Internet of Things weighing monitoring method and system based on iron and steel industry

Abstract

The invention relates to the technical field of iron and steel industry, in particular to an internet of things weighing monitoring method and system based on the iron and steel industry, and the method comprises the following steps: continuously collecting the weight information of steel materials based on a weight sensor installed on a conveyor belt, recording the weight value and the corresponding timestamp of each material in real time, and removing the noise and abnormal values in the data; and obtaining a cleaned weight data set. According to the invention, the real-time change of the weight of the material is accurately monitored through the time sequence analysis and anomaly detection technology, so that the iron and steel industry can accurately control the use of the material, unnecessary waste is avoided, the real-time performance and accuracy of data processing are improved, production personnel are allowed to discover production deviation in real time through continuously updated weight prediction information, and the production efficiency is improved. According to the method, rapid intervention, rapid identification of abnormal weight points and reason analysis are carried out, material loss caused by misoperation or equipment faults can be greatly reduced, and the continuity of the production process is further ensured by adjusting production parameters in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of steel industry, and in particular to an Internet of Things weighing monitoring method and system based on the steel industry. Background Art

[0002] The steel industry is one of the important pillars of the global economy, involving all stages from the processing of raw materials to the production of final products. The industry is particularly dependent on large-scale production facilities and complex mechanical systems to process and transform raw materials, such as iron ore. With the development of technology, Internet of Things (IoT) technology has been introduced into the industry to enhance production efficiency, safety and sustainability. The application of IoT makes real-time data collection and monitoring possible, thereby optimizing energy management, reducing resource waste, and improving the automation level of production lines.

[0003] Among them, the IoT weighing monitoring method of the steel industry involves the use of IoT technology to implement precise weighing monitoring in the steel production process, aiming to monitor the weight of materials in real time, which is essential for ensuring product quality, optimizing production processes and reducing waste of raw materials. By deploying sensors and network equipment, it can automatically collect weight data and realize remote monitoring, so that operators can adjust production parameters in time and improve production efficiency.

[0004] The monitoring of material weight in the existing technology relies on intermittent inspection or non-real-time feedback, resulting in obvious time lags in data collection and processing. Due to the lack of real-time monitoring capabilities, it is difficult to promptly detect and adjust weight errors in production. This delayed response not only affects production efficiency, but also causes waste of raw materials and energy. The limitations of existing technologies in data analysis capabilities, especially in data noise processing and outlier identification in dynamic production environments, make it impossible to optimize the production process, often leading to overproduction or substandard products, adding additional cost burdens. For example, failure to promptly adjust weight deviations caused by fluctuations in raw material quality will cause the finished product to fail to meet quality standards. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an Internet of Things weighing monitoring method and system based on the steel industry.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solution: an Internet of Things weighing monitoring method based on the steel industry comprises the following steps: S1: Based on the weight sensor installed on the conveyor belt, the weight information of steel materials is continuously collected, the weight value and corresponding timestamp of each material are recorded in real time, and the noise and outliers in the data are removed to obtain the cleaned weight data set; S2: Based on the weight data set after cleaning, a time series analysis is performed on the weight change of the material, the weight change trend of the material at different time points is evaluated, and the weight of the material in the future time period is predicted to obtain the material weight prediction information; S3: Based on the material weight prediction information, the difference between the predicted weight of the material and the current monitored weight is calculated, abnormal weight points beyond the normal range are identified, and the detected abnormal points are analyzed to determine potential weighing errors or operational errors, and obtain abnormal analysis records; S4: Based on the abnormal analysis record, the running speed of the conveyor belt and the distribution parameters of the material are adjusted to optimize the weight distribution, and the adjusted weighing process is monitored again to verify the effect of the adjustment execution and obtain the weighing adjustment verification result.

[0007] The present invention is improved in that the steps of recording the weight value of the material and the corresponding timestamp are specifically as follows: S111: Based on the weight sensor installed on the conveyor belt, the steel material passing through is monitored, and the weight of each material is recorded to obtain the original weight record; S112: Based on the original weight record, add a timestamp to each weight data, using the formula: ; Get the weight value and corresponding timestamp of each material, where: It is The timestamp of the material, Indicates the time when the measurement starts. is the time interval between two data recorded by the sensor, Is the material serial number.

