Unattended weighing management system and method
By introducing AI-assisted anomaly detection, blockchain application and adaptive optimization algorithms into the over-pound management system, the problem that existing systems cannot realize AI automated anomaly detection, data secure storage and operation efficiency optimization is solved, and efficient and safe Overweight data management and transaction transparency.
Patent Information
- Application Number
- CN202410560996.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-08
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-05-08
AI Technical Summary
The existing overweight management system cannot realize AI automation abnormality detection, encrypted and secure storage, operation efficiency optimization, and data analysis and processing to generate reports, resulting in manual data tampering, data loss and opacity in transaction data.
An unattended over-weight management system was designed, including AI-assisted exception detection system, blockchain application system and adaptive optimization algorithm system. Anomaly detection is realized through AI-assisted exception detection system, and blockchain application system realizes data encryption storage. and adaptive optimization algorithm systems optimize operating efficiency, and generate system-generated reports through data analysis reports.
It realizes AI automated abnormal detection, ensures data security and transaction transparency, improves the automation level and operation efficiency of the system, reduces manual errors and data loss, and enhances decision-making accuracy.
Smart Images

Figure CN118506341B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of weighing management systems, and in particular to an unattended weighing management system and method. Background Art
[0002] In traditional weighing management systems, personnel are usually required to perform weighing operations on site and manually record vehicle information, cargo information, weight and other data. This method has problems such as high labor costs, low efficiency, and prone to errors. With the development of technologies such as the Internet of Things, artificial intelligence and blockchain, unattended weighing management systems have emerged. This system uses modern technology to achieve automated and intelligent management of the weighing process, thereby improving the accuracy, efficiency and safety of weighing operations, reducing labor costs, and bringing new changes and opportunities to the production and operation of enterprises.
[0003] Therefore, the present application proposes an unattended weighing management system and method to enable the weighing management system to be truly unattended, providing a more intelligent, efficient and safe solution for the production and logistics management of the enterprise.
[0004] 1. Patent document CN115587772B discloses a self-service weighing control method based on warehouse entry and exit management. The above patent realizes the automation and information management of warehouse entry and exit, but the above patent cannot realize the AI automatic anomaly detection function of weighing management.
[0005] 2. Patent document CN113515985B discloses a self-service weighing system, weighing detection method, equipment and storage medium. The above patent realizes the detection of abnormal cheating behavior of weighed vehicles, thereby intelligently managing the vehicle weighing process and improving the reliability of the autonomous weighing method. However, the above patent cannot realize the encrypted and secure storage function of the weighing data.
[0006] 3. Patent document CN114296472B discloses a weighing process test system based on drones. The above patent achieves the technical effects of portability, versatility, safety, reduced workload, simulation of unexpected conditions and high degree of automation, but the above patent cannot achieve the function of optimizing the operating efficiency of the weighing system.
[0007] 4. Patent document CN102235906B discloses a method for automatic weighing of vehicles while they are moving. The above patent can effectively alleviate the logistics pressure of enterprise weighing stations, improve weighing efficiency, and improve the accuracy of weighing at weighing stations. It can also prevent various cheating phenomena such as embezzlement, license plate replacement, incomplete weighing, weighing multiple vehicles at the same time, and substitution of one vehicle for another. However, the above patent cannot realize the function of analyzing and processing weighing data to generate reports.
[0008] In summary, the above patents cannot realize the AI automated anomaly detection function for weighing management, the encrypted and secure storage function for weighing data, the optimization function for the operating efficiency of the weighing system, and the analysis and processing of weighing data to generate reports, resulting in problems such as manual tampering of data, data loss, inability to autonomously detect anomalies in real time, and opaque transaction data;
[0009] To this end, the present application proposes an unattended weighing management system and method that can realize AI automated anomaly detection function for weighing management, encrypted and secure storage function for weighing data, optimization function for operating efficiency of the weighing system, and analysis and processing of weighing data to generate reports. Summary of the invention
[0010] The purpose of the present invention is to provide an unattended weighing management system and method to solve the technical problems proposed in the above background technology that the AI automated anomaly detection function for weighing management, the encrypted and secure storage function for weighing data, the operation efficiency optimization function for the weighing system and the analysis and processing of weighing data to generate reports cannot be realized, resulting in manual data tampering, data loss, inability to autonomously detect anomalies in real time and opaque transaction data.
