... G iL ]; G i1 G i2 ... G iL Let m and m represent the 1st, 2nd, ..., Lth characteristic state parameters, respectively, where L is a positive integer and L≤m;

[0057] In this embodiment, a mixed solution is used as a reference, and the characteristic state parameter of the loading stage is temperature. When the mixed solution exceeds the temperature range, the solution will separate into layers, which will lead to a decrease in its function. If the temperature is too high, it may even lead to loss of function. Therefore, the temperature during loading must be strictly controlled.

[0058] The characteristic state parameters of the storage stage include temperature and pressure. When the pressure exceeds the upper limit, a chemical reaction of dissolved oxygen will occur, which will cause the function of the mixed solution to decline. In this embodiment, the mixed solution is stored under vacuum and inert gas respectively to avoid the influence of pressure. However, the operation is too cumbersome and the temperature will also be affected to some extent. Therefore, pressure detection is used as the characteristic state parameter of the storage stage.

[0059] The fermentation stage, also known as the long-term storage stage, may experience delays in transportation after loading. Therefore, observations are conducted during the long-term storage stage. The long-term storage stage remains highly sensitive to temperature, but as the storage time increases, the temperature variation range of the solution varies with changes in ambient temperature. However, when the pressure stabilizes within a certain range, its sensitivity to pressure decreases significantly. Therefore, the characteristic state parameter of the fermentation stage, also known as the long-term storage stage, is temperature.

[0060] During the transportation phase, due to the different means of transportation, the solution will slosh when bumps occur, causing foam to be generated in the solution. However, no change in its function was found. But the barrel with sealing problems will leak partially. Therefore, the liquid level change range is used as a characteristic state parameter of the transportation phase.

[0061] The environmental factors affecting the changes in the characteristic state parameters are analyzed. The relationship between the characteristic state parameters and the environmental factors is fitted based on several historical observation samples, with the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables. The best fit expression is determined by the goodness-of-fit index.

[0062] Based on historical observation samples, with environmental factors as independent variables and characteristic state parameters as dependent variables, a quantitative relationship model between the two is established through nonlinear regression fitting, and the best model is selected through the goodness-of-fit index to achieve predictive calculation of characteristic parameters.

[0063] By analyzing environmental factors, changes in characteristic parameters can be predicted in advance, allowing for preventative measures to be taken before actual anomalies occur, thus avoiding mass damage to materials.

[0064] Furthermore, based on the n working stages, several environmental factors are obtained when the characteristic state parameters of the smart chip bucket change during historical operation. By using the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables, the relationship between the characteristic state parameters and environmental factors is fitted based on x historical observation samples.

[0065] Among them, several environmental factors that cause the characteristic state parameters of the smart chip bucket to change during the i-th working stage are denoted as [h]. i1 h i2 ... h il ]; h i1 h i2 ... h il Let l represent the 1st, 2nd, ..., lth environmental factors, where l is a positive integer;

[0066] The fitting formula is characterized as: G ij =α0+Σ s α s *h xs +Σ s,e α se *h xs *h xe Among them, G ij Let αj represent the observed value of the j-th feature state parameter in the i-th working stage of the x-th sample, where j is the identifier of the feature state parameter, j is a positive integer, and j∈[1,L]; α0 represents the fitting constant term; αj s α represents the regression coefficient of the s-th first-order term; se h represents the coefficient of the s-th and e-th second-order cross terms; s and e are positive integers, s and e ∈ [1, l]; xs h represents the s-th environmental factor in the x-th sample; xe This represents the e-th environmental factor in the x-th sample;

[0067] In this embodiment, the characteristic state parameters of the storage stage are fitted with environmental factors.

[0068] The characteristic state parameters during the storage stage include temperature and pressure. When the pressure exceeds the upper limit, a chemical reaction of dissolved oxygen occurs, which leads to a decrease in the function of the mixed solution. However, it was found that the ambient temperature affects the temperature inside the container. Therefore, the temperature of the storage environment and the current atmospheric pressure are used as independent variables to fit the characteristic state parameters, and the parameters are solved by the least squares method.

[0069] In this embodiment, the result is obtained through nonlinear regression fitting using the curve_fit function in Python.

