Intelligent dynamic weighing system and weighing method thereof

Abstract

The invention relates to the technical field of weighing measurement and metering, and discloses an intelligent dynamic weighing system and a weighing method thereof, and the system comprises a first-level hierarchical module which is used for achieving local real-time processing through a weighing sensor and a local data collection module; the second-level hierarchical module is used for carrying out fusion and cross validation on multi-sensor data through edge computing nodes; and the three-level hierarchical module is used for carrying out global optimization and monitoring analysis through the cloud computing node. Through hierarchical distribution design, the real-time performance and fault-tolerant capability of the system are remarkably improved, and the problems of precision deviation, overhigh energy consumption and response delay of a traditional system in a high-load and high-frequency dynamic weighing scene are solved. The local adaptive algorithm and the global optimization algorithm cooperate with each other, so that the system can dynamically adjust the weighing parameters in a complex environment, and the weighing precision is ensured not to be affected by the change of the external environment. And a multi-sensor data fusion algorithm is combined with edge calculation, so that the data reliability and the system stability are further improved.

Description

Technical Field

[0001] The invention relates to an intelligent dynamic weighing system and a weighing method thereof, belonging to the technical field of weighing measurement and metering. Background Art

[0002] In traditional intelligent dynamic weighing systems, a single sensor and centralized data processing method are usually used. This method has the following problems in high-load, high-frequency dynamic weighing scenarios: First, the data acquisition accuracy of a single sensor is easily affected by environmental factors (such as vibration, temperature, and electromagnetic interference), resulting in deviations in weighing results; second, the centralized data processing method requires a large amount of raw data to be transmitted to the central processor, resulting in data transmission delays and excessive energy consumption, making it difficult to meet real-time requirements; finally, the system has poor fault tolerance. Once a sensor or data processing node fails, the entire system may not be able to operate normally.

[0003] In order to solve the above problems, the existing technology usually adopts the following methods: first, increase the number of sensors and improve the accuracy through multi-sensor data collection; second, optimize the data transmission protocol and reduce the amount of data transmission; third, introduce redundant design to improve the fault tolerance of the system. However, these methods still have new problems: although increasing the number of sensors improves the accuracy of data collection, the fusion and cross-validation of multi-sensor data require complex algorithm support, which increases the computational burden; although optimizing the data transmission protocol reduces the amount of data, it cannot fundamentally solve the problem of data transmission delay; although redundant design improves the fault tolerance of the system, it increases the complexity and cost of the system. Summary of the invention

[0004] The present invention provides an intelligent dynamic weighing system and a weighing method thereof, the main purpose of which is to solve the problems of accuracy deviation, excessive energy consumption and response delay of traditional dynamic weighing systems under high load and high frequency scenarios.

[0005] To achieve the above-mentioned purpose, the present invention provides an intelligent dynamic weighing system, comprising: a primary level module, comprising a plurality of weighing sensors and a local data acquisition module, wherein the plurality of weighing sensors are arranged near the weighing point and used to collect weighing data in real time; the local data acquisition module is connected to the plurality of weighing sensors and used to perform local real-time processing on the weighing data, including dynamically adjusting the weighing parameters according to the weighing environment through a local adaptive algorithm, and compressing and preprocessing the original weighing data to generate local processed data; a secondary level module, comprising a plurality of edge computing nodes, each edge computing node is connected to at least one primary level module, and used to receive the local processed data, and weight the data from the plurality of primary level modules and cross-validate through a multi-sensor data fusion algorithm to generate fused data; the multi-sensor data fusion algorithm includes dynamically adjusting the weight coefficient according to the historical accuracy data and real-time environmental data of each weighing sensor, wherein the weight coefficient is calculated as follows: ,in, For the The weight coefficient of each sensor, For the Historical accuracy scores of sensors, For the Real-time environment matching of each sensor; The sensor index ranges from 1 to , is the total number of sensors in the system; and Respectively The three-level module includes a cloud computing node, which is used to receive fused data from multiple second-level modules, and globally optimize the fused data through a global optimization algorithm to generate optimized weighing data; the global optimization algorithm includes dynamically adjusting the weighing model according to external environmental parameters, and the external environmental parameters include temperature, humidity and electromagnetic interference intensity; wherein the first-level module, the second-level module and the third-level module realize multi-level linkage through a cross-level collaborative optimization mechanism, including: the second-level module sends feedback instructions to the first-level module according to the fused data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm.

[0006] As an embodiment of the present invention, the local adaptive algorithm includes the following steps: real-time monitoring of the vibration intensity, temperature changes and electromagnetic interference intensity of the weighing environment; dynamically adjusting the sampling frequency and filtering parameters of the weighing sensor according to the monitoring results; when the vibration intensity exceeds a preset threshold, automatically increasing the sampling frequency and enhancing the filtering intensity to ensure the accuracy of the weighing data.

[0007] As an embodiment of the present invention, the multi-sensor data fusion algorithm further includes: dynamically adjusting the weight coefficient of each weighing sensor according to the real-time signal strength and noise level of each weighing sensor; when the signal strength of a weighing sensor is lower than a preset threshold or the noise level is higher than a preset threshold, automatically reducing its weight coefficient and increasing the weight coefficients of other sensors to ensure the reliability of the fused data.

[0008] As an embodiment of the present invention, the global optimization algorithm includes the following steps: real-time monitoring of external environmental parameters, including temperature, humidity and electromagnetic interference intensity; dynamically adjusting the parameters of the weighing model according to the monitoring results to compensate for the impact of environmental changes on weighing accuracy; when the temperature change exceeds a preset range, automatically adjusting the temperature compensation coefficient of the weighing model to ensure the stability of the weighing results.