[0011] To achieve the above object, the present invention provides the following technical solutions: an unattended weighing management system, comprising a management control center and a cloud server, wherein the management control center is connected to the cloud server via the Internet, and an AI-assisted anomaly detection system is designed in the management control center, wherein the AI-assisted anomaly detection system is used to analyze and detect abnormal situations occurring during the weighing process;
[0012] The AI-assisted anomaly detection system comprises: a weighing data analysis unit, a real-time anomaly detection unit and an anomaly processing notification unit, wherein the weighing data analysis unit is connected to the real-time anomaly detection unit via a data interface, and the real-time anomaly detection unit is connected to the anomaly processing notification unit via a signal API interface;
[0013] The weighing data analysis unit integrates a data analysis server and a machine learning algorithm module, and uses the machine learning algorithm module to analyze and learn the historical weighing data in the data analysis server to obtain normal weighing patterns and rules;
[0014] The real-time anomaly detection unit integrates a real-time data processing module and an anomaly detection algorithm module, wherein the real-time data processing module is used to process real-time weighing data, and the anomaly detection algorithm module is used to detect weighing anomalies in real time and issue an alarm signal in time;
[0015] The abnormal handling notification unit integrates an alarm system and a notification device. The alarm system receives an alarm signal transmitted by the abnormal detection algorithm module to trigger the alarm system response. The system automatically takes processing measures and notifies relevant personnel through the notification device to handle the abnormal situation.
[0016] Preferably, the AI-assisted anomaly detection system is connected to an intelligent recognition system via a data cable, and the intelligent recognition system is used to intelligently identify the weighed vehicle and cargo data information;
[0017] The intelligent identification system includes: a vehicle information identification unit, a cargo information identification unit and a data processing and storage unit, and the vehicle information identification unit and the cargo information identification unit are connected to the data processing and storage unit via a data line;
[0018] The vehicle information recognition unit integrates a high-definition camera installed at the entrance, a license plate recognition camera and an RGB sensor. The license plate number recognition adopts a CNN model based on deep learning. The high-definition camera and RGB sensor use image feature extraction and pattern matching algorithms to identify the vehicle model and color.
[0019] The cargo information identification unit integrates laser radar and weighing sensor. The laser radar is used to obtain cargo images, and the weighing sensor is used to measure the weight of the weighing vehicle in real time. The cargo type identification adopts a deep learning model based on image recognition, and the weighing data processing uses a sliding window algorithm to identify the cargo type, quantity and weight in real time and generate complete weighing data.
[0020] The data processing storage unit integrates a database server, which is used to store weighing data, associate the identified vehicle and cargo information with the weighing information, and store it in the data analysis server.
[0021] Preferably, a blockchain application system is designed in the cloud server, and the blockchain application system is used to encrypt and store the vehicle information and weighing data in the database;
[0022] The blockchain application system includes: data encryption storage unit, smart contract execution unit and data traceability audit unit;
[0023] The data encryption storage unit integrates an encryption module and a blockchain node interface. The encryption module is used to encrypt the data. The blockchain node is used to store the encrypted data and connect to the database through an API interface. The weighing data is encrypted and stored in the blockchain in the cloud server.
[0024] The smart contract execution unit integrates smart contract programming tools and smart contract execution nodes. Smart contract programming tools are used to write smart contracts, and smart contract execution nodes are used to execute contracts. Smart contracts are written to define the rights and interests of both parties and the contract terms. The smart contract execution on the blockchain ensures fair and transparent transactions.
[0025] The data traceability and audit unit integrates a blockchain browser and an audit tool. The blockchain browser is used to view data on the blockchain, and the audit tool is used to audit the data. Blockchain technology is used to provide traceability and audit functions for weighing data.
[0026] Preferably, the AI-assisted anomaly detection system is designed with an adaptive optimization algorithm system, which is used to analyze and learn the weighing process and optimize the process;
[0027] The adaptive optimization algorithm system includes: historical data analysis and learning unit, optimization algorithm implementation adjustment unit and effect evaluation feedback unit;
[0028] The historical data analysis and learning unit integrates a historical data processing module and a machine learning model training module. The historical data processing module is used to clean and preprocess historical data, and the machine learning model training module is used to establish an optimization model. The historical weighing data is analyzed and learned through the machine learning algorithm to obtain the rules and characteristics of the weighing process and output the optimization algorithm.