[0008] The present invention is improved in that the steps of obtaining the cleaned weight data set are specifically as follows: S121: Based on the weight value and corresponding timestamp of each material, data exceeding the normal weight range of the material or timestamp error records are removed to obtain initially processed weight data; S122: Continue to analyze the initially processed weight data, and use a box plot method to identify and remove statistical outliers to obtain a cleaned weight data set.

[0009] The present invention is improved in that the step of evaluating the weight change trend is specifically as follows: S211: extracting the weight record of each material at the different time points based on the cleaned weight data set to obtain a data set associated with time and weight; S212: Perform time series analysis on the associated data set of time and weight to evaluate the change trend of material weight using the formula: ; The weight trend analysis results are obtained, where: represents the predicted weight at the current time point, Indicates the weight value at the previous time point, represents the average weight of historical data, is the complement of the smoothing coefficient.

[0010] The present invention is improved in that the step of obtaining the material weight prediction information is specifically as follows: S221: using the ARIMA model, determining the initial prediction parameters according to the historical records of material weight, including the difference degree and the window size of the moving average, and obtaining the adjusted ARIMA model; S222: Based on the adjusted ARIMA model, the formula is: ; Predict the material weight in the future time period to obtain the material weight prediction information, where: Indicates the current time point The predicted weight change value of and Respectively represent the material weight at the previous and previous two time points, is the average of the historical weight changes, and is the adjustment factor.

[0011] The present invention is improved in that the step of identifying the abnormal weight point is specifically as follows: S311: Based on the material weight prediction information, extract the currently monitored material weight, and perform data integration to obtain a data table of current weight and predicted weight; S312: Based on the data table of the current weight and the predicted weight, the difference between the current weight and the predicted weight is calculated using the formula: ; Get the weight difference ,in, Indicates the current weight of the material obtained from real-time monitoring. Represents the predicted value of material weight at the same point in time: S313: Based on the weight difference value, compare with the difference threshold, identify and mark abnormal weight points that exceed the threshold.

[0012] The present invention is improved in that the step of obtaining the abnormal analysis record is specifically as follows: S321: Classify the abnormal weight points, determine the potential abnormal causes according to the weight difference values ​​and the frequency of occurrence, and obtain a preliminary abnormal point classification table; S322: Analyze the preliminary abnormal point classification table, calculate potential weighing errors or operational errors, and obtain abnormal analysis records.

[0013] The present invention is improved in that the steps of obtaining the weighing adjustment verification result are specifically as follows: S411: Based on the abnormal analysis record, collect the real-time weight after the conveyor belt speed adjustment and the material distribution parameter adjustment, including the material weight value, timestamp and the operating parameters of the corresponding equipment, and obtain a new parameter data record table; S412: Analyze the new parameter data record table, and compare the data difference before and after adjustment, using the formula: ; Efficiency index for calculating adjustment effect ,in, Represents the adjusted weight data, Represents the weight data before adjustment; S413: Based on the efficiency index, determine the success degree of the adjustment and obtain a weighing adjustment verification result.