[0009] As an embodiment of the present invention, the cross-level collaborative optimization mechanism further includes: the second-level module sends feedback instructions to the first-level module according to the fusion data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm; through multi-level linkage, the system's adaptability and overall performance in different environments are ensured.

[0010] As an embodiment of the present invention, the edge computing node further includes: a data cache module for temporarily storing local processing data from the first-level hierarchical module; a data verification module for cross-validating the local processing data and eliminating abnormal data; and a data fusion module for generating fused data according to a multi-sensor data fusion algorithm.

[0011] As an embodiment of the present invention, the cloud computing node further includes: a historical data storage module for storing historical weighing data and environmental parameters; a real-time monitoring module for real-time monitoring of the operating status of weighing sensors, edge computing nodes and cloud computing nodes; a fault detection module for automatically taking over the function of the faulty node through a redundant node when a fault is detected.

[0012] As an embodiment of the present invention, the first-level hierarchical module further includes: a redundant weighing sensor, which is used to ensure that the weighing data is not affected by compensating the data of the adjacent nodes when a certain weighing sensor fails; a local data compression module, which is used to compress the weighing data to reduce the amount of data transmitted and reduce energy consumption; the second-level hierarchical module further includes: a dynamic weight adjustment module, which is used to dynamically adjust the weight coefficient of each weighing sensor according to its historical accuracy data and real-time environmental matching degree; a data cross-validation module, which is used to cross-validate data from multiple first-level hierarchical modules to ensure the reliability and consistency of the data; the third-level hierarchical module further includes: a machine learning module, which is used to perform predictive analysis on the fused data and generate weighing trend prediction results; a global optimization module, which is used to dynamically adjust the parameters of the global optimization algorithm according to the prediction results to ensure the long-term stability and accuracy of the system in different environments.

[0013] As an embodiment of the present invention, the weighing method corresponding to the intelligent dynamic weighing system includes the following steps: Step 1, data acquisition and local processing: real-time collection of weighing data by multiple weighing sensors arranged near the weighing point; local real-time processing of the collected weighing data by a local data acquisition module, including dynamically adjusting weighing parameters according to the vibration intensity, temperature change and electromagnetic interference intensity of the weighing environment through a local adaptive algorithm, and compressing and preprocessing the original weighing data to generate local processing data; Step 2, data fusion and cross-validation: transmitting the local processing data to the edge computing node; weighting and cross-validation of data from multiple first-level modules through a multi-sensor data fusion algorithm to generate fused data; the multi-sensor data fusion algorithm includes dynamically adjusting the weight coefficient according to the historical accuracy data and real-time environmental data of each weighing sensor, and the calculation formula of the weight coefficient is: ,in, For the The weight coefficient of each sensor, For the Historical accuracy scores of sensors, For the The real-time environment matching degree of each sensor, is the sensor index, and Respectively The historical accuracy scores and real-time environmental matching degrees of each sensor are collected; Step 3, global optimization and monitoring analysis: the fused data is transmitted to the cloud computing node; the fused data is globally optimized through the global optimization algorithm to generate optimized weighing data; the global optimization algorithm includes dynamically adjusting the weighing model according to the external environmental parameters, and the external environmental parameters include temperature, humidity and electromagnetic interference intensity; Step 4, cross-level collaborative optimization: the second-level module sends feedback instructions to the first-level module according to the fused data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm; through multi-level linkage, the system's adaptability and overall performance in different environments are ensured.

[0014] Compared with the problems described in the background technology, the beneficial effects of the present invention are: through the local real-time processing of the first-level module, the data fusion and cross-validation of the second-level module, and the global optimization of the third-level module, the system realizes the hierarchical distribution of data processing. This architecture not only reduces the data transmission delay, but also ensures that the system can still operate stably when some nodes fail through redundant design and local calculation, significantly improving the real-time and fault-tolerant capabilities of the system. For example, when a weighing sensor fails, the system can compensate the data of the adjacent nodes to ensure that the weighing data is not affected; the local adaptive algorithm of the first-level module can dynamically adjust the sampling frequency and filtering parameters of the weighing sensor according to the vibration intensity, temperature change and electromagnetic interference intensity of the weighing environment, while the global optimization algorithm of the third-level module dynamically adjusts the parameters of the weighing model according to the deviation of historical data and real-time data. This local and global collaborative optimization mechanism enables the system to maintain high-precision weighing under different environments and loads, solving the problem of accuracy deviation of traditional systems in complex environments; the second-level module uses a multi-sensor data fusion algorithm to weight and cross-validate the data from multiple first-level modules, combined with the real-time processing capability of the edge computing node, significantly improving the reliability of the data and weighing accuracy. For example, the algorithm can dynamically adjust the weight coefficient of each weighing sensor according to its real-time working status (such as signal strength, noise level and failure rate) to ensure the accuracy and stability of the weighing data. Through the cross-level collaborative optimization mechanism between the first-level module, the second-level module and the third-level module, the system can achieve multi-level linkage and dynamically adjust the sampling interval, data processing method and fusion strategy of the weighing sensor. This mechanism not only improves the adaptive ability of the system, but also ensures the efficiency and stability of the system in different application scenarios, and solves the problems of response delay and excessive energy consumption of traditional systems in high-load and high-frequency dynamic weighing scenarios. The three-level module predicts and analyzes the fused data through machine learning algorithms, generates weighing trend prediction results, and dynamically adjusts the parameters of the global optimization algorithm according to the prediction results. At the same time, the real-time monitoring module monitors the operating status of the weighing sensor, edge computing node and cloud computing node, and automatically takes over the function of the faulty node through the redundant node when a fault is detected. This design significantly improves the intelligence level and reliability of the system, enabling it to operate efficiently in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is the architecture diagram of the intelligent dynamic weighing system of the present invention.