[0029] The optimization algorithm implementation and adjustment unit integrates the optimization algorithm implementation module and the parameter adjustment module. The optimization algorithm implementation module is used to execute the optimization algorithm, and the parameter adjustment module is used to dynamically adjust the optimization parameters. According to the historical data learning model, the weighing process is adaptively optimized and the best weighing scheme is selected.
[0030] The effect evaluation feedback unit integrates an effect evaluation module and a feedback mechanism module. The effect evaluation module is used to evaluate the optimization results, and the feedback mechanism module is used to feed back the evaluation results to the optimization algorithm, monitor the optimization results, evaluate the effect of the weighing process, and feed back the evaluation results to the optimization algorithm for adjustment.
[0031] Preferably, the management control center is also designed with a data analysis report generation system, which is connected to the cloud server and the adaptive optimization algorithm system through a data Internet interface, and is used to generate data analysis reports in real time;
[0032] The data analysis report generation system includes: a real-time data analysis unit and a report generation and display unit, and the real-time data analysis unit is connected to the report generation and display unit through a network communication interface;
[0033] The real-time data analysis unit integrates a data analysis module and a real-time database. The data analysis module is used to analyze the real-time weighing data and the analysis results are stored in the real-time database.
[0034] The report generation and display unit integrates a report generation module and a data visualization tool, automatically generates a report based on the data analysis results, and displays it on the display screen in the management control center through the data visualization tool.
[0035] Preferably, the algorithm in the anomaly detection algorithm module in the real-time anomaly detection unit is a Z-score algorithm, and the specific formula is as follows:
[0036] Assume that the weighing data set is x1, x2, ..., x n , the mean is x0, the standard deviation is s, then the data point x i The calculation formula is Among them, Zi represents the data point x i The Z-score of the data point is marked as abnormal if the threshold is set to 3 and -3.
[0037] Preferably, the sliding window algorithm formula used in the weighing data processing is as follows:
[0038] Moving average calculation formula:
[0039] Among them, the moving average MA t , t represents the current time point, k represents the window size, x i Represents the weighing data at the i-th time point.
[0040] Preferably, the machine learning model trained by the machine learning model training module in the historical data analysis learning unit is a random forest model, and the formula of the random forest model is as follows:
[0041] For classification problems:
[0042] in, is the predicted category of sample x, f i (x) is the prediction result of the i-th decision tree, and MajorityVote represents the vote to decide the final category;
[0043] For regression problems:
[0044] in, is the predicted value of sample x, f i (x) is the prediction result of the i-th decision tree, and n is the number of decision trees.
[0045] Preferably, the management method comprises the following steps:
[0046] S1. Use the vehicle information recognition module and cargo information recognition module in the intelligent recognition system to monitor the operating status and data collection of the weighing system in real time;
[0047] S2. With the help of AI-assisted anomaly detection system, abnormal situations in the weighing process are monitored, alarms are issued in time, and corresponding measures are taken;
[0048] S3. Use the adaptive optimization algorithm system to adjust the parameters and operation strategies of the weighing system in real time according to the historical data analysis results to optimize the system efficiency and accuracy;
[0049] S4. Analyze the weighing data and generate a weighing report using the data analysis report generation system, and display the report on the display screen of the management control center;
[0050] S5. Regularly perform equipment inspection, sensor calibration, and data cleaning system maintenance and care to ensure long-term and stable operation of the system.
[0051] Preferably, the management method further comprises the following steps:
[0052] S11, the data processing storage unit associates the vehicle and cargo information identified by the vehicle information identification module and the cargo information identification module with the weighing information, and stores them in the data analysis server;
[0053] S31. Use the machine learning model training module in the historical data analysis learning unit to study and analyze historical weighing data, generate reports and provide decision support.
