Intelligent manufacturing production quality control system based on big data technology

Abstract

The invention, which relates to the technical field of the control system, discloses an intelligent manufacturing production quality control system based on a big data technology, comprising a data acquisition module, an edge calculation module, a cloud processor, a management control module and a man-machine interaction module. According to the invention, a product full-life-cycle data platform is comprehensively constructed from multiple angles of energy monitoring, pollution monitoring and resource scheduling, energy efficiency evaluation of product manufacturing and production quality is realized, multi-source data fusion is carried out through the data acquisition module, a unified data platform is constructed, integration and sharing of multi-source heterogeneous data are realized, and cloud edge cooperative computing is combined, so that energy efficiency evaluation of product manufacturing and production quality is realized. Real-time processing and analysis of data are guaranteed, efficiency and real-time performance are improved, global process parameter optimization and man-machine cooperation early warning processing are carried out through the management control module and the man-machine interaction module, the intelligent application level of the product production management system is improved, and the product quality is improved. And finally, the optimized transformation of energy conservation, environment friendliness, pollution reduction and sustainable development in the product manufacturing industry is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of control systems, and particularly to an intelligent manufacturing production quality control system based on big data technology. Background Art

[0002] In the current global context of advocating sustainable development, the product manufacturing industry is facing unprecedented challenges and opportunities. With the rapid development of technology, consumers' requirements for product quality are increasing day by day. At the same time, society's attention to environmental protection and rational utilization of resources is also constantly increasing.

[0003] During the product manufacturing process, multiple links and departments are involved, and the data sources are extensive and diverse in format, including equipment sensor data, production report data, environmental monitoring data, etc. These multi-source heterogeneous data are difficult to effectively integrate and share, resulting in enterprises being unable to comprehensively and accurately understand the production situation, and unable to effectively and timely remind operators to take measures in case of production anomalies, which affects production efficiency and product quality.

[0004] Moreover, traditional data processing methods often adopt centralized computing, transmitting all data to the cloud for processing. This method has problems such as data transmission delay and low processing efficiency, and cannot meet the needs of real-time decision-making. When there is abnormal energy consumption during the production process, due to untimely data processing, measures cannot be taken in time for adjustment, resulting in further aggravation of energy waste.

[0005] Therefore, the existing production quality control systems have problems of low data integration efficiency and insufficient real-time data analysis, as well as defects such as lack of comprehensive energy efficiency evaluation of the entire product life cycle and poor human-computer interaction experience. In view of the above technical defects, a solution is proposed. Summary of the Invention

[0006] The purpose of the present invention is to solve the problems of low data integration efficiency and insufficient real-time data analysis existing in the existing production quality control systems, as well as the defects of lack of comprehensive energy efficiency evaluation of the entire product life cycle and poor human-computer interaction experience.

[0007] To achieve the above purpose, the present invention adopts the following technical solutions: An intelligent manufacturing production quality control system based on big data technology, including a data acquisition module, an edge computing module, a cloud processor, a management control module, and a human-computer interaction module. Among them, the data acquisition module, the edge computing module, the cloud processor, the management control module, and the human-computer interaction module are communicatively connected; The data acquisition module is used to collect production cycle data: the production cycle data includes energy monitoring parameters, pollution monitoring parameters, and resource scheduling parameters; The edge computing module is used to standardize the production cycle data: by means of energy monitoring parameters, pollution monitoring parameters and resource scheduling parameters, the energy consumption degree, environmental pollution degree and resource utilization degree of product manufacturing production are evaluated in turn, and then the product manufacturing production quality is comprehensively evaluated, and the corresponding abnormal warning signals are output; The cloud processor is used to deeply analyze and optimize the production cycle data: by deeply analyzing the energy consumption degree, environmental pollution degree and resource utilization degree of product manufacturing production, a global process parameter optimization scheme is obtained, so as to construct a product full life cycle data chain and realize product quality traceability and prompt; The management control module is used to receive the global process parameter optimization scheme and generate the corresponding management control instructions to realize the adaptive adjustment and optimization of the system; The human-computer interaction module is used to receive the abnormal warning signal and perform the corresponding warning prompt operation, so as to prompt the management personnel to perform the corresponding processing.