[0016] Figure 2 This is a diagram of the cross-level collaborative optimization mechanism of the intelligent dynamic weighing system of the present invention.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0019] The embodiment of the present application provides an intelligent dynamic weighing system and a weighing method thereof. It includes: a first-level hierarchical module, including multiple weighing sensors and a local data acquisition module, wherein the multiple weighing sensors are arranged near the weighing point and used to collect weighing data in real time; the local data acquisition module is connected to the multiple weighing sensors and used to perform local real-time processing of the weighing data, including dynamically adjusting the weighing parameters according to the weighing environment through a local adaptive algorithm, and compressing and preprocessing the original weighing data to generate local processing data; a second-level hierarchical module, including multiple edge computing nodes, each edge computing node is connected to at least one first-level hierarchical module, used to receive the local processing data, and weight the data from multiple first-level hierarchical modules and cross-validate through a multi-sensor data fusion algorithm to generate fused data; the multi-sensor data fusion algorithm includes dynamically adjusting the weight coefficient according to the historical accuracy data and real-time environmental data of each weighing sensor, wherein the weight coefficient is calculated as follows: ,in, For the The weight coefficient of each sensor, For the Historical accuracy scores of sensors, For the Real-time environment matching of each sensor; The sensor index ranges from 1 to , is the total number of sensors in the system; and Respectively The three-level module includes a cloud computing node, which is used to receive fused data from multiple second-level modules, and globally optimize the fused data through a global optimization algorithm to generate optimized weighing data; the global optimization algorithm includes dynamically adjusting the weighing model according to external environmental parameters, and the external environmental parameters include temperature, humidity and electromagnetic interference intensity; wherein the first-level module, the second-level module and the third-level module realize multi-level linkage through a cross-level collaborative optimization mechanism, including: the second-level module sends feedback instructions to the first-level module according to the fused data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm.

[0020] As an embodiment of the present invention, the local adaptive algorithm includes the following steps: real-time monitoring of the vibration intensity, temperature changes and electromagnetic interference intensity of the weighing environment; dynamically adjusting the sampling frequency and filtering parameters of the weighing sensor according to the monitoring results; when the vibration intensity exceeds a preset threshold, automatically increasing the sampling frequency and enhancing the filtering intensity to ensure the accuracy of the weighing data.

[0021] As an embodiment of the present invention, the multi-sensor data fusion algorithm further includes: dynamically adjusting the weight coefficient of each weighing sensor according to the real-time signal strength and noise level of each weighing sensor; when the signal strength of a weighing sensor is lower than a preset threshold or the noise level is higher than a preset threshold, automatically reducing its weight coefficient and increasing the weight coefficients of other sensors to ensure the reliability of the fused data.

[0022] As an embodiment of the present invention, the global optimization algorithm includes the following steps: real-time monitoring of external environmental parameters, including temperature, humidity and electromagnetic interference intensity; dynamically adjusting the parameters of the weighing model according to the monitoring results to compensate for the impact of environmental changes on weighing accuracy; when the temperature change exceeds a preset range, automatically adjusting the temperature compensation coefficient of the weighing model to ensure the stability of the weighing results.

[0023] As an embodiment of the present invention, the cross-level collaborative optimization mechanism further includes: the second-level module sends feedback instructions to the first-level module according to the fusion data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm; through multi-level linkage, the system's adaptability and overall performance in different environments are ensured.

[0024] As an embodiment of the present invention, the edge computing node further includes: a data cache module for temporarily storing local processing data from the first-level hierarchical module; a data verification module for cross-validating the local processing data and eliminating abnormal data; and a data fusion module for generating fused data according to a multi-sensor data fusion algorithm.

[0025] As an embodiment of the present invention, the cloud computing node further includes: a historical data storage module for storing historical weighing data and environmental parameters; a real-time monitoring module for real-time monitoring of the operating status of weighing sensors, edge computing nodes and cloud computing nodes; a fault detection module for automatically taking over the function of the faulty node through a redundant node when a fault is detected.

[0026] As an embodiment of the present invention, the first-level hierarchical module further includes: a redundant weighing sensor, which is used to ensure that the weighing data is not affected by compensating the data of the adjacent nodes when a certain weighing sensor fails; a local data compression module, which is used to compress the weighing data to reduce the amount of data transmitted and reduce energy consumption; the second-level hierarchical module further includes: a dynamic weight adjustment module, which is used to dynamically adjust the weight coefficient of each weighing sensor according to its historical accuracy data and real-time environmental matching degree; a data cross-validation module, which is used to cross-validate data from multiple first-level hierarchical modules to ensure the reliability and consistency of the data; the third-level hierarchical module further includes: a machine learning module, which is used to perform predictive analysis on the fused data and generate weighing trend prediction results; a global optimization module, which is used to dynamically adjust the parameters of the global optimization algorithm according to the prediction results to ensure the long-term stability and accuracy of the system in different environments.