[0054] Compared with the prior art, the present invention has the following beneficial effects:
[0055] 1. The present invention realizes the AI automated abnormality detection function for weighing management by designing an AI-assisted abnormality detection system, solves the problem of low accuracy in manual judgment of weighing abnormalities, helps the system to detect abnormal weighing problems in time, reduces system failures and downtime, and improves the degree of automation;
[0056] 2. The present invention realizes the encrypted and secure storage function of weighing data by designing a blockchain application system, solves the problems of manual data tampering and data loss, improves the accuracy of system weighing, and strengthens the security of data management and transactions;
[0057] 3. The present invention realizes the optimization function of the operating efficiency of the weighing system by designing an adaptive optimization algorithm system, helps the system to achieve automatic optimization, adapt to the ever-changing environment and needs, and improves the system's adaptive ability and intelligence level;
[0058] 4. The present invention realizes the function of generating reports by analyzing and processing weighing data through the design of a data analysis report generation system, thereby improving data visualization, enhancing decision-making accuracy, and facilitating decision-making management and transaction transparency. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] Figure 1 It is a schematic diagram of the weighing management process of the present invention. DETAILED DESCRIPTION
[0060] 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.
[0061] Example 1
[0062] See also Figure 1 , an embodiment provided by the present invention: an unattended weighing management system, comprising a management control center and a cloud server, wherein the management control center is connected to the cloud server via the Internet, and an AI-assisted anomaly detection system is designed in the management control center, wherein the AI-assisted anomaly detection system is used to analyze and detect abnormal situations occurring during the weighing process;
[0063] The AI-assisted anomaly detection system comprises: a weighing data analysis unit, a real-time anomaly detection unit and an anomaly processing notification unit, wherein the weighing data analysis unit is connected to the real-time anomaly detection unit via a data interface, and the real-time anomaly detection unit is connected to the anomaly processing notification unit via a signal API interface;
[0064] The weighing data analysis unit integrates a data analysis server and a machine learning algorithm module, and uses the machine learning algorithm module to analyze and learn the historical weighing data in the data analysis server to obtain normal weighing patterns and rules;
[0065] The real-time anomaly detection unit integrates a real-time data processing module and an anomaly detection algorithm module, wherein the real-time data processing module is used to process real-time weighing data, and the anomaly detection algorithm module is used to detect weighing anomalies in real time and issue an alarm signal in time;
[0066] The abnormality handling notification unit integrates an alarm system and a notification device. The alarm system receives an alarm signal transmitted by the abnormality detection algorithm module to trigger the alarm system response. The system automatically takes processing measures and notifies relevant personnel through the notification device to handle the abnormal situation.
[0067] Furthermore, the weighing system collects real-time weighing data, including vehicle information, cargo information and weight data. The weighing data analysis unit cleans and preprocesses the collected data, including removing duplicate data, filling missing values, data conversion, etc., extracts meaningful features, such as vehicle type, cargo type, weighing time, etc., and uses the machine learning algorithm module to analyze and learn historical weighing data to obtain normal weighing patterns and rules;
[0068] The real-time anomaly detection unit uses the real-time data processing module to analyze and learn the historical weighing data, obtain the normal weighing mode and rules, extract features from the real-time data, such as vehicle type, cargo type, weight, etc., use the anomaly detection algorithm module to perform anomaly detection on the real-time data, determine whether the current weighing data is abnormal, and send the abnormal information to the abnormal processing notification unit;
[0069] The exception handling notification unit receives the exception information sent by the real-time exception detection unit, including the exception data and the exception type, and further analyzes and verifies the exception data to determine the exact cause and impact of the exception. According to the severity and impact of the exception, the alarm system is triggered and corresponding handling measures are taken, such as stopping the weighing operation, notifying relevant personnel, etc., and an exception handling notification is sent to relevant personnel, including exception information, handling measures and suggestions, to ensure that the exception is handled and resolved in a timely manner.
[0070] Example 2
[0071] See also Figure 1 , an embodiment provided by the present invention: an unattended weighing management system, wherein the AI-assisted anomaly detection system is connected to an intelligent recognition system via a data line, and the intelligent recognition system is used to intelligently recognize the weighing vehicle and cargo data information;
[0072] The intelligent identification system includes: a vehicle information identification unit, a cargo information identification unit and a data processing and storage unit, and the vehicle information identification unit and the cargo information identification unit are connected to the data processing and storage unit via a data line;
[0073] The vehicle information recognition unit integrates a high-definition camera installed at the entrance, a license plate recognition camera and an RGB sensor. The license plate number recognition adopts a CNN model based on deep learning. The high-definition camera and RGB sensor use image feature extraction and pattern matching algorithms to identify the vehicle model and color.
