Intelligent anomaly detection and early warning cloud weighing system

Abstract

The invention discloses an intelligent anomaly detection and early warning cloud weighing system, relates to the technical field of Internet of Things and information, and solves the problems that it is difficult to extract new features from real-time data to construct a temperature and humidity comprehensive compensation index, it is difficult to carry out anomaly judgment according to a difference value between a first weight and a second weight, and it is difficult to carry out early warning. The weight compensation multi-element linear regression model construction method comprises the following modules: a data acquisition module used for acquiring weight, temperature and humidity data, and a weight compensation multi-element linear regression model construction module used for constructing a weight compensation multi-element linear regression model according to a first weight, a temperature comprehensive compensation index and a humidity comprehensive compensation index; the cloud server storage module is used for recording and storing various real-time data into a database; the data processing and analysis module is used for constructing temperature comprehensive compensation indexes and humidity comprehensive compensation indexes of different intervals; the abnormity detection process and early warning module is used for judging whether abnormity exists or not, obtaining a weight theoretical value if the abnormity exists, and constructing a weight compensation multiple linear regression model to obtain a weight compensation value if the abnormity exists.

Description

Technical Field

[0001] The present invention belongs to the field of Internet of Things and information technology, and specifically relates to an intelligent abnormality detection and early warning cloud weighing system. Background Art

[0002] With the continuous advancement of science and technology, the research on intelligent weighing systems is constantly developing in all walks of life, especially in the logistics industry. The accuracy of intelligent weighing systems has been improved through sensor technology and model construction. For this reason, an intelligent anomaly detection and early warning cloud weighing system came into being. The collected temperature and humidity data are divided into intervals and classified according to the set segmentation standards, and the temperature comprehensive compensation index and humidity comprehensive compensation index are constructed. The anomaly judgment is made according to the difference between the first weight and the second weight of the measured object. If it is normal, the average of the first and second weights is used as the theoretical weight value. If it is abnormal, a weight compensation multivariate linear regression model is constructed to obtain the weight compensation value, which improves the accuracy and stability of the weighing system.

[0003] Although the existing anomaly detection and early warning cloud weighing system has realized cloud weighing to a certain extent, the existing anomaly detection and early warning cloud weighing system requires professional operation, resulting in large data errors and thus inaccurate results. In addition, the sensor will be affected by the environment, thus affecting the accuracy of the weighing system. The existing cloud weighing system is difficult to extract new features from real-time data to construct a comprehensive compensation index for temperature and humidity, and it is difficult to make anomaly judgments based on the difference between the first weight and the second weight. There is a lack of a weight compensation multivariate linear regression model based on the first weight, the comprehensive temperature compensation index and the comprehensive humidity compensation index. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent abnormality detection and early warning cloud weighing system for solving the following technical problems: Although the existing anomaly detection and warning cloud weighing system has realized cloud weighing to a certain extent, it is difficult for the existing cloud weighing system to extract new features from real-time data to construct a comprehensive compensation index for temperature and humidity, and it is difficult to make anomaly judgments based on the difference between the first weight and the second weight. There is a lack of a weight compensation multivariate linear regression model based on the first weight, the temperature comprehensive compensation index and the humidity comprehensive compensation index.

[0005] To solve the above problems, the first aspect of the present invention provides an intelligent abnormality detection and early warning cloud weighing system, comprising the following modules: Data acquisition module: two weight measuring sensors and one temperature and humidity measuring sensor are installed to collect the first weight, second weight, temperature and humidity data in real time through the sensors, and transmit the collected data to the cloud server; Cloud server storage module: receives the real-time first weight, second weight, temperature and humidity data from the data acquisition module, and records and stores various real-time data in the database; Data processing and analysis module: record the temperature and humidity data of each weighing, collect historical data sets containing temperature, humidity and weighing results, pre-process the collected temperature, humidity and weight data, assign intervals to the temperature and humidity data and classify them according to the set segmentation standards, analyze the real-time data to extract new features, and construct the temperature comprehensive compensation index and humidity comprehensive compensation index based on the extracted new features; Abnormal detection process and early warning module: judge whether there is an abnormality based on the difference between the first weight and the second weight of the measured object. If there is no abnormality, obtain the theoretical weight value based on the first weight and the second weight. If there is an abnormality, construct a weight compensation multivariate linear regression model based on the first weight, temperature comprehensive compensation index and humidity comprehensive compensation index.

