A sandstone production process monitoring management system based on big data
The sand and gravel production process monitoring and management system, established using big data technology, has solved the decision-making conflict problem caused by the independent operation of modules in the sand and gravel production system. It has realized the automatic coordination of production planning and quality monitoring, thereby improving production efficiency and product quality stability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HANGZHOU GAOXUN INTERNET OF THINGS TECH CO LTD
- Filing Date
- 2026-02-04
- Publication Date
- 2026-06-05
AI Technical Summary
In existing sand and gravel production systems, each module operates independently, resulting in a lack of coordination mechanisms between production planning, process monitoring, and quality inspection. This leads to decision-making conflicts and an excessive burden of manual coordination, affecting production efficiency and product quality stability.
By establishing a big data-based monitoring and management system for sand and gravel production processes, including data acquisition and transmission, data standardization processing, event detection and rule matching, collaborative control and chain response, as well as optimization and decision support modules, automatic collaboration and intelligent decision-making in production planning, process control and quality monitoring can be achieved.
It enables the automatic conversion of production instructions into quality thresholds and equipment control parameters, automatically identifies and resolves module conflicts, improves the overall efficiency and responsiveness of production management, reduces reliance on manual judgment, and enhances production management efficiency and product quality stability.
Smart Images
Figure REF-OBJ-1770192826283-000002 
Figure REF-OBJ-1770192826283-000003 
Figure REF-OBJ-1770192826283-000004
Abstract
Description
Technical Field
[0001] This invention relates to the field of sand and gravel production technology, and in particular to a sand and gravel production process monitoring and management system based on big data. Background Technology
[0002] With the continuous expansion of infrastructure construction, the quality control and production efficiency of sand and gravel aggregates, as important building materials, are receiving increasing attention. In recent years, sand and gravel production enterprises have gradually introduced information management systems, deploying various sensors and equipment to collect and monitor production data. However, these systems often merely transform traditional manual records into electronic records, lacking effective collaboration mechanisms between different business modules.
[0003] The above-disclosed technical solutions have at least the following technical problems: In traditional methods, because system modules such as production planning, process monitoring, and quality inspection operate independently, data cannot be automatically transferred and transformed between modules, leading to a disconnect between production instructions and quality control requirements. Specifically, the output targets issued by the production planning module cannot be automatically converted into monitoring thresholds by the quality module; equipment adjustments by the process control module cannot consider quality constraints in real time; and when decision-making conflicts arise between modules based on their own objectives, manual coordination is required, significantly increasing the decision-making burden on operators and affecting production efficiency and product quality stability. To address these problems, this invention proposes a solution. Summary of the Invention
[0004] This application provides a big data-based sand and gravel production process monitoring and management system, which solves the problems of decision-making conflicts and excessive manual coordination caused by the independent operation of each production module and the lack of a collaborative mechanism in the prior art. It realizes automatic collaboration and intelligent decision-making of production planning, process control and quality monitoring, and significantly improves management efficiency and product quality stability.
[0005] This application provides a big data-based sand and gravel production process monitoring and management system, including: a data acquisition and transmission module: used to acquire multi-source heterogeneous process data in real time and transmit the multi-source heterogeneous process data to a central server; Data standardization processing module: used to preprocess the received multi-source heterogeneous process data and generate standardized real-time production data streams; Event detection and rule matching module: used to establish an event-condition-action rule library, detect predefined events in the production process in real time, and match the corresponding processing rules; Collaborative control and chain response module: used to make decisions and schedule according to the priority of processing rules, and automatically execute preset collaborative control actions; Optimization and Decision Support Module: This module is used to dynamically optimize the parameters of the rule base based on historical data and effect evaluation results of the processing rules.
[0006] Further, the preprocessing steps for the received multi-source heterogeneous process data include: Preprocessing includes data quality assessment and outlier correction; The steps involved in data quality assessment include: For each collected data point, the reliability of that data set is calculated using the following formula: ; In the formula, Indicates data credibility. This indicates the deviation between the current sampled value and the historical average value of this parameter. This represents the historical arithmetic mean of the monitored parameter over the past production cycle. This indicates the data collection time interval between the current data point and the previous data point. Indicates the maximum allowed data collection time interval. Indicates the sensor calibration coefficient. , , These are the deviation weighting factor, the time weighting factor, and the calibration weighting factor, respectively.
[0007] Furthermore, the steps for generating standardized real-time production data streams include: Based on data credibility Multi-source heterogeneous process data is filtered and weighted to generate a real-time production data stream in a unified format; When data credibility If the data point is below a preset confidence threshold, it is marked and excluded. The retained data points are normalized to unify their numerical range to the [0,1] interval; Add timestamps and data source identifiers to form a standardized real-time production data stream.