[0027] As an embodiment of the present invention, the weighing method corresponding to the intelligent dynamic weighing system includes the following steps: Step 1, data acquisition and local processing: real-time collection of weighing data by multiple weighing sensors arranged near the weighing point; local real-time processing of the collected weighing data by a local data acquisition module, including dynamically adjusting weighing parameters according to the vibration intensity, temperature change and electromagnetic interference intensity of the weighing environment through a local adaptive algorithm, and compressing and preprocessing the original weighing data to generate local processing data; Step 2, data fusion and cross-validation: transmitting the local processing data to the edge computing node; weighting and cross-validation of data from multiple first-level modules through a multi-sensor data fusion algorithm to generate fused data; the multi-sensor data fusion algorithm includes dynamically adjusting the weight coefficient according to the historical accuracy data and real-time environmental data of each weighing sensor, and the calculation formula of the weight coefficient is: ,in, For the The weight coefficient of each sensor, For the Historical accuracy scores of sensors, For the The real-time environment matching degree of each sensor, is the sensor index, and Respectively The historical accuracy scores and real-time environmental matching degrees of each sensor are collected; Step 3, global optimization and monitoring analysis: the fused data is transmitted to the cloud computing node; the fused data is globally optimized through the global optimization algorithm to generate optimized weighing data; the global optimization algorithm includes dynamically adjusting the weighing model according to the external environmental parameters, and the external environmental parameters include temperature, humidity and electromagnetic interference intensity; Step 4, cross-level collaborative optimization: the second-level module sends feedback instructions to the first-level module according to the fused data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm; through multi-level linkage, the system's adaptability and overall performance in different environments are ensured.

[0028] Embodiment 1: The present invention relates to the field of intelligent dynamic weighing technology, and specifically to a dynamic weighing system based on a distributed multi-level architecture. For example, in logistics warehousing scenarios, high-frequency and high-load dynamic weighing requirements are extremely common. Traditional centralized weighing systems often have difficulty in coping with weighing requirements in complex environments due to their reliance on a single sensor and a central processor. Taking a large logistics center as an example, it needs to process thousands of cargo weighings every day. Traditional weighing systems are prone to accuracy deviation, response delay and excessive energy consumption under high load. In an embodiment of the present invention, the system includes a first-level module, a second-level module and a third-level module. The first-level module consists of a plurality of weighing sensors and a local data acquisition module. The weighing sensors are arranged at key positions of the cargo conveyor belt for real-time collection of cargo weight data. The local data acquisition module dynamically adjusts the sampling frequency and filtering parameters of the weighing sensor according to the vibration intensity, temperature change and electromagnetic interference intensity of the weighing environment through a local adaptive algorithm to generate local processing data. For example, when the environmental vibration intensity is high, the system automatically increases the sampling frequency and adjusts the filtering parameters to ensure the accuracy of the weighing data. The secondary level module consists of multiple edge computing nodes, each of which is connected to multiple primary level modules, receives local processing data and performs data fusion and cross-validation through a multi-sensor data fusion algorithm. The algorithm dynamically adjusts the weight coefficient of each weighing sensor according to its real-time working status (such as signal strength, noise level, and failure rate) to generate fused data. For example, when the signal strength of a weighing sensor decreases due to a failure, the system automatically reduces its weight coefficient to ensure the reliability of the weighing data. The tertiary level module consists of cloud computing nodes, receives fused data from multiple secondary level modules, and performs global optimization and monitoring analysis through a global optimization algorithm. The global optimization algorithm dynamically adjusts the parameters of the weighing model according to the deviation between historical weighing data and real-time weighing data to ensure the stability of weighing accuracy under different environments. For example, when the system detects that the temperature change causes weighing deviation, it automatically adjusts the parameters of the weighing model to ensure that the weighing result is not affected.

[0029] Example 2: In industrial production scenarios, such as a steel plant that needs to weigh thousands of tons of steel every day, traditional weighing systems have problems with accuracy deviation, response delay and excessive energy consumption in high-load and high-frequency scenarios. In this embodiment, the system includes a first-level module, a second-level module and a third-level module. The first-level module consists of multiple weighing sensors and a local data acquisition module. The weighing sensors are arranged at key positions of the steel conveyor belt to collect steel weight data in real time. The local data acquisition module dynamically adjusts the sampling frequency and filtering parameters of the weighing sensor according to the vibration intensity, temperature change and electromagnetic interference intensity of the weighing environment through a local adaptive algorithm to generate local processing data. For example, when the environmental vibration intensity is high, the system automatically increases the sampling frequency and adjusts the filtering parameters to ensure the accuracy of the weighing data. The second-level module consists of multiple edge computing nodes, each of which is connected to multiple first-level modules, receives local processing data and performs data fusion and cross-validation through a multi-sensor data fusion algorithm. The algorithm dynamically adjusts its weight coefficient according to the real-time working status of each weighing sensor (such as signal strength, noise level and failure rate) to generate fused data. For example, when a weighing sensor fails and the signal strength drops, the system automatically reduces its weight coefficient to ensure the reliability of the weighing data. The three-level module is composed of cloud computing nodes, which receive fused data from multiple second-level modules and perform global optimization and monitoring analysis through a global optimization algorithm. The global optimization algorithm dynamically adjusts the parameters of the weighing model based on the deviation between historical weighing data and real-time weighing data to ensure the stability of weighing accuracy in different environments. For example, when the system detects that temperature changes cause weighing deviations, it automatically adjusts the parameters of the weighing model to ensure that the weighing results are not affected.