[0006] As a further solution of the present invention, the two weight measuring sensors and one temperature and humidity measuring sensor are installed, the first weight, the second weight, the temperature and the humidity data are collected in real time by the sensors, and the collected data is transmitted to the cloud server, including the following steps: Two weight sensors are installed at both ends of the conveyor belt in the logistics center. The first weight sensor is installed at the entrance of the conveyor belt to measure the first weight, and the second weight sensor is installed at the exit of the conveyor belt to measure the second weight. The temperature sensor and humidity sensor are installed in the center of the ceiling to measure the temperature and humidity respectively. The sensors collect data at preset time intervals and transmit the collected data to the cloud server.

[0007] As a further solution of the present invention, the receiving of the real-time first weight, second weight, temperature and humidity data from the data acquisition module, and recording and storing the multiple real-time data in a database comprises the following steps: A service for receiving data is deployed on the cloud server, which processes the connection request from the data acquisition module, receives the collected real-time first weight, second weight, temperature and humidity data, creates a database for storing data on the cloud server, and parses the received various real-time data through the server program, and records and stores them in the created database.

[0008] As a further solution of the present invention, the recording of temperature and humidity data at each weighing, collecting a historical data set containing temperature, humidity and weighing results, and preprocessing the collected temperature, humidity and weight data include the following steps: The temperature, humidity and weight data obtained from each weighing are used to construct a historical data set. Missing values, abnormal values ​​and duplicate data in real-time weight, temperature and humidity data are checked and processed. Filtering technology is applied to remove high-frequency noise in the data and then normalized. Normalized operation calculation formula: in, Any of the normalized real-time weight, temperature and humidity data, Any of the real-time weight, temperature and humidity data actually measured, Any one of the minimum values ​​of real-time weight, temperature and humidity data, Any one of the maximum values ​​of real-time weight, temperature and humidity data.

[0009] As a further solution of the present invention, the temperature and humidity data are allocated intervals and classified according to the set segmentation standard, the real-time data is analyzed to extract new features, and the temperature comprehensive compensation index and the humidity comprehensive compensation index are constructed according to the extracted new features, including the following steps: The temperature data is considered as a temperature interval when it increases by 5°C, and the humidity data is considered as a humidity interval when it increases by 10%RH. The recorded data are sorted from small to large, and the temperature and humidity intervals are classified according to the set segmentation standards. The data in each temperature and humidity interval are analyzed respectively to construct the temperature comprehensive compensation index and humidity comprehensive compensation index.

[0010] As a further solution of the present invention, the construction of the temperature comprehensive compensation index and the humidity comprehensive compensation index comprises the following steps: The temperature change rate is obtained by the difference and time interval of adjacent data points, and the temperature average value, standard deviation, maximum temperature value and minimum temperature value in the time window are determined. Based on the acquired data, a new temperature feature is constructed to obtain a comprehensive temperature compensation index; Temperature comprehensive compensation index calculation formula: in, is the comprehensive temperature compensation index, is the total number of detections, For the The temperature value obtained by the test is For the The temperature value obtained by the test is is the time interval, is the maximum temperature in the test results, is the minimum temperature value in the test results, is the average temperature in the test results, is the temperature standard deviation in the test results; The humidity change rate is obtained by the difference and time interval of adjacent data points, and the maximum humidity value, minimum humidity value and average humidity value in the time window are determined. The humidity fluctuation rate is obtained according to the ratio of the difference between the maximum humidity value and the average humidity value and the difference between the minimum humidity value and the average humidity value. The humidity change rate and humidity fluctuation rate are combined to construct a new humidity feature and obtain a comprehensive humidity compensation index. Humidity comprehensive compensation index calculation formula: in, is the humidity comprehensive compensation index, is the total number of detections, For the The humidity value obtained by the test is For the The humidity value obtained by the test is is the time interval, is the maximum humidity value in the test results, is the minimum humidity value in the test results. is the average humidity value in the test results; All temperature and humidity data are obtained, and the interval corresponding to the data is determined according to the segmentation standard, and the data is substituted to obtain the temperature comprehensive compensation index and humidity comprehensive compensation index in different temperature and humidity intervals.