[0074] The cargo information identification unit integrates laser radar and weighing sensor. The laser radar is used to obtain cargo images, and the weighing sensor is used to measure the weight of the weighing vehicle in real time. The cargo type identification adopts a deep learning model based on image recognition, and the weighing data processing uses a sliding window algorithm to identify the cargo type, quantity and weight in real time and generate complete weighing data.
[0075] The data processing and storage unit is integrated with a database server, which is used to store weighing data, associate the identified vehicle and cargo information with the weighing information, and store it in a data analysis server;
[0076] Further, the vehicle information recognition unit obtains the image and video data of the vehicle through the high-definition camera, license plate recognition camera and RGB sensor installed at the entrance, performs image processing and feature extraction on the license plate area, uses the license plate recognition algorithm to recognize the license plate number, associates the recognized license plate number with information such as entry and exit time to form vehicle information data, stores the vehicle information data in a database or a data processing storage unit to ensure data security and accessibility, uses laser radar to perform image processing and feature extraction on the cargo feature area, uses the object recognition algorithm to recognize the cargo type, uses the weighing sensor to obtain the cargo weight data, associates the recognized cargo type and weight data with information such as entry and exit time to form complete cargo information data;
[0077] The data processing and storage unit receives the identification result data from the vehicle information identification unit and the cargo information identification unit, cleans, deduplicates and formats the received data to ensure data quality and consistency, stores the cleaned and pre-processed data in the data analysis server for subsequent data analysis and report generation, and manages the stored data, including data backup, recovery, permission control, etc., to ensure data security and reliability.
[0078] Example 3
[0079] See also Figure 1 , an embodiment provided by the present invention: an unattended weighing management system, wherein a blockchain application system is designed in the cloud server, and the blockchain application system is used to encrypt and process the vehicle information and weighing data in the database for storage;
[0080] The blockchain application system includes: data encryption storage unit, smart contract execution unit and data traceability audit unit;
[0081] The data encryption storage unit integrates an encryption module and a blockchain node interface. The encryption module is used to encrypt the data. The blockchain node is used to store the encrypted data and connect to the database through an API interface. The weighing data is encrypted and stored in the blockchain in the cloud server.
[0082] The smart contract execution unit integrates smart contract programming tools and smart contract execution nodes. Smart contract programming tools are used to write smart contracts, and smart contract execution nodes are used to execute contracts. Smart contracts are written to define the rights and interests of both parties and the contract terms. The smart contract execution on the blockchain ensures fair and transparent transactions.
[0083] The data traceability audit unit integrates a blockchain browser and an audit tool. The blockchain browser is used to view data on the blockchain, and the audit tool is used to audit the data. The blockchain technology is used to provide the traceability and audit functions of the weighing data.
[0084] Furthermore, the data generated by the weighing system, such as vehicle information, cargo information and weight data, is first encrypted by an encryption algorithm to ensure the privacy and security of the data. The encrypted data is packaged into transactions and broadcast and verified through the blockchain network. The nodes in the blockchain network verify and confirm the data transactions to ensure the integrity and authenticity of the data. The verified transaction data is added to the blocks of the blockchain to form an unalterable data record.
[0085] Write smart contracts according to business needs, including data verification, transaction logic, and permission control, and deploy the written smart contracts to the blockchain network to become executable smart contracts. Users initiate transaction requests through the system to trigger the execution of smart contracts. Smart contracts automatically execute transactions according to preset logic and conditions, and update the data status on the blockchain. The execution results of smart contracts are recorded on the blockchain, forming an unalterable transaction record;
[0086] Authorized users can trace back all historical records and transaction processes of data through the blockchain network to achieve data integrity and credibility. The data traceability audit unit on the blockchain network can audit the data to verify the authenticity and compliance of the data. The audit system will generate a data audit report, including information such as data source, transaction records, and compliance verification. The audit results are stored on the blockchain to form verifiable audit evidence to ensure the credibility and transparency of the data. Authorized users can query the audit results through the blockchain network to ensure the compliance and security of the data.
[0087] Example 4
[0088] See also Figure 1 , an embodiment provided by the present invention: an unattended weighing management system, wherein the AI-assisted anomaly detection system is designed with an adaptive optimization algorithm system, and the adaptive optimization algorithm system is used to analyze and learn the weighing process and optimize the process;
[0089] The adaptive optimization algorithm system includes: historical data analysis and learning unit, optimization algorithm implementation adjustment unit and effect evaluation feedback unit;
[0090] The historical data analysis and learning unit integrates a historical data processing module and a machine learning model training module. The historical data processing module is used to clean and preprocess historical data, and the machine learning model training module is used to establish an optimization model. The historical weighing data is analyzed and learned through the machine learning algorithm to obtain the rules and characteristics of the weighing process and output the optimization algorithm.