[0030] Embodiment 3: The intelligent dynamic weighing system of this embodiment still adopts a distributed multi-level architecture, including three-level modules: a first-level module, a second-level module and a third-level module. By introducing an improved edge computing algorithm and a data caching mechanism in the second-level module, the real-time and fault-tolerance of the system are enhanced. First-level module: It includes multiple weighing sensors and local data acquisition modules. The sensors are located at key positions and collect weighing data in real time. The local data acquisition module dynamically adjusts the sampling frequency and filtering parameters according to environmental monitoring data (such as vibration intensity, temperature changes, etc.). In order to reduce the amount of data transmission and reduce energy consumption, all original weighing data are compressed and local processing data is generated. Second-level module: It includes multiple edge computing nodes, each of which is responsible for receiving data from multiple first-level modules. Data caching and cross-validation functions are implemented inside each edge node. Specifically, the data caching module temporarily stores data from the first-level module, and the data verification module removes abnormal data to ensure data accuracy. Three-level module: Global optimization analysis is performed in the cloud, and the parameters of the weighing model are dynamically adjusted through the global optimization algorithm in combination with historical data and real-time data. The cloud module is also responsible for real-time monitoring and fault detection. Once a system failure is detected, data will be automatically taken over through redundant nodes to ensure that the system is not interrupted.

[0031] The core advantage of this embodiment lies in its cross-level collaborative optimization mechanism, which not only improves the weighing accuracy, but also significantly reduces data transmission delay and energy consumption. Local adaptive algorithm: This algorithm monitors the vibration intensity, temperature changes and electromagnetic interference intensity in the environment in real time, and automatically adjusts the sampling frequency and filtering parameters of the weighing sensor according to environmental changes. For example, when the system detects that the vibration intensity exceeds the set threshold, it automatically increases the sampling frequency and strengthens the filtering to ensure the accuracy and stability of data collection; Multi-sensor data fusion: The edge computing node in the secondary level module adopts a multi-sensor data fusion algorithm to dynamically adjust the weight coefficient of each sensor according to the real-time signal strength and noise level. The weight coefficient calculation formula is as follows: ,in, For the The weight coefficient of each sensor, For the Historical accuracy scores of sensors, For the The real-time environment matching degree of each sensor, is the total number of sensors in the system, and Respectively The historical accuracy score and real-time environmental matching degree of each sensor; Global optimization algorithm: The three-level module dynamically adjusts the parameters of the weighing model through the global optimization algorithm in the cloud, combined with external environmental parameters (such as temperature, humidity, magnetic field interference, etc.). When the external environmental parameters change, the system automatically adjusts the compensation coefficient in the model to cope with the impact of environmental changes on weighing accuracy; Cross-level collaborative optimization: The second-level module sends feedback instructions to the first-level module based on the fusion data to adjust the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module based on the optimized data to adjust the weight coefficient of the multi-sensor data fusion algorithm. Through this mechanism, the system can achieve adaptive adjustment according to different environments and load conditions to ensure the optimization of overall performance. In specific implementation, the local adaptive algorithm and the multi-sensor data fusion algorithm can be implemented through existing embedded systems and computing platforms. Each edge computing node includes a data cache module, a data verification module, and a data fusion module, which can complete real-time data processing with low latency. In implementation, it is first necessary to arrange sensors according to the application scenario, ensure that the weighing sensors cover key locations, and configure the data acquisition module. Secondly, configure the edge computing node to process and optimize the data. Finally, the entire system is monitored in real time and optimized globally through the cloud module, and a distributed architecture with three-level modules is implemented, so that the system can achieve efficient data processing and optimization in a high-load, high-frequency dynamic weighing environment.

[0032] Example 4: In this example, the optimization focus of the local adaptive algorithm is to monitor the vibration intensity, temperature change and electromagnetic interference intensity in the weighing environment in real time, and dynamically adjust the sampling frequency and filtering parameters of the weighing sensor according to these real-time data. In order to improve the accuracy of the weighing data, we have added a threshold judgment mechanism for vibration intensity and temperature change. When the system detects that the environmental vibration intensity exceeds the set threshold, it will automatically increase the sampling frequency of the weighing sensor and enhance the filtering intensity to ensure the accuracy of the data; when the temperature change exceeds the preset range, the system will automatically adjust the temperature compensation coefficient of the sensor to ensure the stability of the weighing result. This optimization step is performed by the embedded computing module, which can respond to environmental changes in real time and ensure the reliability and accuracy of the weighing data. For the optimization of the multi-sensor data fusion algorithm, we introduce a more accurate environmental matching model to monitor the working status of each weighing sensor in detail. When the signal strength of the sensor is lower than the preset threshold or the noise level exceeds the specified range, the system will automatically reduce its weight in the data fusion and increase the weight of other sensors. The dynamic adjustment of this weight is based on the historical accuracy score of each sensor ( ) and real-time environment matching ( Specifically, the weight coefficient is calculated based on the following two main parameters: historical accuracy score ( ) reflects the performance of the sensor in past data, and the real-time environment matching degree ( ) characterizes the degree of adaptability of the current environment to the sensor output data. The weight coefficient is dynamically adjusted through the following path: Calculate the historical accuracy score and real-time environment matching degree of each sensor. Dynamically adjust the sensor weight according to the historical accuracy score and environment matching degree to ensure that the actual operating status and current environmental factors of each sensor are taken into account during data fusion.