[0011] As a further solution of the present invention, judging whether there is an abnormality according to the difference between the first weight and the second weight of the measured object, if there is no abnormality, obtaining the weight theoretical value according to the first weight and the second weight, and if there is an abnormality, constructing a weight compensation multivariate linear regression model according to the first weight, the temperature comprehensive compensation index and the humidity comprehensive compensation index, including the following steps: If 0≤ <5, it means that there is no abnormality in the weight of the measured object, no warning signal is issued, and the theoretical weight value is obtained based on the average value of the first weight and the second weight; in, is the first weight of the object being measured, is the second weight of the object being measured; Theoretical weight calculation formula: in, is the theoretical value of the weight of the object being measured, is the first weight of the object being measured, is the second weight of the object being measured; like If ≥5, it means that the weight of the measured object is abnormal, and an early warning signal is issued. The weight compensation multivariate linear regression model is fitted according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index to compensate for the weight of the measured object.

[0012] As a further solution of the present invention, the weight compensation multivariate linear regression model is obtained by fitting the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index to compensate the weight of the measured object, comprising the following steps: For each interval, a multivariate linear regression model is constructed, and the corresponding interval is determined according to the temperature and humidity data to obtain the corresponding temperature comprehensive compensation index and humidity comprehensive compensation index. The temperature comprehensive compensation index and humidity comprehensive compensation index are used as independent variables, and the weight compensation multivariate linear regression model is obtained by fitting according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index. The model is trained using the collected data, and the new object's first weight, temperature comprehensive compensation index and humidity comprehensive compensation index are input into the trained model as input data, and the model will output the corresponding weight compensation value; in, is the weight of the object after compensation, is the first weight of the object being measured, is the comprehensive temperature compensation index, is the humidity comprehensive compensation index, , and are the coefficients obtained from regression analysis.

[0013] Beneficial effects of the present invention: The present invention combines multiple sensors to monitor the environment of the logistics center and the weight of the measured object in real time, thereby improving the comprehensiveness and accuracy of the measurement results; The present invention divides the collected temperature and humidity data into intervals and classifies them according to the set segmentation standard, so that data processing is more refined, which helps to more accurately analyze and understand the changes in environmental conditions, construct a temperature comprehensive compensation index and a humidity comprehensive compensation index, realize data correction, and improve the intelligence level of the system; The present invention determines the abnormal situation of the system by using the difference between the first weight and the second weight of the measured object as a judgment condition. If the system is normal, the average value of the first weight and the second weight is used as the theoretical weight value. If the system is abnormal, a weight compensation multivariate linear regression model is constructed according to the first weight, the temperature comprehensive compensation index and the humidity comprehensive compensation index to obtain the weight compensation value. By introducing the temperature comprehensive compensation index and the humidity comprehensive compensation index, the weighing data is corrected, thereby reducing the influence of environmental factors on the measurement results and improving the accuracy of the measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a module flow chart of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1 As shown, the present invention is an intelligent abnormality detection and early warning cloud weighing system, comprising the following modules: Data acquisition module: two weight measuring sensors and one temperature and humidity measuring sensor are installed to collect the first weight, second weight, temperature and humidity data in real time through the sensors, and transmit the collected data to the cloud server; Cloud server storage module: receives the real-time first weight, second weight, temperature and humidity data from the data acquisition module, and records and stores various real-time data in the database; Data processing and analysis module: record the temperature and humidity data of each weighing, collect historical data sets containing temperature, humidity and weighing results, pre-process the collected temperature, humidity and weight data, assign intervals to the temperature and humidity data and classify them according to the set segmentation standards, analyze the real-time data to extract new features, and construct the temperature comprehensive compensation index and humidity comprehensive compensation index based on the extracted new features; Abnormal detection process and early warning module: judge whether there is an abnormality based on the difference between the first weight and the second weight of the measured object. If there is no abnormality, obtain the theoretical weight value based on the first weight and the second weight. If there is an abnormality, construct a weight compensation multivariate linear regression model based on the first weight, temperature comprehensive compensation index and humidity comprehensive compensation index.