[0091] The optimization algorithm implementation and adjustment unit integrates the optimization algorithm implementation module and the parameter adjustment module. The optimization algorithm implementation module is used to execute the optimization algorithm, and the parameter adjustment module is used to dynamically adjust the optimization parameters. According to the historical data learning model, the weighing process is adaptively optimized and the best weighing scheme is selected.
[0092] The effect evaluation feedback unit integrates an effect evaluation module and a feedback mechanism module. The effect evaluation module is used to evaluate the optimization results, and the feedback mechanism module is used to feed back the evaluation results to the optimization algorithm, monitor the optimization results, evaluate the effect of the weighing process, and feed back the evaluation results to the optimization algorithm for adjustment;
[0093] Furthermore, historical data during the operation of the system is collected, including weighing data, system operating parameters, environmental conditions, etc. The historical data analysis and learning unit performs pre-processing operations such as cleaning, denoising, and feature extraction on the collected historical data for subsequent data analysis. The historical data is analyzed and learned using machine learning, data mining and other technologies to discover the correlation and regularity between the data. Based on the results of the historical data analysis, a model suitable for the system is trained, such as a prediction model, a classification model, etc.;
[0094] The optimization algorithm implementation adjustment unit sets the initial parameters and operation strategy of the system according to the model and rules obtained by the historical data analysis learning unit, and continuously monitors the real-time operation data of the system, including weighing efficiency, accuracy, system resource utilization, etc. According to the real-time data monitored and the model obtained by the historical data analysis learning unit, a suitable optimization algorithm, such as genetic algorithm, particle swarm algorithm, etc., is selected. The system parameters are adjusted through the optimization algorithm implementation adjustment unit to optimize the operation efficiency and performance of the system, and the effect of the adjusted system parameters is monitored in real time to obtain real-time feedback data;
[0095] The effect evaluation feedback unit conducts effect evaluation based on real-time monitoring feedback data, including indicators such as system operating efficiency, accuracy, and resource utilization, analyzes the evaluation results, evaluates the effect and impact of system parameter adjustments, and adjusts the strategies and parameters of the optimization algorithm implementation adjustment unit according to the evaluation results to optimize the system's adaptive capabilities. The evaluation results and adjustment strategies are fed back to the historical data analysis and learning unit and the optimization algorithm implementation adjustment unit to form a closed-loop feedback.
[0096] Example 5
[0097] See also Figure 1 , an embodiment provided by the present invention: an unattended weighing management system, wherein the management control center is further designed with a data analysis report generation system, the data analysis report generation system is connected to a cloud server and an adaptive optimization algorithm system through a data Internet interface, and the data analysis report generation system is used to generate a data analysis report in real time;
[0098] The data analysis report generation system includes: a real-time data analysis unit and a report generation and display unit, and the real-time data analysis unit is connected to the report generation and display unit through a network communication interface;
[0099] The real-time data analysis unit integrates a data analysis module and a real-time database. The data analysis module is used to analyze the real-time weighing data and the analysis results are stored in the real-time database.
[0100] The report generation and display unit integrates a report generation module and a data visualization tool, automatically generates reports based on data analysis results, and displays them on the display screen in the management control center through the data visualization tool;
[0101] Furthermore, the real-time data analysis unit is responsible for collecting real-time data during the operation of the system, including weighing data, system operating parameters, environmental conditions, etc., and performing pre-processing operations such as cleaning, denoising, and feature extraction on the collected real-time data to ensure the quality and accuracy of the data, monitor the system operation status and indicators, such as weighing efficiency, accuracy, resource utilization, etc., and promptly detect and respond to abnormal situations;
[0102] The report generation and display unit receives the analysis results and monitoring data provided by the real-time data analysis unit, summarizes and organizes the data, selects a report template based on the type and content of the report to be generated, fills the data analysis results and monitoring data into the report template, generates complete report content, and uses the report generation tool to generate a report file, including text, charts, statistical data, etc. The generated report is displayed on the interface of the report generation and display unit, provided to users for viewing and downloading, and displayed on the display screen in the management control center through data visualization tools.