[0033] The key to the global optimization algorithm is to further optimize the fused data from modules at all levels through cloud computing nodes. The optimization process mainly includes adjusting the various parameters of the weighing model according to external environmental parameters (such as temperature, humidity, electromagnetic interference intensity, etc.). In this embodiment, we introduced a real-time environmental monitoring module that can automatically update the optimization parameters when the environment changes. For example, when the temperature change causes weighing deviation, the system will automatically adjust the temperature compensation coefficient through the real-time optimization algorithm, and dynamically adjust the weighing model according to the deviation of historical data and real-time data to ensure the weighing accuracy and stability under different environments. In order to further improve the real-time and fault tolerance of the system, this embodiment refines the cross-level collaborative optimization mechanism. In the secondary level module, the edge computing node is not only responsible for data fusion and cross-validation, but also can send instructions to the primary level module through the feedback mechanism to dynamically adjust the parameters of the local adaptive algorithm. At the same time, the third level module can send optimization instructions to the secondary level module according to the optimized weighing data, and dynamically adjust the weight coefficient of the multi-sensor data fusion algorithm. Through this cross-level collaborative optimization, the system can dynamically adjust the operating parameters of each level in a complex environment to ensure high efficiency and stability under different load conditions. And in order to further improve the fault tolerance of the system, this embodiment adds redundant weighing sensors and redundant computing nodes. When a weighing sensor fails, the system can ensure the continuity and stability of the weighing data through data compensation of adjacent nodes. When a computing node fails, the redundant computing node will automatically take over the task to ensure that the system will not suffer performance degradation due to a single point failure. The specific implementation path of this embodiment includes hardware requirements: multiple high-precision weighing sensors and edge computing nodes need to be configured, and the integration of cloud computing platforms is supported; data acquisition and processing: weighing data is processed in real time through local data acquisition modules, and adjusted through adaptive algorithms to ensure data accuracy and real-time performance; edge computing and cloud collaboration: local data is processed in real time through edge computing nodes, and data is further optimized in the cloud through global optimization algorithms to ensure the accuracy and stability of weighing results; algorithm and module configuration: ensure that the configuration of each module can realize data flow and collaborative work, and the feedback mechanism and instruction transmission between modules can realize real-time adjustment and optimization.

[0034] Embodiment 5: This embodiment further optimizes the system's fault tolerance, real-time performance, and accuracy of multi-sensor data fusion. In this embodiment, the intelligent dynamic weighing system still adopts a three-level hierarchical module architecture, namely, a first-level module, a second-level module, and a third-level module. Each module will continue to perform task assignments according to the original design, but the collaboration between modules at each level is optimized to ensure rapid data transmission and processing. Enhancement of the first-level module: The first-level module consists of multiple weighing sensors and local data acquisition modules, and the weighing sensors are arranged in key positions to achieve real-time data acquisition. The local data acquisition module further introduces an adaptive adjustment algorithm to dynamically detect and adjust the weighing parameters, especially in environments with high vibration intensity, temperature changes, and electromagnetic interference intensity. The system will automatically increase the sampling frequency of the sensor and optimize the filtering parameters to improve the accuracy and stability of data acquisition. Optimization of the second-level module: The second-level module consists of multiple edge computing nodes, each of which receives local processing data from multiple first-level modules. In order to further optimize the efficiency of data fusion and cross-validation, the second-level module will use an improved multi-sensor data fusion algorithm. By analyzing the signal strength and noise level from multiple sensors in real time, the system dynamically adjusts the sensor weight coefficient based on each sensor's historical accuracy score and environmental matching. The implementation path of this process is as follows: Data is transmitted from the primary module to the secondary module. The secondary module calculates the real-time signal strength and noise level of each sensor. The weight coefficient is dynamically adjusted based on the real-time data to increase the data weight of sensors with higher signal strength and lower noise. Improvements to the third-level module: The cloud computing node of the third-level module receives the fused data from the second-level module and further processes the data through a global optimization algorithm. The global optimization algorithm will take into account the impact of external environmental parameters (such as temperature, humidity, and electromagnetic interference) on the weighing model to ensure that when the environment changes, the system can automatically adjust the compensation coefficient to maintain the stability of the weighing results.

[0035] In order to ensure the stability of the system in high-load and high-frequency dynamic weighing scenarios, this embodiment particularly strengthens the fault-tolerant mechanism. The first-level module introduces a redundant sensor design. Once a weighing sensor fails, the system can automatically compensate the data of the adjacent sensors to ensure that the weighing data is not affected. The redundant design can ensure that the entire system can still work normally without obvious data deviations in the event of a sensor failure. The second-level module further strengthens the fault-tolerant processing and adds data caching, abnormal data detection and automatic adjustment functions. The data cache module will cache data from the first-level module to prevent calculation interruptions due to data loss. The data verification module cross-validates the received data, eliminates abnormal data, and passes reliable data to the next-level module.