[0018] Specifically, the installed multiple sensors collect data every 10 minutes, use HTTP POST data transmission method for data transmission, and send the data to the cloud server by configuring the Wi-Fi network connection method; use Python to write a Flack application as a server program, deploy the written server program on the cloud server, parse the formatted data through the program, and insert it into the database; after cleaning and filtering the data, use Min-Max normalization to scale the data to a specific range, and save the data after the data is cleaned; based on the acquired data, new features of temperature and humidity are respectively constructed to obtain the temperature comprehensive compensation index and the humidity comprehensive compensation index; judge whether there is an abnormality according to the weight difference of the measured object. If it is normal, the average of the first weight and the second weight of the measured object is used as the theoretical weight value; if it is abnormal, a weight compensation multivariate linear regression model is constructed according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index.

[0019] In one embodiment of the present invention, the two weight measuring sensors and one temperature and humidity measuring sensor are installed, the first weight, the second weight, the temperature and the humidity data are collected in real time by the sensors, and the collected data is transmitted to the cloud server, including the following steps: Two weight sensors are installed at both ends of the conveyor belt in the logistics center. The first weight sensor is installed at the entrance of the conveyor belt to measure the first weight, and the second weight sensor is installed at the exit of the conveyor belt to measure the second weight. The temperature sensor and humidity sensor are installed in the center of the ceiling to measure the temperature and humidity respectively. The sensors collect data at preset time intervals and transmit the collected data to the cloud server.

[0020] Specifically, weight, temperature and humidity data are collected through various sensors, and the sensors are made to collect data every 10 minutes. The HTTP POST data transmission method is used for data transmission, and the data is sent to the cloud server by configuring the Wi-Fi network connection method.

[0021] In one embodiment of the present invention, the receiving of the real-time first weight, second weight, temperature and humidity data from the data acquisition module, and recording and storing the multiple real-time data in a database comprises the following steps: A service for receiving data is deployed on the cloud server, which processes the connection request from the data acquisition module, receives the collected real-time first weight, second weight, temperature and humidity data, creates a database for storing data on the cloud server, and parses the received various real-time data through the server program, and records and stores them in the created database.

[0022] Specifically, a Flack application is written in Python as a server program to receive connection requests and real-time data sent by the data acquisition module. The written server program is deployed on the cloud server. The formatted data is parseable through the program and inserted into the database. The database system is installed and configured on the cloud server, and the database table structure is designed to store the collected first weight, second weight, temperature and humidity data.

[0023] In one embodiment of the present invention, recording the temperature and humidity data at each weighing, collecting a historical data set including temperature, humidity and weighing results, and preprocessing the collected temperature, humidity and weight data include the following steps: The temperature, humidity and weight data obtained from each weighing are used to construct a historical data set. Missing values, abnormal values ​​and duplicate data in real-time weight, temperature and humidity data are checked and processed. Filtering technology is applied to remove high-frequency noise in the data and then normalized. Normalized operation calculation formula: in, Any of the normalized real-time weight, temperature and humidity data, Any of the real-time weight, temperature and humidity data actually measured, Any one of the minimum values ​​of real-time weight, temperature and humidity data, Any one of the maximum values ​​of real-time weight, temperature and humidity data.

[0024] Specifically, the data processing library of the programming language is used to read the data, check whether there are missing values, delete the rows or columns containing missing values, identify and delete outliers through statistical analysis methods, use the deduplication function to remove duplicate data, use a low-pass filter to remove high-frequency noise, and use Min-Max normalization to scale the data to a specific range, and save the data after the data is processed.

[0025] In one embodiment of the present invention, the temperature and humidity data are allocated intervals and classified according to the set segmentation standard, the real-time data is analyzed to extract new features, and the temperature comprehensive compensation index and the humidity comprehensive compensation index are constructed according to the extracted new features, including the following steps: The temperature data is considered as a temperature interval when it increases by 5°C, and the humidity data is considered as a humidity interval when it increases by 10%RH. The recorded data are sorted from small to large, and the temperature and humidity intervals are classified according to the set segmentation standards. The data in each temperature and humidity interval are analyzed respectively to construct the temperature comprehensive compensation index and humidity comprehensive compensation index.

[0026] Specifically, the temperature and humidity data are sorted from small to large and segmentation standards are set. Every increase of 5°C is regarded as a temperature interval, and every increase of 10%RH is regarded as a humidity interval. The temperature and humidity data are classified according to the set segmentation standards, and the data in each temperature and humidity interval are analyzed to calculate the temperature comprehensive compensation index and humidity comprehensive compensation index.