[0008] Furthermore, the specific process of standardization processing is as follows: S2-1, estimate and fill in the missing values: ; Among them, mark the data value corresponding to any time node as , then is the previous time node, and the corresponding data value is , is the next time node, and the corresponding data value is ; S2-2, screen and process the outliers: If , then determine that Xi is an outlier; Among them, aveX is the parameter average value, and stdX is the parameter standard deviation; S2-3, determine the linear correlation between variables: Pearson correlation coefficient γxy: ; Among them, is the overall covariance of the two variables, and stdX and stdY are the standard deviations of variable X and variable Y respectively; γxy is positively correlated with , and negatively correlated with stdX and stdY; the value range of the Pearson correlation coefficient γxy is [-1,1]; Set the evaluation interval of the overall Pearson correlation coefficient γxy, and evaluate the correlation between variable X and variable Y through interval comparison, so as to determine and perform regression analysis; S2-4, output the positive correlation variable set Qa and the negative correlation variable set Qb; Mark any element of the set of positively correlated variables \(Q_a\) as variable \(A_q\); Mark any element of the set of negatively correlated variables \(Q_b\) as variable \(B_q\); S2-5, assign a corresponding weight coefficient \(\varphi_q\) to variable \(q\) according to the strength of correlation; Mark the weight coefficient of the positively correlated variable \(A_q\) as \(\varphi_{aq}\), and mark the weight coefficient of the negatively correlated variable \(B_q\) as \(\varphi_{bq}\); S2-6, comprehensively obtain the standardized parameter value \(\varPsi\) through \(n_1\) positively correlated variables \(A_q\), \(n_2\) negatively correlated variables \(B_q\) and their corresponding weight coefficients.

[0009] Furthermore, set the data acquisition period \(T_x\) to collect production cycle data, collect production cycle data through sensors, SCADA systems and MES systems, and align the timestamps of the production cycle data. The specific parameters are as follows: Energy monitoring parameters include electricity consumption information, water consumption information and gas consumption information; Electricity consumption information includes total power consumption, power consumption of each device, power, voltage and current; Water consumption information includes total water consumption, water consumption of each process, water pressure and water flow; Gas consumption information includes total gas consumption, gas consumption of each device, gas pressure and gas flow; Pollution monitoring parameters include solid waste information, gaseous waste information, liquid waste information and noise information; Solid waste information includes solid waste generation and hazardous waste generation; Gaseous waste information includes waste gas emissions, SO₂ concentration, NOₓ concentration and dust concentration; Liquid waste information includes wastewater emissions, COD concentration, BOD concentration, ammonia nitrogen concentration and total phosphorus concentration; Noise information includes noise intensity and noise frequency; Resource scheduling parameters include raw material information, personnel information and equipment information; Raw material information includes raw material input and raw material utilization rate; Personnel information includes the number of personnel, personnel working hours and personnel working hour utilization rate; Equipment information includes equipment operation time, equipment utilization rate and equipment failure rate.

[0010] Furthermore, the specific steps for standardizing the production cycle data and evaluating the energy consumption level, environmental pollution level and resource utilization level of product manufacturing production are as follows: By standardizing the energy monitoring parameters, the standardized parameter values of electricity consumption information, water consumption information, and gas consumption information are obtained. Furthermore, the energy consumption assessment index ECindex is comprehensively obtained. By setting the assessment interval of the energy consumption assessment index ECinde and conducting interval comparison, the energy consumption degree of product manufacturing production is evaluated; By standardizing the pollution monitoring parameters, the standardized parameter values of solid waste information, gas waste information, liquid waste information, and noise information are obtained. Furthermore, the environmental pollution assessment index EPindex is comprehensively obtained. By setting the assessment interval of the environmental pollution assessment index EPindex and conducting interval comparison, the environmental pollution degree of product manufacturing production is evaluated; By standardizing the resource scheduling parameters, the standardized parameter values of raw material information, personnel information, and equipment information are obtained. Furthermore, the resource utilization assessment index RUindex is comprehensively obtained. By setting the assessment interval of the resource utilization assessment index RUindex and conducting interval comparison, the resource utilization degree of product manufacturing production is evaluated; By combining the energy consumption assessment index ECindex, the environmental pollution assessment index EPindex, and the resource utilization assessment index RUindex, the product manufacturing production quality coefficient PQ is comprehensively evaluated; By setting the assessment interval of the product manufacturing production quality coefficient PQ and conducting interval comparison to determine the risk level of the product manufacturing production quality, an abnormal warning signal of the corresponding level is output.