[0103] Working principle: First, the vehicle and cargo information are identified through the vehicle information identification module and the cargo information identification module, and the type and weight information of the cargo are identified and recorded. The data information of these vehicles and cargoes is stored in the database to form a complete weighing data record;
[0104] Then, the AI-assisted anomaly detection system learns and analyzes the historical weighing data stored in the database, and detects and warns of abnormal situations. At the same time, the weighing system process is optimized through the adaptive optimization algorithm system.
[0105] Finally, the data analysis report generation system analyzes the weighing data and generates weighing and optimization data reports; at the same time, the smart contract execution module in the blockchain application system is used to adjust and optimize system parameters, thereby continuously improving the performance and efficiency of the weighing management system, and encrypting and storing the weighing process data to ensure data reliability and security.
[0106] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
Claims
1. An unattended weighing management system, Features: It includes a management control center and a cloud server. The management control center is connected to the cloud server via the Internet. An AI-assisted anomaly detection system is designed in the management control center. The AI-assisted anomaly detection system is used to analyze and detect abnormal situations that occur during the weighing process; The AI-assisted anomaly detection system comprises: a weighing data analysis unit, a real-time anomaly detection unit and an anomaly processing notification unit, wherein the weighing data analysis unit is connected to the real-time anomaly detection unit via a data interface, and the real-time anomaly detection unit is connected to the anomaly processing notification unit via a signal API interface; The weighing data analysis unit integrates a data analysis server and a machine learning algorithm module, and uses the machine learning algorithm module to analyze and learn the historical weighing data in the data analysis server to obtain normal weighing patterns and rules; The real-time anomaly detection unit integrates a real-time data processing module and an anomaly detection algorithm module, wherein the real-time data processing module is used to process real-time weighing data, and the anomaly detection algorithm module is used to detect weighing anomalies in real time and issue an alarm signal in time; The abnormality handling notification unit integrates an alarm system and a notification device. The alarm system receives an alarm signal transmitted by the abnormality detection algorithm module to trigger the alarm system response. The system automatically takes processing measures and notifies relevant personnel through the notification device to handle the abnormal situation. The cloud server is designed with a blockchain application system, which is used to encrypt and store vehicle information and weighing data in the database; The blockchain application system includes: data encryption storage unit, smart contract execution unit and data traceability audit unit; The data encryption storage unit integrates an encryption module and a blockchain node interface. The encryption module is used to encrypt the data. The blockchain node is used to store the encrypted data and connect to the database through an API interface. The weighing data is encrypted and stored in the blockchain in the cloud server. The smart contract execution unit integrates smart contract programming tools and smart contract execution nodes. Smart contract programming tools are used to write smart contracts, and smart contract execution nodes are used to execute contracts. Smart contracts are written to define the rights and interests of both parties and the contract terms. The smart contract execution on the blockchain ensures fair and transparent transactions. The data traceability audit unit integrates a blockchain browser and an audit tool. The blockchain browser is used to view data on the blockchain, and the audit tool is used to audit the data. Blockchain technology is used to provide the traceability and audit functions of weighing data. The AI-assisted anomaly detection system is designed with an adaptive optimization algorithm system, which is used to analyze and learn the weighing process and optimize the process; The adaptive optimization algorithm system includes: historical data analysis and learning unit, optimization algorithm implementation adjustment unit and effect evaluation feedback unit; The historical data analysis and learning unit integrates a historical data processing module and a machine learning model training module. The historical data processing module is used to clean and preprocess historical data, and the machine learning model training module is used to establish an optimization model. The historical weighing data is analyzed and learned through the machine learning algorithm to obtain the rules and characteristics of the weighing process and output the optimization algorithm. The optimization algorithm implementation and adjustment unit integrates the optimization algorithm implementation module and the parameter adjustment module. The optimization algorithm implementation module is used to execute the optimization algorithm, and the parameter adjustment module is used to dynamically adjust the optimization parameters. According to the historical data learning model, the weighing process is adaptively optimized and the best weighing scheme is selected. The effect evaluation feedback unit integrates an effect evaluation module and a feedback mechanism module. The effect evaluation module is used to evaluate the optimization results, and the feedback mechanism module is used to feed back the evaluation results to the optimization algorithm, monitor the optimization results, evaluate the effect of the weighing process, and feed back the evaluation results to the optimization algorithm for adjustment.