[0070] Furthermore, it also includes predicting characteristic state parameters based on environmental factors obtained from the fitting formula, with the predicted value denoted as G. ij ’ ;

[0071] By adjusting the observed samples x that fit the relationship between the characteristic state parameters and environmental factors based on the difference between the observed and predicted values ​​of the characteristic state parameters, the best fitting expression is determined, specifically:

[0072] R 2 =1-[Σ x (G ij ’ -G ij ) 2 ] / [Σ x (G ij -G ij ’’ ) 2 ]; where R 2 R represents the goodness-of-fit index. 2 ∈[0,1];G ij ’’ This represents the historical average value of the characteristic state parameters;

[0073] Different historical observation samples x were selected for fitting, and the goodness-of-fit index R was calculated based on the magnitude of the goodness-of-fit index. 2 The largest fitted expression is determined as the best fitted expression;

[0074] It should be noted that the selection of different observation samples x requires that they be for the same loaded material. Different numbers of continuous observation samples can be selected, and the best fitting expression with the highest goodness of fit index is selected as the best fitting expression.

[0075] Real-time acquisition of current environmental factors, calculation of predicted values ​​of characteristic state parameters under the current environment through the best fit expression, calculation of characteristic error degree based on preset characteristic state parameter benchmark values, and real-time warning when exceeding preset threshold.

[0076] By presetting the baseline values ​​of characteristic parameters for each stage and combining them with the predicted values ​​of current environmental factors, the characteristic error degree is calculated. When the error exceeds the preset threshold, a real-time warning is triggered to achieve timely response to abnormal states.

[0077] By quantifying error rates and thresholds, anomalies can be quickly identified and warnings issued, preventing substances from losing their function due to persistent anomalies.

[0078] Furthermore, specifically:

[0079] For each working stage, baseline environmental conditions are set for the characteristic state parameters, where the baseline environmental factors for the i-th working stage are denoted as [h]. i1,base h i2,base ... h il,base ], h i1,base h i2,base ... h il,base These are the first, second, ..., lth reference environmental factors; the preset reference values ​​G for the characteristic state parameters under these reference environments are... ij，base ;

[0080] Real-time collection of current environmental factors [h i1,curr h i2,curr ... h il,curr ]; where h i1,curr h i2,curr ... h il,curr Let each of the following represent the first, second, ..., lth current environmental factors; substitute them into the best-fit expression to calculate the predicted value G of the characteristic state parameter under the current environment. ij，pred ’ ;

[0081] Based on the preset characteristic state parameter reference value G under the reference environment of the i-th working stage. ij，base and the predicted value G of the characteristic state parameters under the current environment ij，pred ’ Calculate the characteristic error E ij =|G ij，pred ’ -G ij，base | If the feature error exceeds its preset threshold Th ij If so, the characteristic state parameter is marked as abnormal, and a real-time warning is issued;

[0082] By using the best-fit expression, the first step is to make predictive prevention by monitoring before the characteristic state parameters are actually detected, thus preventing the massive waste of resources caused by the destruction of materials in batches.

[0083] Second, prevent environmental factors from causing additional changes in characteristic state parameters, thus avoiding detection anomalies caused by process-related environmental influences.

[0084] Optionally, the preset threshold Th ij One calculation method is as follows:

[0085] Collect normal samples from group F (environmental factors fluctuate within a reasonable range, and material state parameters are normal) from the i-th stage in history; for each group of samples, calculate the environmental factors [h]. i1,F h i2,F ... h il,F The corresponding predicted value G ij，pred，F ’ ;

[0086] Calculate the historical normal error: E ij,F =|G ij，pred,F ’ -G ij，base |;Calculate the mean μ of the normal samples in group F. ij and standard deviation σ ij Preset threshold Th ij =μ ij +bσ ij Where b is the safety factor.

[0087] In this embodiment, b is selected as 3, which can be set according to the actual situation;

[0088] The IoT-based smart bucket status data monitoring and management system includes a data acquisition module, a feature selection module, a fitting analysis module, and a real-time early warning module.

[0089] The data acquisition module is used to collect physical state data of the substances inside the bucket in real time based on the multi-sensor monitoring device integrated in the smart bucket body; it also includes the collection of environmental data. The multi-sensor monitoring device transmits the physical state data of the substances inside the bucket through the Internet of Things.

[0090] The feature selection module is used to divide the smart chip bucket into several working stages and obtain the characteristic state parameters of different working stages through data stage analysis.

[0091] The fitting analysis module is used to analyze the environmental factors that affect the changes in the characteristic state parameters. By using the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables, it fits the relationship between the changes in the characteristic state parameters and the environmental factors based on several historical observation samples; and determines the best fitting expression through the goodness-of-fit index.