[0036] The local adaptive algorithm monitors the vibration intensity, temperature and electromagnetic interference of the weighing environment in real time. The sampling frequency and filtering parameters of the sensor are dynamically adjusted. If the vibration intensity exceeds the preset threshold, the sampling frequency is increased and the filtering intensity is strengthened. When the environmental changes cause data deviation, the sensor sampling frequency and filtering parameters are automatically adjusted to ensure the accuracy of the data. In the secondary module, the multi-sensor data fusion algorithm dynamically adjusts the weight coefficient of the sensor by analyzing the real-time signal strength and noise level of each sensor. When the signal strength of a sensor is lower than the set threshold, or the noise level exceeds the preset range, the system will automatically reduce the weight coefficient of the sensor and enhance the weight of other sensors to ensure the accuracy and stability of the data fusion results. The global optimization algorithm realizes global optimization through cloud computing nodes, dynamically adjusts the compensation coefficient of the weighing model according to the deviation between historical data and real-time data, and when the ambient temperature fluctuates greatly, the system will automatically adjust the temperature compensation coefficient to ensure the stability of the weighing results under different environmental conditions. At the same time, when implementing the intelligent dynamic weighing system, it is necessary to configure multiple high-precision weighing sensors and arrange them reasonably in key positions, such as on a cargo conveyor belt, to ensure that each object can be weighed accurately in real time when passing through the conveyor belt. Each sensor needs to be connected to a local data acquisition module, which processes the data collected by the sensor through an adaptive algorithm and adjusts relevant parameters in real time. In the secondary level module, multiple edge computing nodes need to be configured to receive local data from the primary module and perform data fusion and verification. The data fusion algorithm will dynamically adjust the weight coefficient based on the historical accuracy score of the sensor and the real-time environmental data to ensure the accuracy and reliability of the weighing data. The cloud computing node will further optimize the data from the secondary module and adjust the model parameters through a global optimization algorithm to ensure the stability of the weighing accuracy. In addition, the system also needs to implement redundant sensor design and fault detection mechanisms to ensure that the system can automatically switch and continue to provide accurate weighing data when a fault occurs.

[0037] Example 6: This example optimizes the data weight allocation and cross-validation steps in the multi-sensor data fusion algorithm to ensure that the system's data fusion process is more accurate, reliable, and has efficient real-time processing capabilities. For example, when the system is running, multiple weighing sensors are arranged in different positions to collect weighing data in real time. These sensor data are affected by environmental factors (such as temperature, humidity, vibration, etc.) and the performance of the sensor itself, so they need to be preliminarily processed by the local data acquisition module. This module dynamically adjusts the sampling frequency and filtering parameters according to the working environment of each sensor through a local adaptive algorithm. For example, when the sensor is in a high vibration environment, its sampling frequency will increase and the filtering intensity will be enhanced to remove more noise and ensure the accuracy of the data. The sensor data collected by each first-level module will be further processed and fused through the edge computing node. The key step in data fusion is weight allocation, that is, adjusting the importance of each sensor in the fusion process according to its historical accuracy score and real-time environmental matching; historical accuracy score ( ): reflects the accuracy of the sensor in the past work, which can be obtained through the analysis of system historical data; real-time environment matching ( ): Based on the real-time monitoring data, it indicates the working matching degree of the sensor in the current state; the calculation of the weight coefficient is realized in the following way: first, a weight is assigned to each sensor according to the historical accuracy score and real-time environmental matching degree of each sensor; when the historical accuracy of the sensor is high and the current environmental impact is small (that is, the real-time environmental matching degree is high), its weight will increase accordingly, otherwise it will decrease; in order to ensure the reliability and consistency of the data, this embodiment introduces a cross-validation mechanism. The specific process of cross-validation is as follows: the data from multiple first-level hierarchical modules are checked for consistency to ensure that the difference between the sensor data does not exceed the preset threshold; if the data of a certain sensor deviates from the data of other sensors beyond the set range, the data is judged to be abnormal and automatically removed. At this time, the system will estimate and fill in the missing values ​​based on the data of other sensors to ensure the integrity of the data; this process is automatically executed using an algorithm without manual intervention, thereby ensuring efficient data processing. After the fusion data is completed, the cross-validated data will be further input into the third-level hierarchical module (that is, the cloud computing node) for global optimization. This step uses a global optimization algorithm to fine-tune the fusion data and compensate for the impact of environmental factors on weighing accuracy. For example, the system will dynamically adjust the optimization coefficient according to parameters such as ambient temperature and humidity to ensure that the system can always maintain high-precision weighing under different environmental conditions. And the system ensures dynamic feedback and adaptive adjustment of modules at each level through a cross-level collaborative optimization mechanism: after data fusion and cross-validation, the secondary level module will feed back the data to the primary level module to adjust the filter parameters and sampling frequency in the local data acquisition module. The third level module sends adjustment instructions to the secondary level module through the feedback mechanism based on the optimized weighing data to optimize the weight distribution algorithm. This embodiment ensures that data with higher accuracy can be given priority in a multi-sensor environment by dynamically adjusting the weight coefficient of the sensor, thereby improving the data fusion accuracy of the entire system. Its cross-validation mechanism: introduces a cross-validation algorithm to effectively eliminate abnormal data and ensure the consistency and reliability of the final fused data; real-time environmental adaptability: through real-time monitoring and adjustment of system parameters, it ensures that under different environmental conditions, the system can dynamically adjust to cope with external changes.

[0038] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limiting. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An intelligent dynamic weighing system, characterized in that: include: A first-level module includes a plurality of weighing sensors and a local data acquisition module, wherein the plurality of weighing sensors are arranged near the weighing point and are used to collect weighing data in real time; the local data acquisition module is connected to the plurality of weighing sensors and is used to perform local real-time processing of the weighing data, including dynamically adjusting the weighing parameters according to the weighing environment through a local adaptive algorithm, and compressing and preprocessing the original weighing data to generate local processed data; a second-level module includes a plurality of edge computing nodes, each of which is connected to at least one first-level module, and is used to receive the local processed data, and weight the data from the plurality of first-level modules and cross-validate through a multi-sensor data fusion algorithm to generate fused data; the multi-sensor data fusion algorithm includes dynamically adjusting the weight coefficient according to the historical accuracy data and real-time environmental data of each weighing sensor, wherein the weight coefficient is calculated as follows: ,in, For the The weight coefficient of each sensor, For the Historical accuracy scores of sensors, For the Real-time environment matching of each sensor; The sensor index ranges from 1 to , is the total number of sensors in the system; and Respectively The three-level module includes a cloud computing node, which is used to receive fused data from multiple second-level modules, and globally optimize the fused data through a global optimization algorithm to generate optimized weighing data; the global optimization algorithm includes dynamically adjusting the weighing model according to external environmental parameters, and the external environmental parameters include temperature, humidity and electromagnetic interference intensity; wherein the first-level module, the second-level module and the third-level module realize multi-level linkage through a cross-level collaborative optimization mechanism, including: the second-level module sends feedback instructions to the first-level module according to the fused data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm.