[0014] An IoT weighing monitoring system based on the steel industry, the system includes: The weight data acquisition module continuously collects the weight information of steel materials based on the weight sensor installed on the conveyor belt, records the weight value and corresponding timestamp of each material, and removes noise and outliers in the data to obtain a cleaned weight data set; The weight data analysis module performs a time series analysis on the weight change of the material based on the cleaned weight data set, evaluates the weight change trend of the material at different time points, and predicts the weight of the material in the future time period to obtain the material weight prediction information; The abnormality detection module calculates the difference between the predicted weight of the material and the current monitored weight based on the material weight prediction information, identifies abnormal weight points beyond the normal range, and analyzes the detected abnormal points to determine potential weighing errors or operational errors, and obtains abnormality analysis records; The adjustment execution optimization module adjusts the running speed of the conveyor belt according to the abnormal analysis record, optimizes the material distribution and weight distribution, and obtains weighing adjustment information; The monitoring and verification module monitors the weighing process after adjustment again based on the weighing adjustment information, verifies the execution effect of the adjustment, and obtains a weighing adjustment verification result.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by applying time series analysis and anomaly detection technology, the real-time changes of material weight are accurately monitored, so that the steel industry can accurately control the use of materials, avoid unnecessary waste, and improve the real-time and accuracy of data processing. Through the continuously updated weight prediction information, production personnel are allowed to immediately discover production deviations and intervene quickly. The rapid identification and cause analysis of abnormal weight points can greatly reduce material losses caused by operational errors or equipment failures. By adjusting production parameters in real time, the continuity of the production process is further ensured, which significantly improves the operating efficiency and product quality of the entire production line, while reducing energy consumption and costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention proposes a flow chart of an Internet of Things weighing monitoring method based on the steel industry; Figure 2 A flow chart for recording the weight value of a material and the corresponding timestamp in the present invention; Figure 3 is a flow chart for obtaining a weight data set after cleaning in the present invention; Figure 4 is a flow chart for evaluating the weight change trend in the present invention; Figure 5 It is a flow chart for obtaining material weight prediction information in the present invention; Figure 6 This is a flow chart for identifying abnormal weight points in the present invention; Figure 7 A flowchart for obtaining abnormal analysis records in the present invention; Figure 8 The figure is a flow chart for obtaining the weighing adjustment verification result in the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0019] Example See also Figure 1 The present invention provides a technical solution: an Internet of Things weighing monitoring method based on the steel industry, comprising the following steps: S1: Based on the weight sensor installed on the conveyor belt, the weight information of steel materials is continuously collected, the weight value and corresponding timestamp of each material are recorded in real time, and the noise and outliers in the data are removed to obtain the cleaned weight data set; S2: Based on the weight data set after cleaning, perform time series analysis on the weight change of the material, evaluate the weight change trend of the material at different time points, and predict the weight of the material in the future time period to obtain the material weight prediction information; S3: Based on the material weight prediction information, calculate the difference between the predicted weight of the material and the current monitored weight, identify abnormal weight points beyond the normal range, and analyze the detected abnormal points to determine potential weighing errors or operational errors, and obtain abnormal analysis records; S4: Based on the abnormal analysis records, the conveyor belt running speed and material distribution parameters are adjusted to optimize the weight distribution, and the adjusted weighing process is monitored again to verify the adjustment execution effect and obtain the weighing adjustment verification result.

[0020] The cleaned weight data set includes the filtered weight value, time calibration, and abnormal indicator identification information. The material weight prediction information includes the trend analysis results, predicted weight value, and key time points. The abnormal analysis records include abnormal point identification, error source analysis results, and operational error information. The weighing adjustment verification results include the weighing adjustment effect, real-time monitoring feedback information, and effect confirmation status.

[0021] See also Figure 2 , the specific steps for recording the weight value of the material and the corresponding timestamp are: S111: Based on the weight sensor installed on the conveyor belt, the steel material passing through is monitored, and the weight of each material is recorded to obtain the original weight record; Based on the weight sensor, the weight information of each piece of steel material passing through is continuously collected. It is necessary to continuously detect the collected weight signal according to the moving speed of the material on the conveyor belt and the sampling frequency of the sensor, capture the specific moment when the material on the conveyor belt passes through the sensor, and require the sensor to output the weight signal during the execution process. At the same time, the collected data is arranged in sequence as a time series record, and the time point when the material enters the sensor collection range is determined through the synchronization timing mechanism. Combined with the specific collection order of each weight data and the synchronization of the conveyor belt movement, the weight record is ensured to be complete and in time sequence. The interference noise in the weight signal is eliminated by screening the abnormal values ​​of data fluctuations, and the fixed threshold method is used to eliminate the records that do not meet the reasonable weight range of steel materials, and finally the original weight record after preliminary processing is generated.

[0022] S112: Based on the original weight record, add a timestamp to each weight data using the formula: ; Get the weight value and corresponding timestamp of each material, where: It is The timestamp of the material, Indicates the time when the measurement starts, that is, when the first material starts to pass through the conveyor belt and is recorded by the sensor. It is the time interval between two data recorded by the sensor, that is, the time difference between two consecutive materials. It is the material serial number, indicating the number of materials that pass through the sensor, and is used to calculate the timestamp of each material; The following data were collected: (Unix timestamp 2024-10-27, 17:06:40, representing a specific start time), seconds (time interval between materials), and (the fifth material), then the timestamp of the fifth material is calculated as: ; The result shows that the fifth material passed through the sensor at timestamp 1730020050 (2024-10-27, 17:07:30). This timestamp converted into a regular date and time represents the exact monitoring time of the material, which is crucial for logistics tracking and quality management.