[0027] In one embodiment of the present invention, the step of constructing the temperature comprehensive compensation index and the humidity comprehensive compensation index comprises the following steps: The temperature change rate is obtained by the difference and time interval of adjacent data points, and the temperature average value, standard deviation, maximum temperature value and minimum temperature value in the time window are determined. Based on the acquired data, a new temperature feature is constructed to obtain a comprehensive temperature compensation index; Temperature comprehensive compensation index calculation formula: in, is the comprehensive temperature compensation index, is the total number of detections, For the The temperature value obtained by the test is For the The temperature value obtained by the test is is the time interval, is the maximum temperature in the test results, is the minimum temperature value in the test results, is the average temperature in the test results, is the temperature standard deviation in the test results; The humidity change rate is obtained by the difference and time interval of adjacent data points, and the maximum humidity value, minimum humidity value and average humidity value in the time window are determined. The humidity fluctuation rate is obtained according to the ratio of the difference between the maximum humidity value and the average humidity value and the difference between the minimum humidity value and the average humidity value. The humidity change rate and humidity fluctuation rate are combined to construct a new humidity feature and obtain a comprehensive humidity compensation index. Humidity comprehensive compensation index calculation formula: in, is the humidity comprehensive compensation index, is the total number of detections, For the The humidity value obtained by the test is For the The humidity value obtained by the test is is the time interval, is the maximum humidity value in the test results, is the minimum humidity value in the test results. is the average humidity value in the test results; All temperature and humidity data are obtained, and the interval corresponding to the data is determined according to the segmentation standard, and the data is substituted to obtain the temperature comprehensive compensation index and humidity comprehensive compensation index in different temperature and humidity intervals.

[0028] Specifically, the temperature change rate is calculated according to the temperature difference and time interval of adjacent data points, a time window of 10 minutes is selected, and the statistical characteristics of the temperature are calculated, including calculating the temperature average, standard deviation, maximum and minimum values ​​in each time window, and the statistical data are merged back into the original data set. New temperature characteristics are constructed based on the acquired data to obtain a comprehensive temperature compensation index; the humidity change rate is calculated according to the humidity difference and time interval of adjacent data points, a time window of 10 minutes is selected, and the statistical characteristics of the humidity are calculated, including calculating the humidity average, maximum and minimum values ​​in each time window, and the humidity fluctuation rate is obtained according to the difference ratio between the maximum value and the minimum value and the average value, and the statistical characteristics are merged back into the original data set. The humidity change rate and the humidity fluctuation rate are jointly constructed to obtain a new humidity characteristic to obtain a comprehensive humidity compensation index.

[0029] In one embodiment of the present invention, judging whether there is an abnormality according to the difference between the first weight and the second weight of the measured object, if there is no abnormality, obtaining the weight theoretical value according to the first weight and the second weight, and if there is an abnormality, constructing a weight compensation multivariate linear regression model according to the first weight, the temperature comprehensive compensation index and the humidity comprehensive compensation index, comprises the following steps: If 0≤ <5, it means that there is no abnormality in the weight of the measured object, no warning signal is issued, and the theoretical weight value is obtained based on the average value of the first weight and the second weight; in, is the first weight of the object being measured, is the second weight of the object being measured; Theoretical weight calculation formula: in, is the theoretical value of the weight of the object being measured, is the first weight of the object being measured, is the second weight of the object being measured; like If ≥5, it means that the weight of the measured object is abnormal, and an early warning signal is issued. The weight compensation multivariate linear regression model is fitted according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index to compensate for the weight of the measured object.

[0030] Specifically, whether there is an abnormality is determined based on the weight difference of the measured object. If the weight difference is within the preset range, it is considered that there is no abnormality. The average of the first weight and the second weight of the measured object is used as the theoretical weight value. If it exceeds the preset range, it is considered that there is an abnormality. A weight compensation multivariate linear regression model is constructed based on the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index.

[0031] In one embodiment of the present invention, the weight compensation multivariate linear regression model is obtained by fitting the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index to compensate the weight of the measured object, comprising the following steps: For each interval, a multivariate linear regression model is constructed, and the corresponding interval is determined according to the temperature and humidity data to obtain the corresponding temperature comprehensive compensation index and humidity comprehensive compensation index. The temperature comprehensive compensation index and humidity comprehensive compensation index are used as independent variables, and the weight compensation multivariate linear regression model is obtained by fitting according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index. The model is trained using the collected data, and the new object's first weight, temperature comprehensive compensation index and humidity comprehensive compensation index are input into the trained model as input data, and the model will output the corresponding weight compensation value; in, is the weight of the object after compensation, is the first weight of the object being measured, is the comprehensive temperature compensation index, is the humidity comprehensive compensation index, , and are the coefficients obtained from regression analysis.