[0008] Furthermore, the steps to establish an event-condition-action rule base include: Define event types, which include material quality exceeding standards events, critical equipment abnormal operating status events, and production plan change events; Define conditional judgment parameters, which include event duration threshold, parameter deviation magnitude threshold, and data trend change rate; Define action types, which include instructions to adjust equipment operating parameters, push early warning information to managers at different levels, and control instructions to intervene in the material flow of the production line.
[0009] Furthermore, the steps for making decisions and scheduling based on the priority of processing rules, and automatically executing preset collaborative control actions, include: The priority of each triggered rule is obtained through the rule priority calculation formula; The coordinated control actions are executed sequentially from highest to lowest priority. The priority calculation formula is as follows: ; In the formula, This indicates the priority score of the rule. This represents the scope coefficient of the impact of the rule on the production process after its execution. This represents the urgency coefficient of the event corresponding to the rule. This represents the consistency coefficient between the current rule and the queue of triggered rules. , , These are the weights for the scope of impact, urgency, and rule consistency, respectively.
[0010] Furthermore, the sensor calibration coefficients are obtained as follows: During the equipment calibration cycle, the sensor calibration coefficient is calculated using the following formula: ; In the formula, This represents the standard calibration value input from a standard metrology device. This indicates the actual reading output by the sensor after receiving the standard calibration value. To calibrate the attenuation factor.
[0011] Furthermore, the method for obtaining the rate of change of data trend is as follows: When determining a trend, select at least three different historical time scales and calculate the weighted average rate of change of the trend using the following formula: ; In the formula, Indicates the rate of change of data trend. Indicates the number of time scales used. Indicates the sequence number of the time scale. Indicates the first Weighting coefficients for each time scale This represents the sensor sample value at the current moment. Indicates the first Historical arithmetic mean over a given time scale.
[0012] Furthermore, the formula for calculating the consistency coefficient of the rule queue is as follows: ; In the formula, This indicates the cumulative number of times that the action of the currently pending rule is mutually exclusive with the actions in the queue of already triggered pending rules. This indicates the total number of times the rule has been triggered in the history.
[0013] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: By establishing an event-condition-action rule base and a chain response mechanism, previously independent modules such as production planning, process monitoring, and quality inspection are tightly linked. This allows production instructions to be automatically converted into specific quality thresholds and equipment control parameters, effectively breaking down information barriers between modules and achieving fully automated collaborative control, thus improving the overall integrity and responsiveness of production management. Furthermore, during collaborative control, rule priority calculation and conflict resolution mechanisms automatically identify and adjudicate conflicting instructions from different modules, prioritizing the execution of critical actions and avoiding confusion in internal decision-making logic, ensuring the stability and consistency of the production process. Moreover, during system operation, a dynamic optimization mechanism based on historical execution data and effect evaluation continuously learns and adjusts rule parameters, making the system more adaptable and gradually reducing reliance on manual judgment. This liberates operators from complex multi-objective decision-making dilemmas, truly achieving intelligent workload reduction and improving the efficiency and reliability of production management. Attached Figure Description
[0014] Figure 1 This is a schematic diagram of the structure of a sand and gravel production process monitoring and management system based on big data, provided in an embodiment of this application. Detailed Implementation
[0015] This application provides a big data-based sand and gravel production process monitoring and management system, which solves the problems of decision-making conflicts and heavy manual coordination burden caused by the independent operation of each production module in the prior art. By establishing an event-condition-action rule base, implementing rule priority calculation and chain response mechanism, it realizes the automatic conversion of production instructions to quality control parameters, intelligent adjudication of conflict instructions, and collaborative management and control of the entire process.
[0016] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.
[0017] like Figure 1As shown, this application embodiment provides a big data-based sand and gravel production process monitoring and management system, including: a data acquisition and transmission module: used to collect multi-source heterogeneous process data, including material weight, particle size distribution, equipment operating parameters and logistics weighing, in real time through a sensor network deployed on the sand and gravel production line, and transmit the multi-source heterogeneous process data to the central server; The sensor network includes electronic belt scales, online particle size analyzers, motor current sensors, PLC controllers, and weighbridge load cells. The electronic belt scales are installed on the crusher and finished sand conveyor belts for real-time material weight measurement; the online particle size analyzer is installed above the finished sand conveyor belt and detects particle size distribution using laser diffraction; the motor current sensors are connected to the crusher and screening machine motors to monitor equipment operating parameters; the PLC controller collects status signals from each piece of equipment on the production line; and the weighbridge load cells are located at the material inlet and outlet to record material weighing data.