[0023] See also Figure 3 , the steps to obtain the cleaned weight data set are as follows: S121: Based on the weight value and corresponding timestamp of each material, data exceeding the normal weight range of the material or records with wrong timestamps are removed to obtain initially processed weight data; Each weight value and timestamp is extracted from the original record as the data to be processed, the weight threshold parameters of the sensor are called, the weight values ​​are compared one by one, and the records that exceed the normal weight range of the material are removed. According to the synchronization records of the time management system, each timestamp is checked against the standard format one by one, and the timestamp records with logical errors or discontinuities are screened out and eliminated. By gradually updating the cleaned data set, it is ensured that only weight values ​​and timestamps that are within the normal characteristic range of the material are retained to obtain the initially processed weight data.

[0024] S122: Continuously analyzing the initially processed weight data, using a box plot method to identify and remove statistical outliers, and obtaining a cleaned weight data set; Analyze their distribution characteristics one by one, call the box plot method to determine the upper and lower quartile ranges of the data by calculating the quantiles, and calculate the interquartile range based on this, set the elimination threshold to the range outside the upper and lower limits, and eliminate outliers by threshold judgment on all weight records. Further check the uniformity of the distribution of the eliminated data to ensure that the data set after elimination still conforms to the statistical distribution law. By cleaning up the outliers in the final data set, update and save all weight records that meet the statistical standards, and obtain the cleaned weight data set.

[0025] See also Figure 4 , the evaluation steps of weight change trend are as follows: S211: based on the cleaned weight data set, extract the weight record of each material at the difference time point to obtain a data set associated with time and weight; By parsing the timestamp and corresponding weight value of each record in the data set, valid data points that meet the time series requirements are screened out, records with incomplete or duplicate timestamps are removed, and the remaining data are sorted according to timestamps to ensure the continuity and logical correctness of the time series. On this basis, the correspondence between time points and weight values ​​is mapped to a time-weight association structure, which uses timestamps as indexes and weight values ​​as data units. Through the mapping relationship, the weight value at any time point can be easily called, and further analysis and calculation can be performed to obtain a time-weight association data set.

[0026] S212: Perform time series analysis on the time and weight correlation data set to evaluate the change trend of material weight using the formula: ; The weight trend analysis results are obtained, where: represents the predicted weight at the current time point, Indicates the weight value at the previous time point, which is the weight of the most recent record obtained directly from the data set. Represents the average weight of historical data, which is calculated by taking the arithmetic mean of all past data weights. is the complement of the smoothing coefficient, which is used to adjust the contribution of the historical average weight to the forecast. Increase, Decrease, indicating that the impact of the historical average weight on the prediction is reduced; The following data were collected: kg represents the actual weight at the previous time point, kg represents the historical average weight, Represents the smoothing coefficient and calculates the predicted weight at the current time point: ; ; The result shows that based on historical data and recent actual weight, the predicted weight at the current point in time is 148 kg. The prediction helps understand the weight trend over time and provides data support for further material management decisions.

[0027] See also Figure 5 , the specific steps for obtaining material weight prediction information are: S221: using the ARIMA model, determining the initial prediction parameters according to the historical records of material weight, including the difference degree and the window size of the moving average, and obtaining the adjusted ARIMA model; The ARIMA model is used to separate the medium- and long-term trends, seasonal components, and random fluctuations of historical data. The differential scores are calculated step by step based on the characteristics of the data. The differential scores are determined by monitoring the changes in the autocorrelation function (ACF) and partial autocorrelation function (PACF) of the data. It is analyzed whether first-order or high-order differential operations are needed to eliminate non-stationarity. At the same time, the sliding window technology is used to perform moving average processing on the data. The model fitting effect is observed by gradually adjusting the window size. The optimal window size is selected based on the window range with large fluctuations, thereby eliminating the impact of short-term fluctuations and optimizing the prediction accuracy. The adjusted ARIMA model is generated, which can accurately capture the fluctuation law and prediction trend of material weight.

[0028] S222: Based on the adjusted ARIMA model, using the formula: ; Predict the material weight in the future time period to obtain the material weight prediction information, where: Indicates the current time point The predicted weight change value, that is, from the time point arrive The predicted weight difference is and Represents the material weight at the previous and two previous time points, respectively, and is directly obtained from the time series data. is the average of the historical weight changes, and is the adjustment factor, Used to control the influence of weight change between the last two time points. For adjustment Contribution to forecasting; The parameter values ​​are as follows kg The weight of the material at the previous time point, kg Material weight at the first two time points, kg Average value of historical weight changes, Adjust the coefficient to enhance the impact of weight changes at recent time points, Adjustment coefficient to mitigate the impact of historical average changes. Calculation process: ; The result shows that based on the given parameters and historical data, the predicted weight change at a future point in time is an increase of 5 kg.