[0032] Specifically, the first weight of the measured object is measured, and the corresponding range interval is found according to the obtained temperature and humidity data, and then substituted into the corresponding temperature comprehensive compensation index and humidity comprehensive compensation index. The temperature and humidity comprehensive compensation index are used as independent variables, and the weight compensation multivariate linear regression model is obtained by fitting according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index, and the weight compensation value of the measured object is obtained by prediction; wherein, , and is the coefficient obtained from regression analysis, is 0.2, is 0.5, is 0.3.

[0033] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent abnormality detection and early warning cloud weighing system, characterized in that: Includes the following modules: Data acquisition module: two weight measuring sensors and one temperature and humidity measuring sensor are installed to collect the first weight, second weight, temperature and humidity data in real time through the sensors, and transmit the collected data to the cloud server; Cloud server module: receives the real-time first weight, second weight, temperature and humidity data from the data acquisition module, and records and stores various real-time data in the database; Data processing and analysis module: record the temperature and humidity data of each weighing, collect historical data sets containing temperature, humidity and weighing results, pre-process the collected temperature, humidity and weight data, assign intervals to the temperature and humidity data and classify them according to the set segmentation standards, analyze the real-time data to extract new features, and construct the temperature comprehensive compensation index and humidity comprehensive compensation index based on the extracted new features; Abnormal detection process and early warning module: judge whether there is an abnormality based on the difference between the first weight and the second weight of the measured object. If there is no abnormality, obtain the theoretical weight value based on the first weight and the second weight. If there is an abnormality, construct a weight compensation multivariate linear regression model based on the first weight, temperature comprehensive compensation index and humidity comprehensive compensation index.

2. The intelligent abnormality detection and early warning cloud weighing system according to claim 1 is characterized in that: The method comprises installing two weight measuring sensors and one temperature and humidity measuring sensor, collecting first weight, second weight, temperature and humidity data in real time through the sensors, and transmitting the collected data to a cloud server, including the following steps: Two weight sensors are installed at both ends of the conveyor belt in the logistics center. The first weight sensor is installed at the entrance of the conveyor belt to measure the first weight, and the second weight sensor is installed at the exit of the conveyor belt to measure the second weight. The temperature sensor and humidity sensor are installed in the center of the ceiling to measure the temperature and humidity respectively. The sensors collect data at preset time intervals and transmit the collected data to the cloud server.

3. The intelligent abnormality detection and early warning cloud weighing system according to claim 1 is characterized in that: The method comprises the following steps: receiving the real-time first weight, second weight, temperature and humidity data from the data acquisition module, and recording and storing the multiple real-time data in a database: A service for receiving data is deployed on the cloud server, which processes the connection request from the data acquisition module, receives the collected real-time first weight, second weight, temperature and humidity data, creates a database for storing data on the cloud server, and parses the received various real-time data through the server program, and records and stores them in the created database.

4. The intelligent abnormality detection and early warning cloud weighing system according to claim 1 is characterized in that: The method of recording the temperature and humidity data at each weighing, collecting a historical data set including temperature, humidity and weighing results, and performing a preprocessing operation on the collected temperature, humidity and weight data comprises the following steps: The temperature, humidity and weight data obtained from each weighing are used to construct a historical data set. Missing values, abnormal values ​​and duplicate data in real-time weight, temperature and humidity data are checked and processed. Filtering technology is applied to remove high-frequency noise in the data and then normalized. Normalized operation calculation formula: Where X is any one of the normalized real-time weight, temperature and humidity data, X is any one of the real-time weight, temperature and humidity data actually measured, Any one of the minimum values ​​of real-time weight, temperature and humidity data, It is any one of the maximum values ​​of real-time weight, temperature and humidity data.