[0018] Data standardization processing module: used to preprocess the received multi-source heterogeneous process data, including data cleaning, data normalization and data alignment, to generate a standardized real-time production data stream with credibility identifiers for subsequent analysis by the rule engine; Event detection and rule matching module: used to establish a configurable event-condition-action rule base, monitor the standardized real-time production data stream through the rule engine, detect predefined events in the production process in real time, and match the corresponding processing rules; The steps for matching the corresponding processing rules include: constructing a network-like pattern matching structure using the Rete algorithm to process the rule conditions. Specifically, this involves: generating a condition network from all conditions in the rule base; then feeding the facts from the standardized data stream into the network for pattern matching; and triggering a rule when all condition nodes of a certain rule are activated.
[0019] Collaborative Control and Chain Response Module: When a predefined event is detected and meets the triggering conditions of the corresponding processing rule, it makes decisions and schedules according to the priority of the processing rule, and automatically executes the preset collaborative control action; the new data generated by the execution of the collaborative control action will be used as a new event to trigger subsequent rules, thereby forming a cross-module rule chain response; The steps for detecting whether a predefined event meets the triggering conditions of the corresponding processing rule include: First, defining the triggering conditions, including the event type, duration, and threshold; second, evaluating the event data in real time and checking whether it meets the conditions through logical expressions, such as meeting the deviation requirements within a specified time after the event occurs; then, if the conditions are met, the system initiates rule execution; finally, recording the judgment results for subsequent analysis.
[0020] The optimization and decision support module is used to generate visualized early warning information and decision support reports based on historical data and effect evaluation results of the processing rules, and to dynamically optimize the parameters of the rule base using the evaluation results.
[0021] Dynamic optimization employs a parameter adjustment method based on historical data. The steps include: collecting rule execution records, including response time, execution success rate, and cost data. Response time refers to the time interval from event triggering to rule action completion; execution success rate is the ratio of the number of successful rule executions to the total number of triggers; cost data includes quantitative indicators such as computational resource consumption and manual intervention costs. Evaluation metrics for the effectiveness of calculation rules: ; In the formula, This represents the rule response time, expressed in seconds, which is the time difference between the moment the event is triggered and the moment the action is completed. The success rate of rule execution is dimensionless, with a value range of [0,1], and is obtained through statistics from historical execution records. The cost of rule enforcement is dimensionless, and it comprehensively quantifies resource consumption and operational costs. These are weighting coefficients, dimensionless, configured according to business needs, reflecting the relative importance of each evaluation dimension.
[0022] when When the rate decreases by more than 10% for three consecutive cycles, parameter optimization is initiated: the rule-based condition parameters are adjusted using gradient descent, with the adjustment range being... ,in The learning rate controls the step size of parameter adjustments; This indicates that the rule effectiveness evaluation index applies to specific conditional parameters. The partial derivatives reflect the sensitivity of the effect of parameter changes. Through iterative adjustment... It converges towards the optimization direction.
[0023] Further, the preprocessing steps for the received multi-source heterogeneous process data include: Preprocessing includes data quality assessment and outlier correction; The steps involved in data quality assessment include: For each collected data point, the reliability of that data set is calculated using the following formula: ; In the formula, This indicates the reliability of the data; a higher value indicates more reliable data quality. This indicates the deviation between the current sampled value and the historical average value of this parameter. This represents the historical arithmetic mean of the monitored parameter over the past production cycle. This indicates the data collection time interval between the current data point and the previous data point. Indicates the maximum allowed data collection time interval. Indicates the sensor calibration coefficient. , , These are the deviation weighting factor, the time weighting factor, and the calibration weighting factor, all of which are constants preset based on empirical data and satisfy the following conditions: .
[0024] The steps for outlier correction include: The moving average of a data series is calculated using the sliding window method. and standard deviation The calculation formula is as follows: ; In the formula, This is a moving average, reflecting the central tendency of the data series, with the same units as the original data. Standard deviation measures the dispersion of data; its unit is the same as the original data. The size of the sliding window determines the timeliness and stability of the statistical features. These are the original sampling data points; the units depend on the specific monitoring parameters. An improved Raida criterion is used for outlier identification, with the following criteria: ; in, The statistical significance coefficient is determined based on the normality test results of the data distribution. For datasets that are approximately normally distributed, Take the quantiles of the standard normal distribution; for skewed datasets, Adjustments should be made according to the actual distribution characteristics.