[0092] The real-time early warning module is used to collect current environmental factors in real time, calculate the predicted value of the characteristic state parameters under the current environment through the best fitting expression, calculate the characteristic error degree based on the preset characteristic state parameter benchmark value, and issue a real-time early warning when the preset threshold is exceeded.

[0093] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for monitoring and managing the status data of a smart bucket based on the Internet of Things, characterized in that: Specifically, the steps include the following: The smart bucket integrates a multi-sensor monitoring device to collect real-time physical state data of the substances inside the bucket; the multi-sensor monitoring device transmits the physical state data of the substances inside the bucket through the Internet of Things. The smart chip bucket is divided into several working stages, and the characteristic state parameters of different working stages are obtained through phased data analysis. The environmental factors affecting the changes in the characteristic state parameters are analyzed. The relationship between the characteristic state parameters and the environmental factors is fitted based on several historical observation samples, with the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables. The best fit expression is determined by the goodness-of-fit index. Real-time acquisition of current environmental factors, calculation of predicted values ​​of characteristic state parameters under the current environment through the best fit expression, calculation of characteristic error degree based on preset characteristic state parameter benchmark values, and real-time warning when exceeding preset threshold.

2. The IoT-based smart chip bucket status data monitoring and management method according to claim 1, characterized in that: include: The physical state data of the substances inside the smart chip bucket are collected throughout its entire working cycle. The smart chip bucket is divided into several working stages [s1, s2, ..., sn], where s1, s2, ..., sn represent the 1st, 2nd, ..., nth working stages, respectively. The physical state data of the substances inside the bucket in each stage are extracted, and the normalization method is used to unify the dimensions between different physical state data.

3. The IoT-based smart chip bucket status data monitoring and management method according to claim 2, characterized in that: include: For the i-th working stage, the real-time physical state data of the substances inside the container is represented as [k i1 k i2 ..., k im ]; where k i1 k i2 ..., k im Let represent the 1st, 2nd, ..., mth real-time physical state data in the i-th working stage, where m is a positive integer; Based on the preset expected physical state data [K] for the i-th working stage i1 K i2 ... K im ], where K i1 K i2 ... K im These represent the 1st, 2nd, ..., mth desired physical state data in the preset i-th working stage; the actual physical state error is calculated and represented as: W_i = Σ u=1 u=m |k iu -K iu | / m; where W_i represents the actual physical state error of the i-th working stage; u represents the identifier of the working stage.

4. The IoT-based smart chip bucket status data monitoring and management method according to claim 3, characterized in that: include: The system acquires real-time physical state data and actual physical state error of the X-wheel smart chip bucket when loading corresponding materials in historical data, and analyzes the characteristic state parameters affecting different working stages, specifically: ; where y v represents the correlation coefficient between the v-th physical state data and the actual physical state error in different working stages, where v is a positive integer and v∈[1,m]; This indicates the acquisition of the v-th real-time physical state data during the i-th working stage when the X-wheel smart chip bucket was loaded with the corresponding material in the history; This indicates the acquisition of the v-th desired physical state data in the i-th working stage when the X-wheel Smart Chip Bucket is loaded with the corresponding material in the history; This represents the average value of the actual physical state error when the X-wheel Smart Chip Bucket was loaded with the corresponding material in the historical data. Based on the above method, the correlation coefficients between several physical state data and the actual physical state error in different working stages are calculated sequentially. Physical state data exceeding a preset correlation coefficient threshold are identified as characteristic state parameters. Characteristic state parameters for different working stages are then obtained. The characteristic state parameter for the i-th working stage is denoted as [G]. i1 G i2 ... G iL ]; G i1 G i2 ... G iL Let L represent the 1st, 2nd, ..., Lth characteristic state parameters, where L is a positive integer and L≤m.

5. The IoT-based smart chip bucket status data monitoring and management method according to claim 4, characterized in that: include: Based on the n working stages, several environmental factors are obtained when the characteristic state parameters of the smart chip bucket change during historical operation. By using the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables, the relationship between the characteristic state parameters and environmental factors is fitted based on x historical observation samples. Among them, several environmental factors that cause the characteristic state parameters of the smart chip bucket to change during the i-th working stage are denoted as [h]. i1 h i2 ... h il ]; h i1 h i2 ... h il Let l represent the 1st, 2nd, ..., lth environmental factors, where l is a positive integer; The fitting formula is characterized as: G ij =α0+Σ s α s *h xs +Σ s,e α se *h xs *h xe Among them, G ij Let αj represent the observed value of the j-th feature state parameter in the i-th working stage of the x-th sample, where j is the identifier of the feature state parameter, j is a positive integer, and j∈[1,L]; α0 represents the fitting constant term; αj s α represents the regression coefficient of the s-th first-order term; se h represents the coefficient of the s-th and e-th second-order cross terms; s and e are positive integers, s and e ∈ [1, l]; xs h represents the s-th environmental factor in the x-th sample; xe This represents the e-th environmental factor in the x-th sample.