5. The intelligent abnormality detection and early warning cloud weighing system according to claim 1 is characterized in that: The temperature and humidity data are assigned intervals and classified according to the set segmentation standard, the real-time data is analyzed to extract new features, and the temperature comprehensive compensation index and the humidity comprehensive compensation index are constructed according to the extracted new features, including the following steps: The temperature data is considered as a temperature interval when it increases by 5°C, and the humidity data is considered as a humidity interval when it increases by 10%RH. The recorded data are sorted from small to large, and the temperature and humidity intervals are classified according to the set segmentation standards. The data in each temperature and humidity interval are analyzed respectively to construct the temperature comprehensive compensation index and humidity comprehensive compensation index.

6. The intelligent abnormality detection and early warning cloud weighing system according to claim 5 is characterized in that: The step of constructing the temperature comprehensive compensation index and the humidity comprehensive compensation index comprises the following steps: The temperature change rate is obtained by the difference and time interval of adjacent data points, and the temperature average value, standard deviation, maximum temperature value and minimum temperature value in the time window are determined. Based on the acquired data, a new temperature feature is constructed to obtain a comprehensive temperature compensation index; Temperature comprehensive compensation index calculation formula: in, is the comprehensive temperature compensation index, N is the total number of detections, is the temperature value obtained from the i-th detection, For the The temperature value obtained by the test is is the time interval, is the maximum temperature in the test results, is the minimum temperature value in the test results, is the average temperature in the test results, is the temperature standard deviation in the test results; The humidity change rate is obtained by the difference and time interval of adjacent data points, and the maximum humidity value, minimum humidity value and average humidity value in the time window are determined. The humidity fluctuation rate is obtained according to the ratio of the difference between the maximum humidity value and the average humidity value and the difference between the minimum humidity value and the average humidity value. The humidity change rate and humidity fluctuation rate are combined to construct a new humidity feature and obtain a comprehensive humidity compensation index. Humidity comprehensive compensation index calculation formula: in, is the humidity comprehensive compensation index, N is the total number of tests, is the humidity value obtained from the jth detection, For the The humidity value obtained by the test is is the time interval, is the maximum humidity value in the test results, is the minimum humidity value in the test results. is the average humidity value in the test results; All temperature and humidity data are obtained, and the interval corresponding to the data is determined according to the segmentation standard, and the data is substituted to obtain the temperature comprehensive compensation index and humidity comprehensive compensation index in different temperature and humidity intervals.

7. The intelligent abnormality detection and early warning cloud weighing system according to claim 1 is characterized in that: The method comprises the following steps: judging whether there is an abnormality according to the difference between the first weight and the second weight of the measured object, obtaining a weight theoretical value according to the first weight and the second weight if there is no abnormality, and constructing a weight compensation multivariate linear regression model according to the first weight, the temperature comprehensive compensation index and the humidity comprehensive compensation index if there is an abnormality: If 0≤ <5, it means that there is no abnormality in the weight of the measured object, no warning signal is issued, and the theoretical weight value is obtained based on the average value of the first weight and the second weight; in, is the first weight of the object being measured, is the second weight of the object being measured; Theoretical weight calculation formula: in, is the theoretical value of the weight of the object being measured, is the first weight of the object being measured, is the second weight of the object being measured; like If ≥5, it means that the weight of the measured object is abnormal, and an early warning signal is issued. The weight compensation multivariate linear regression model is fitted according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index to compensate for the weight of the measured object.

8. The intelligent abnormality detection and early warning cloud weighing system according to claim 7 is characterized in that: The weight compensation multivariate linear regression model is obtained by fitting the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index to compensate the weight of the measured object, including the following steps: For each interval, a multivariate linear regression model is constructed, and the corresponding interval is determined according to the temperature and humidity data to obtain the corresponding temperature comprehensive compensation index and humidity comprehensive compensation index. The temperature comprehensive compensation index and humidity comprehensive compensation index are used as independent variables, and the weight compensation multivariate linear regression model is obtained by fitting according to the first weight of the measured object, the temperature comprehensive compensation index and the humidity comprehensive compensation index. The model is trained using the collected data, and the new object's first weight, temperature comprehensive compensation index and humidity comprehensive compensation index are input into the trained model as input data, and the model will output the corresponding weight compensation value; in, is the weight of the object after compensation, is the first weight of the object being measured, is the comprehensive temperature compensation index, is the humidity comprehensive compensation index, , and are the coefficients obtained from regression analysis.

Images