[0029] See also Figure 6 ,The specific steps for identifying abnormal weight points are: S311: Based on the material weight prediction information, extract the currently monitored material weight, and perform data integration to obtain a data table of current weight and predicted weight; Extract the currently monitored material weight, match the current weight and predicted weight by timestamp, ensure that there is a set of monitoring data and corresponding predicted data at each time point, eliminate data loss or misalignment by checking the consistency of timestamps, format all matching data to form a structured data table containing time, current weight and predicted weight. The time field in the data table ensures the continuity of the data, and the current weight and predicted weight fields are used for subsequent difference calculations to obtain a data table of current weight and predicted weight.

[0030] S312: Based on the data table of current weight and predicted weight, the difference between the current weight and the predicted weight is calculated using the formula: ; Get the weight difference ,in, Indicates the current weight of the material obtained from real-time monitoring. Represents the predicted value of material weight at the same point in time; At a certain point in time, the actual weight obtained from monitoring is 120kg, while the predicted weight given by the forecast is 95kg. Substituting into the formula, the calculation process is as follows: ; The result shows that at the current monitoring moment, the actual weight of the material is 25kg higher than the predicted weight, indicating the existence of an abnormal situation or system error, and the cause needs to be further analyzed.

[0031] S313: Based on the weight difference value, compare with the difference threshold, identify and mark abnormal weight points exceeding the threshold; Call the weight difference calculation result, compare all the difference values ​​with the set threshold one by one, use the judgment conditions to identify all weight points that exceed the threshold, mark the points in chronological order, and the marked abnormal point data includes time, current weight, predicted weight and difference value. At the same time, add an abnormal flag field to the data to ensure that the marked results can be directly associated with other data. Store the marked abnormal weight points for subsequent abnormal analysis and processing, and identify and mark abnormal weight points that exceed the threshold.

[0032] See also Figure 7 , the specific steps for obtaining abnormal analysis records are: S321: Classify the abnormal weight points, determine the potential abnormal causes according to the size of their weight difference values ​​and the frequency of occurrence, and obtain a preliminary abnormal point classification table; The collected abnormal weight point data are arranged in chronological order, grouped according to the size of the weight difference value, the weight range of the group is set and the corresponding time frequency is marked, the weight difference value is calculated as the difference between the actual weighing value and the standard reference value, and the time interval and the frequency of the difference are recorded. Then, the grouped data are called for standard deviation analysis, the fluctuation range of each group of data is calculated, and a fixed threshold is used to determine whether the difference between the data is significant. At the same time, frequency statistics are performed on the groups where abnormal points appear according to the time frequency to determine the potential cause of the abnormality. Subsequently, a preliminary classification table is established for the abnormal groups with significant differences based on the frequency classification results. The weight difference range, occurrence time interval and frequency of each group are recorded in the classification table, and finally a preliminary abnormal point classification table is output.

[0033] S322: Analyze the preliminary abnormal point classification table, calculate potential weighing errors or operating errors, and obtain abnormal analysis records; Select groups with large fluctuations in weight differences for analysis, calculate the average difference, peak value and minimum value of each group based on the data in the classification table, and trace back the abnormal points in the group one by one. By matching the operation time and equipment status recorded in the weighing operation log, determine whether there is a weighing equipment error or human operation error. In the case of equipment error, call the weighing deviation parameter in the equipment self-test record and compare it with the difference value in the abnormal point classification table to calculate the deviation amplitude. For human operation errors, combine the operation time point with the equipment abnormality log record to check whether there are abnormal operation steps. Finally, output the abnormal analysis record according to the above analysis steps.

[0034] See also Figure 8 , the specific steps for obtaining the weighing adjustment verification results are: S411: Based on the abnormal analysis record, collect the real-time weight after the conveyor belt speed adjustment and the material distribution parameter adjustment, including the material weight value, timestamp and the operating parameters of the corresponding equipment, and obtain a new parameter data record table; Real-time data is extracted from the weight monitoring of the conveyor belt, including weight values, timestamps, and adjusted equipment operating parameters. Record entries are established based on each sampling point. The extracted weight data is then initially checked for completeness and accuracy, including eliminating sampling errors, data loss, and illogical data values, and matching timestamps with equipment parameters to form associations. At the same time, all extracted parameters are standardized to ensure consistent data formats and facilitate subsequent analysis. Finally, the cleaned weight values, timestamps, and equipment operating parameters are integrated into a new parameter data record table.