2. The intelligent dynamic weighing system according to claim 1, characterized in that: The local adaptive algorithm comprises the following steps: Real-time monitoring of the vibration intensity, temperature changes and electromagnetic interference intensity of the weighing environment; dynamically adjust the sampling frequency and filtering parameters of the weighing sensor according to the monitoring results; when the vibration intensity exceeds the preset threshold, automatically increase the sampling frequency and enhance the filtering intensity.

3. The intelligent dynamic weighing system according to claim 1, characterized in that: The multi-sensor data fusion algorithm further includes: dynamically adjusting the weight coefficient of each weighing sensor according to the real-time signal strength and noise level of each weighing sensor; when the signal strength of a weighing sensor is lower than a preset threshold or the noise level is higher than a preset threshold, automatically reducing its weight coefficient and increasing the weight coefficients of other sensors.

4. The intelligent dynamic weighing system according to claim 1, characterized in that: The global optimization algorithm includes the following steps: real-time monitoring of external environmental parameters, including temperature, humidity and electromagnetic interference intensity; dynamically adjusting the parameters of the weighing model according to the monitoring results to compensate for the impact of environmental changes on weighing accuracy; and automatically adjusting the temperature compensation coefficient of the weighing model when the temperature changes exceed the preset range.

5. The intelligent dynamic weighing system according to claim 1, characterized in that: The cross-level collaborative optimization mechanism further includes: the second-level module sends feedback instructions to the first-level module according to the fusion data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm; through multi-level linkage, the system's adaptability and overall performance in different environments are ensured.

6. The intelligent dynamic weighing system according to claim 1, characterized in that: The edge computing node further includes: a data cache module for temporarily storing local processing data from the first-level hierarchical module; a data verification module for cross-validating the local processing data and eliminating abnormal data; and a data fusion module for generating fused data according to a multi-sensor data fusion algorithm.

7. The intelligent dynamic weighing system according to claim 1, characterized in that: The cloud computing node further includes: a historical data storage module for storing historical weighing data and environmental parameters; a real-time monitoring module for real-time monitoring of the operating status of the weighing sensor, edge computing node and cloud computing node; and a fault detection module for automatically taking over the function of the faulty node through a redundant node when a fault is detected.

8. The intelligent dynamic weighing system according to claim 1, characterized in that: The first-level module further includes: redundant weighing sensors, which are used to ensure that the weighing data is not affected by compensating the data of the adjacent nodes when a weighing sensor fails; a local data compression module, which is used to compress the weighing data to reduce the amount of data transmission and reduce energy consumption; the second-level module further includes: a dynamic weight adjustment module, which is used to dynamically adjust the weight coefficient of each weighing sensor according to its historical accuracy data and real-time environmental matching degree; a data cross-validation module, which is used to cross-validate the data from multiple first-level modules to ensure the reliability and consistency of the data; the third-level module further includes: a machine learning module, which is used to perform predictive analysis on the fused data and generate weighing trend prediction results; a global optimization module, which is used to dynamically adjust the parameters of the global optimization algorithm according to the prediction results.

9. The intelligent dynamic weighing system according to claim 1, characterized in that: The weighing method corresponding to the intelligent dynamic weighing system includes the following steps: Step 1, data acquisition and local processing: real-time collection of weighing data by multiple weighing sensors arranged near the weighing point; local real-time processing of the collected weighing data by using a local data acquisition module, including dynamically adjusting weighing parameters according to the vibration intensity, temperature change and electromagnetic interference intensity of the weighing environment through a local adaptive algorithm, and compressing and preprocessing the original weighing data to generate local processing data; Step 2, data fusion and cross-validation: transmitting the local processing data to the edge computing node; weighting and cross-validation of data from multiple first-level modules through a multi-sensor data fusion algorithm to generate fused data; the multi-sensor data fusion algorithm includes dynamically adjusting the weight coefficient according to the historical accuracy data and real-time environmental data of each weighing sensor, and the calculation formula of the weight coefficient is: ,in, For the The weight coefficient of each sensor, For the Historical accuracy scores of sensors, For the The real-time environment matching degree of each sensor, is the sensor index, and Respectively The historical accuracy scores and real-time environmental matching degrees of each sensor are collected; Step 3, global optimization and monitoring analysis: the fused data is transmitted to the cloud computing node; the fused data is globally optimized through the global optimization algorithm to generate optimized weighing data; the global optimization algorithm includes dynamically adjusting the weighing model according to the external environmental parameters, and the external environmental parameters include temperature, humidity and electromagnetic interference intensity; Step 4, cross-level collaborative optimization: the second-level module sends feedback instructions to the first-level module according to the fused data, and dynamically adjusts the parameters of the local adaptive algorithm; the third-level module sends optimization instructions to the second-level module according to the optimized weighing data, and dynamically adjusts the weight coefficient of the multi-sensor data fusion algorithm.

Images