2. An unattended weighing management system according to claim 1, Features: The AI-assisted anomaly detection system is connected to an intelligent recognition system via a data cable, and the intelligent recognition system is used to intelligently identify the weighing vehicle and cargo data information; The intelligent identification system includes: a vehicle information identification unit, a cargo information identification unit and a data processing and storage unit, and the vehicle information identification unit and the cargo information identification unit are connected to the data processing and storage unit via a data line; The vehicle information recognition unit integrates a high-definition camera installed at the entrance, a license plate recognition camera and an RGB sensor. The license plate number recognition adopts a CNN model based on deep learning. The high-definition camera and RGB sensor use image feature extraction and pattern matching algorithms to identify the vehicle model and color. The cargo information identification unit integrates laser radar and weighing sensor. The laser radar is used to obtain cargo images, and the weighing sensor is used to measure the weight of the weighing vehicle in real time. The cargo type identification adopts a deep learning model based on image recognition, and the weighing data processing uses a sliding window algorithm to identify the cargo type, quantity and weight in real time and generate complete weighing data. The data processing storage unit integrates a database server, which is used to store weighing data, associate the identified vehicle and cargo information with the weighing information, and store it in the data analysis server.
3. The unattended weighing management system according to claim 1, Features: The management control center is also designed with a data analysis report generation system, which is connected to the cloud server and the adaptive optimization algorithm system through a data Internet interface, and is used to generate data analysis reports in real time; The data analysis report generation system includes: a real-time data analysis unit and a report generation and display unit, and the real-time data analysis unit is connected to the report generation and display unit through a network communication interface; The real-time data analysis unit integrates a data analysis module and a real-time database. The data analysis module is used to analyze the real-time weighing data and the analysis results are stored in the real-time database. The report generation and display unit integrates a report generation module and a data visualization tool, automatically generates a report based on the data analysis results, and displays it on the display screen in the management control center through the data visualization tool.
4. The unattended weighing management system according to claim 1, Features: The algorithm in the anomaly detection algorithm module in the real-time anomaly detection unit is the Z-score algorithm, and the specific formula is as follows: Assume that the weighing data set is x 1 , x 2 , ..., x n , the average value is x 0 , the standard deviation is s, then the data point x i The calculation formula is , where Zi represents the data point x i The Z-score of the data point is marked as abnormal if the threshold is set to 3 and -3.
5. The unattended weighing management system according to claim 2, Features: The sliding window algorithm formula used in the weighing data processing is as follows: Moving average calculation formula: , Among them, the moving average MA t , t represents the current time point, k represents the window size, x i Represents the weighing data at the i-th time point.
6. The unattended weighing management system according to claim 4, Features: The machine learning model trained by the machine learning model training module in the historical data analysis learning unit is a random forest model. The formula of the random forest model is as follows: For classification problems: , in, is the predicted category of sample x, f i (x) is the prediction result of the i-th decision tree, and MajorityVote represents the vote to decide the final category; For regression problems: , in, is the predicted value of sample x, f i (x) is the prediction result of the i-th decision tree, and n is the number of decision trees.
7. An unattended weighing management method, applicable to an unattended weighing management system according to any one of claims 1 to 6, Features: The management method comprises the following steps: S1. Use the vehicle information recognition module and cargo information recognition module in the intelligent recognition system to monitor the operating status and data collection of the weighing system in real time; S2. With the help of AI-assisted anomaly detection system, abnormal situations in the weighing process are monitored, alarms are issued in time, and corresponding measures are taken; S3. Use the adaptive optimization algorithm system to adjust the parameters and operation strategies of the weighing system in real time according to the historical data analysis results to optimize the system efficiency and accuracy; S4. Analyze the weighing data and generate a weighing report using the data analysis report generation system, and display the report on the display screen of the management control center; S5. Regularly perform equipment inspection, sensor calibration, and data cleaning system maintenance and care to ensure long-term and stable operation of the system.
8. An unattended weighing management method according to claim 7, Features: The management method further comprises the following steps: S11, the data processing storage unit associates the vehicle and cargo information identified by the vehicle information identification module and the cargo information identification module with the weighing information, and stores them in the data analysis server; S31. Use the machine learning model training module in the historical data analysis learning unit to study and analyze historical weighing data, generate reports and provide decision support.