[0035] S412: Analyze the new parameter data record table, and compare the data differences before and after adjustment, using the formula: ; Efficiency index for calculating adjustment effect ,in, Represents the adjusted weight data, which is the actual measured material weight after the conveyor belt speed and material distribution parameters are adjusted. The weight data before adjustment refers to the weight of the material measured under the existing settings before any adjustment is made. The weight data before and after adjustment are as follows: The data set is [10, 20, 30] kg Corresponding The data set is [15, 18, 35] kg Substituting the formula into the calculation process is as follows: ; ; The result shows that the adjusted weight data changed by about 0.0386, or 3.86%, compared with the original data, indicating that the adjustment has a certain effect, but the change is not large and needs further optimization.

[0036] S413: Based on the efficiency index, determine the success degree of the adjustment and obtain the weighing adjustment verification result; The calculated efficiency index is compared with the preset success standard, and the value is classified and judged by the standard value. For example, whether the efficiency index is lower than the preset adjustment success threshold. If it is lower than the threshold, the adjusted parameter configuration is considered reasonable, otherwise the adjustment parameters need to be further refined. Then, combined with the adjustment effect recorded in the historical data, the efficiency index change trend is analyzed, the abnormal frequency rate after adjustment is counted and marked, and a summary report is formed on the weighing operation after adjustment. The report includes efficiency index analysis, the basis for judging success or failure, and necessary further adjustment suggestions, and the weighing adjustment verification result is obtained.

[0037] The IoT weighing monitoring system based on the steel industry includes: The weight data acquisition module continuously collects the weight information of steel materials based on the weight sensor installed on the conveyor belt, records the weight value and corresponding timestamp of each material, and removes noise and outliers in the data to obtain a cleaned weight data set; The weight data analysis module performs time series analysis on the weight changes of materials based on the cleaned weight data set, evaluates the weight change trend of materials at different time points, and predicts the weight of materials in the future time period to obtain material weight prediction information; The abnormality detection module calculates the difference between the predicted weight of the material and the current monitored weight based on the material weight prediction information, identifies abnormal weight points that exceed the normal range, and analyzes the detected abnormal points to determine potential weighing errors or operational errors, and obtains abnormal analysis records; The adjustment execution optimization module adjusts the conveyor belt's running speed according to the abnormal analysis records, optimizes the material distribution and weight distribution, and obtains weighing adjustment information; The monitoring and verification module monitors the weighing process after adjustment again based on the weighing adjustment information, verifies the execution effect of the adjustment, and obtains the weighing adjustment verification result.

[0038] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. The Internet of Things weighing monitoring method based on the steel industry is characterized by: The following steps are involved: Based on the weight sensor installed on the conveyor belt, the weight information of steel materials is continuously collected, the weight value and corresponding timestamp of each material are recorded in real time, and the noise and outliers in the data are removed to obtain the cleaned weight data set; Based on the cleaned weight data set, a time series analysis is performed on the weight change of the material, the weight change trend of the material at different time points is evaluated, and the weight of the material in a future time period is predicted to obtain the material weight prediction information; Based on the material weight prediction information, the difference between the predicted weight of the material and the current monitored weight is calculated, abnormal weight points beyond the normal range are identified, and the detected abnormal points are analyzed to determine potential weighing errors or operational errors, and obtain abnormal analysis records; Based on the abnormal analysis records, the running speed of the conveyor belt and the distribution parameters of the material are adjusted to optimize the weight distribution, and the adjusted weighing process is monitored again to verify the adjustment execution effect and obtain the weighing adjustment verification result.

2. The Internet of Things weighing monitoring method based on the steel industry according to claim 1 is characterized in that: The specific steps for recording the weight value of the material and the corresponding timestamp are: Based on the weight sensor installed on the conveyor belt, the steel materials passing through are monitored, and the weight of each material is recorded to obtain the original weight record; Based on the original weight record, a timestamp is added to each weight data, using the formula: ； Get the weight value and corresponding timestamp of each material, where: It is The timestamp of the material, Indicates the time when the measurement starts. is the time interval between two data recorded by the sensor, Is the material serial number.