[0025] The identified outliers are replaced using a weighted replacement method based on temporal correlation. The replacement formula is as follows: ; in, The replaced value. and These are the values of the sampling points adjacent to the outlier, respectively. , , For the weighting coefficients, satisfying The weighting coefficients are determined based on a combination of the temporal distance between each neighboring point and the outlier, and the data reliability.
[0026] Repeat the above steps until no new outliers are detected in the data sequence, ensuring the stability of the correction results.
[0027] Furthermore, the steps for generating standardized real-time production data streams include: Based on data credibility Multi-source heterogeneous process data is filtered and weighted to generate a real-time production data stream in a unified format; Specifically, this includes: when data credibility If the data point is below a preset confidence threshold, it is marked and excluded. The retained data points are normalized to unify their numerical range to the [0,1] interval; Add timestamps and data source identifiers to form a standardized real-time production data stream.
[0028] The weighted fusion employs a credibility-based adaptive weighting method, the specific steps of which include: for each data source, weighting based on its credibility... As the initial weight, i.e. ; Normalize the weights: ,in This represents the sum of the initial weights of all data sources, followed by a weighted average: ; Calculate the difference coefficient for each data source (The ratio of standard deviation to mean), when At that time, the second weighting is initiated: ; Then, the weights are normalized and the fusion calculation is performed again.
[0029] in, For the first The initial weights of each data source reflect the original credibility level of a single data source; they are dimensionless. For the first The reliability of data from each data source The weights are normalized and standardized to ensure that the sum of the weights is 1. They are dimensionless. These are the source data values, the raw monitoring data from various data sources, with units consistent with the specific monitoring parameters. To integrate the data values, the final result after weighted calculation has units consistent with the source data.
[0030] Furthermore, the steps to establish an event-condition-action rule base include: Define event types, which include material quality exceeding standards events, critical equipment abnormal operating status events, and production plan change events; Define conditional judgment parameters, which include event duration threshold, parameter deviation magnitude threshold, and data trend change rate; Define action types, which include instructions to adjust equipment operating parameters, push early warning information to managers at different levels, and control instructions to intervene in the material flow of the production line.
[0031] Furthermore, the steps for making decisions and scheduling based on the priority of processing rules, and automatically executing preset collaborative control actions, include: The priority of each triggered rule is obtained through the rule priority calculation formula; The coordinated control actions are executed sequentially from highest to lowest priority. The priority calculation formula is as follows: ; In the formula, This indicates the priority score of the rule. This represents the impact coefficient on the production process after the rule is executed. This coefficient is based on the number of devices affected by the rule. and number of process steps The quantitative results are as follows: ,in For the total number of devices, This represents the total number of process steps. , These are the weighting coefficients. This represents the urgency coefficient of the event corresponding to the rule. The coefficient is assigned from a table based on the event type: 1.0 for quality and safety events, 0.8 for equipment failure events, and 0.6 for efficiency anomaly events. This assignment table is based on historical event processing priority statistics. This represents the consistency coefficient between the current rule and the queue of triggered rules. , , These are the influence scope weight, urgency weight, and rule consistency weight, all of which are constants preset based on expert experience and satisfy the following conditions: .
[0032] Furthermore, the sensor calibration coefficients are obtained as follows: During the equipment calibration cycle, the sensor calibration coefficient is calculated using the following formula: ; In the formula, This represents the standard calibration value input from a standard metrology device. This indicates the actual reading output by the sensor after receiving the standard calibration value. To calibrate the attenuation factor, its value is set between 0.8 and 0.95, depending on the degree of sensor aging and the severity of the working environment. The harsher the environment or the more severe the equipment aging, the higher the value. The closer the value is to 0.95.
[0033] Furthermore, the method for obtaining the rate of change of data trend is as follows: When determining a trend, select at least three different historical time scales, such as short-term, medium-term, and long-term, to reflect instantaneous changes, recent trends, and long-term directions, respectively. Calculate the weighted average rate of change of the trend using the following formula: ; In the formula, Indicates the rate of change of data trend. Indicates the number of time scales used, and its value is an integer greater than or equal to 3. Indicates the sequence number of the time scale. Indicates the first The weighting coefficients for each time scale are as follows: the closer the time scale is to the current time, the greater its weighting coefficient, and the sum of all weighting coefficients is 1. This represents the sensor sample value at the current moment. Indicates the first Historical arithmetic mean over a given time scale.