6. The IoT-based smart chip bucket status data monitoring and management method according to claim 5, characterized in that: Also includes: Based on the fitted formula, environmental factors are used to predict characteristic state parameters, and the predicted value is denoted as G. ij ’ ; By adjusting the observed samples x that fit the relationship between the characteristic state parameters and environmental factors based on the difference between the observed and predicted values ​​of the characteristic state parameters, the best fitting expression is determined, specifically: R 2 =1-[Σ x (G ij ’ -G ij ) 2 ] / [Σ x (G ij -G ij ’’ ) 2 ]; where R 2 R represents the goodness-of-fit index. 2 ∈[0,1];G ij ’’ This represents the historical average value of the characteristic state parameters; Different historical observation samples x were selected for fitting, and the goodness-of-fit index R was calculated based on its magnitude. 2 The largest fitted expression is determined as the best fitted expression.

7. The IoT-based smart chip bucket status data monitoring and management method according to claim 6, characterized in that: Specifically: For each working stage, baseline environmental conditions are set for the characteristic state parameters, where the baseline environmental factors for the i-th working stage are denoted as [h]. i1,base h i2,base ... h il,base ], h i1,base h i2,base ... h il,base These are the first, second, ..., lth reference environmental factors; the preset reference values ​​G for the characteristic state parameters under these reference environments are... ij，base ; Real-time collection of current environmental factors [h i1,curr h i2,curr ... h il,curr ]; where h i1,curr h i2,curr ... h il,curr Let each of the following represent the first, second, ..., lth current environmental factors; substitute them into the best-fit expression to calculate the predicted value G of the characteristic state parameter under the current environment. ij，pred ’ ; Based on the preset characteristic state parameter reference value G under the reference environment of the i-th working stage. ij，base and the predicted value G of the characteristic state parameters under the current environment ij，pred ’ Calculate the characteristic error E ij =|G ij，pred ’ -G ij，base | If the feature error exceeds its preset threshold Th ij If the characteristic state parameter is abnormal, a real-time warning will be issued.

8. The IoT-based smart chip bucket status data monitoring and management method according to claim 7, characterized in that: include: Wherein, the preset threshold Th ij One calculation method is as follows: Collect F normal samples from the i-th stage of history; for each sample group, calculate the environmental factors [h]. i1,F h i2,F ... h il,F The corresponding predicted value G ij，pred，F ’ ; Calculate the historical normal error: E ij,F =|G ij，pred,F ’ -G ij，base |;Calculate the mean μ of the normal samples in group F. ij and standard deviation σ ij Preset threshold Th ij =μ ij +bσ ij Where b is the safety factor.

9. A smart chip bucket status data monitoring and management system based on the Internet of Things, employing the smart chip bucket status data monitoring and management method based on the Internet of Things as described in any one of claims 1-8, characterized in that: It includes a data acquisition module, a feature selection module, a fitting analysis module, and a real-time early warning module; The data acquisition module is used to collect physical state data of the substances inside the bucket in real time based on the multi-sensor monitoring device integrated in the smart bucket body; it also includes the collection of environmental data. The multi-sensor monitoring device transmits the physical state data of the substances inside the bucket through the Internet of Things. The feature selection module is used to divide the smart chip bucket into several working stages, and obtain the feature state parameters of different working stages through phased data analysis. The fitting analysis module is used to analyze the environmental factors that affect the changes in the characteristic state parameters. By using the environmental factors as independent variables and the corresponding characteristic state parameters as dependent variables, it fits the relationship between the changes in the characteristic state parameters and the environmental factors based on several historical observation samples; and determines the best fitting expression through the goodness-of-fit index. The real-time early warning module is used to collect current environmental factors in real time, calculate the predicted value of the characteristic state parameter under the current environment through the best fitting expression, calculate the characteristic error degree based on the preset characteristic state parameter benchmark value, and issue a real-time early warning when the preset threshold is exceeded.

Images