3. The Internet of Things weighing monitoring method based on the steel industry according to claim 1 is characterized in that: The steps for obtaining the cleaned weight data set are specifically as follows: Based on the weight value and corresponding timestamp of each material, data exceeding the normal weight range of the material or records with wrong timestamps are removed to obtain initially processed weight data; The initially processed weight data is continuously analyzed, and statistical outliers are identified and removed using a box plot method to obtain a cleaned weight data set.

4. The Internet of Things weighing monitoring method based on the steel industry according to claim 1 is characterized in that: The steps for evaluating the weight change trend are specifically as follows: Based on the cleaned weight data set, extract the weight record of each material at the difference time point to obtain a data set associated with time and weight; A time series analysis is performed on the data set of the association between time and weight to evaluate the change trend of the material weight using the formula: ； The weight trend analysis results are obtained, where: represents the predicted weight at the current time point, Indicates the weight value at the previous time point, represents the average weight of historical data, is the complement of the smoothing coefficient.

5. The Internet of Things weighing monitoring method based on the steel industry according to claim 1 is characterized in that: The steps for obtaining the material weight prediction information are specifically as follows: Using the ARIMA model, the initial prediction parameters, including the difference score and the moving average window size, are determined based on the historical records of material weight to obtain the adjusted ARIMA model. Based on the adjusted ARIMA model, the formula is adopted: ； Predict the material weight in the future time period to obtain the material weight prediction information, where: Indicates the current time point The predicted weight change value of and Respectively represent the material weight at the previous and previous two time points, is the average of the historical weight changes, and is the adjustment factor.

6. The Internet of Things weighing monitoring method based on the steel industry according to claim 1 is characterized in that: The specific steps of identifying abnormal weight points are: Based on the material weight prediction information, extract the currently monitored material weight, and perform data integration to obtain a data table of current weight and predicted weight; Based on the data table of current weight and predicted weight, the difference between the current weight and the predicted weight is calculated using the formula: ； Get the weight difference ,in, Indicates the current weight of the material obtained from real-time monitoring. Represents the predicted value of material weight at the same point in time; Based on the weight difference value, a comparison is made with a difference threshold value, and abnormal weight points exceeding the threshold value are identified and marked.

7. The Internet of Things weighing monitoring method based on the steel industry according to claim 1 is characterized in that: The steps for obtaining the abnormal analysis record are specifically as follows: Classify the abnormal weight points, determine the potential abnormal causes according to the size and frequency of their weight difference values, and obtain a preliminary abnormal point classification table; The preliminary abnormal point classification table is analyzed to calculate potential weighing errors or operational errors to obtain abnormal analysis records.

8. The Internet of Things weighing monitoring method based on the steel industry according to claim 1 is characterized in that: The steps for obtaining the weighing adjustment verification result are specifically as follows: Based on the abnormal analysis record, collect the real-time weight after the conveyor belt speed adjustment and the material distribution parameter adjustment, including the material weight value, timestamp and the operating parameters of the corresponding equipment, and obtain a new parameter data record table; The new parameter data record table is analyzed, and the data difference before and after adjustment is compared, and the formula is used: ； Efficiency index for calculating adjustment effect ,in, Represents the adjusted weight data, Represents the weight data before adjustment; Based on the efficiency index, the success degree of the adjustment is judged, and a weighing adjustment verification result is obtained.

9. The IoT weighing monitoring system based on the steel industry is characterized by: According to any one of claims 1 to 8, the Internet of Things weighing monitoring method based on the steel industry is implemented, and the system comprises: The weight data acquisition module continuously collects the weight information of steel materials based on the weight sensor installed on the conveyor belt, records the weight value and corresponding timestamp of each material, and removes noise and outliers in the data to obtain a cleaned weight data set; The weight data analysis module performs a time series analysis on the weight change of the material based on the cleaned weight data set, evaluates the weight change trend of the material at different time points, and predicts the weight of the material in the future time period to obtain the material weight prediction information; The abnormality detection module calculates the difference between the predicted weight of the material and the current monitored weight based on the material weight prediction information, identifies abnormal weight points beyond the normal range, and analyzes the detected abnormal points to determine potential weighing errors or operational errors, and obtains abnormality analysis records; The adjustment execution optimization module adjusts the running speed of the conveyor belt according to the abnormal analysis record, optimizes the material distribution and weight distribution, and obtains weighing adjustment information; The monitoring and verification module monitors the weighing process after adjustment again based on the weighing adjustment information, verifies the execution effect of the adjustment, and obtains a weighing adjustment verification result.

Images