[0034] Furthermore, the formula for calculating the consistency coefficient of the rule queue is as follows: ; In the formula, This indicates the cumulative number of times that the action of the currently pending rule is mutually exclusive with the actions in the queue of already triggered pending rules. This indicates the total number of times the rule has been triggered in the history.
[0035] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.
[0036] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.
[0037] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0038] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.
[0039] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0040] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A big data-based monitoring and management system for sand and gravel production processes, characterized in that, include: Data acquisition and transmission module: used to acquire multi-source heterogeneous process data in real time and transmit the multi-source heterogeneous process data to the central server; Data standardization processing module: used to preprocess the received multi-source heterogeneous process data and generate standardized real-time production data streams; Event detection and rule matching module: used to establish an event-condition-action rule library, detect predefined events in the production process in real time, and match the corresponding processing rules; Collaborative control and chain response module: used to make decisions and schedule according to the priority of processing rules, and automatically execute preset collaborative control actions; Optimization and Decision Support Module: This module is used to dynamically optimize the parameters of the rule base based on historical data and effect evaluation results of the processing rules.
2. The sand and gravel production process monitoring and management system based on big data as described in claim 1, characterized in that, The preprocessing steps for the received multi-source heterogeneous process data include: Preprocessing includes data quality assessment and outlier correction; The steps involved in data quality assessment include: For each collected data point, the reliability of that data set is calculated using the following formula: ； In the formula, Indicates data credibility. This indicates the deviation between the current sampled value and the historical average value of this parameter. This represents the historical arithmetic mean of the monitored parameter over the past production cycle. This indicates the data collection time interval between the current data point and the previous data point. Indicates the maximum allowed data collection time interval. Indicates the sensor calibration coefficient. , , These are the deviation weighting factor, the time weighting factor, and the calibration weighting factor, respectively.
3. The sand and gravel production process monitoring and management system based on big data as described in claim 1, characterized in that, The steps to generate standardized real-time production data streams include: Based on data credibility Multi-source heterogeneous process data is filtered and weighted to generate a real-time production data stream in a unified format; When data credibility If the data point is below a preset confidence threshold, it is marked and excluded. The retained data points are normalized to unify their numerical range to the [0,1] interval; Add timestamps and data source identifiers to form a standardized real-time production data stream.
4. The sand and gravel production process monitoring and management system based on big data as described in claim 1, characterized in that, The steps to build an event-condition-action rule base include: Define event types, which include material quality exceeding standards events, critical equipment abnormal operating status events, and production plan change events; Define conditional judgment parameters, which include event duration threshold, parameter deviation magnitude threshold, and data trend change rate; Define action types, which include instructions to adjust equipment operating parameters, push early warning information to managers at different levels, and control instructions to intervene in the material flow of the production line.
5. The sand and gravel production process monitoring and management system based on big data as described in claim 1, characterized in that, The steps for making decisions and scheduling based on the priority of processing rules, and automatically executing preset collaborative control actions include: The priority of each triggered rule is obtained through the rule priority calculation formula; The coordinated control actions are executed sequentially from highest to lowest priority. The priority calculation formula is as follows: ； In the formula, This indicates the priority score of the rule. This represents the scope coefficient of the impact of the rule on the production process after its execution. This represents the urgency coefficient of the event corresponding to the rule. This represents the consistency coefficient between the current rule and the queue of triggered rules. , , These are the weights for the scope of impact, urgency, and rule consistency, respectively.
6. The sand and gravel production process monitoring and management system based on big data as described in claim 2, characterized in that, The sensor calibration coefficients are obtained as follows: During the equipment calibration cycle, the sensor calibration coefficient is calculated using the following formula: ； In the formula, This represents the standard calibration value input from a standard metrology device. This indicates the actual reading output by the sensor after receiving the standard calibration value. To calibrate the attenuation factor.
7. The sand and gravel production process monitoring and management system based on big data as described in claim 4, characterized in that, The method for obtaining the rate of change of data trend is as follows: When determining a trend, select at least three different historical time scales and calculate the weighted average rate of change of the trend using the following formula: ； In the formula, Indicates the rate of change of data trend. Indicates the number of time scales used. Indicates the sequence number of the time scale. Indicates the first Weighting coefficients for each time scale This represents the sensor sample value at the current moment. Indicates the first Historical arithmetic mean over a given time scale.
8. The sand and gravel production process monitoring and management system based on big data as described in claim 5, characterized in that, The formula for calculating the consistency coefficient of the rule queue is: ； In the formula, This indicates the cumulative number of times that the action of the currently pending rule is mutually exclusive with the actions in the queue of already triggered pending rules. This indicates the total number of times the rule has been triggered in the history.