[0011] Furthermore, the specific process of the global process parameter optimization scheme is as follows: Through the data acquisition period Tx, N0 groups of data samples are obtained. Any group of the energy consumption assessment index, environmental pollution assessment index, and resource utilization assessment index in the samples are respectively marked as ECi, EPi, and RUi; Mark the product manufacturing production quality coefficient PQ as the true value , build a deep learning model for time series analysis. By deeply analyzing the sample data of product manufacturing production, the predicted value of the product manufacturing production quality coefficient is obtained : ; Furthermore, the predicted value of the product manufacturing production quality coefficient is obtained and the true value The sum of squared errors SSe between them: ; By minimizing SSe to obtain the parameters , , , ; Then, , where V is the sample data matrix and y is the target vector; Then calculate the SHAP values for model feature impact analysis: For any feature j, its SHAP value Φj is: ; where F is the set of all features, S is any feature subset of F, is the number of elements in subset S, is the predicted value of the model when only using feature subset S; Mark the SHAP values of the features of the energy consumption evaluation index ECi, the environmental pollution evaluation index EPi, and the resource utilization evaluation index RUi as Φ1, Φ2, and Φ3 respectively. Furthermore, the specific steps for standardizing the energy monitoring parameters are as follows: The energy monitoring parameters include electricity consumption information, water consumption information, and gas consumption information; Perform standardization processing on the electricity consumption information, water consumption information, and gas consumption information respectively to obtain the standardized parameter values Ψpw of the electricity consumption information, the standardized parameter values Ψwt of the water consumption information, and the standardized parameter values Ψgs of the gas consumption information; Furthermore, comprehensively obtain the energy consumption evaluation index ECindex to evaluate the energy consumption level of product manufacturing production.

[0012] Furthermore, the specific steps for standardizing the pollution monitoring parameters are as follows: The pollution monitoring parameters include solid waste information, gaseous waste information, liquid waste information, and noise information; Perform standardization processing on the solid waste information, gaseous waste information, liquid waste information, and noise information respectively to obtain the standardized parameter values Ψgf of the solid waste information, the standardized parameter values Ψqf of the gaseous waste information, the standardized parameter values Ψyf of the liquid waste information, and the standardized parameter values Ψzs of the noise information; Furthermore, comprehensively obtain the environmental pollution evaluation index EPindex to evaluate the environmental pollution level of product manufacturing production.

[0013] Furthermore, the specific steps for standardizing the resource scheduling parameters are as follows: The resource scheduling parameters include raw material information, personnel information, and equipment information; Perform standardization processing on the raw material information, personnel information, and equipment information respectively to obtain the standardized parameter values Ψcl of the raw material information, the standardized parameter values Ψry of the personnel information, and the standardized parameter values Ψsb of the equipment information; Furthermore, comprehensively obtain the resource utilization evaluation index RUindex to evaluate the resource utilization level of product manufacturing production.

[0014] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are: The present invention comprehensively constructs a product full - life - cycle data platform from multiple perspectives of energy monitoring, pollution monitoring, and resource scheduling, realizes the energy - efficiency evaluation of product manufacturing production quality, and through the data acquisition module, conducts multi - source data fusion, constructs a unified data platform, realizes the integration and sharing of multi - source heterogeneous data, and then combines the edge computing module and the cloud processor to realize cloud - edge collaborative computing, ensuring real - time data processing and analysis, improving efficiency and real - time performance. Furthermore, through the management control module and the human - machine interaction module, it optimizes global process parameters and conducts human - machine collaboration early - warning processing, realizes the simple and easy - to - use effect of human - machine interaction, and improves the intelligent application level of the product production management system, ultimately realizing the optimized transformation of product manufacturing industry towards energy conservation, greenness, pollution reduction, and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The connection schematic diagram of the system modules of the present invention is shown; Figure 2 The step schematic diagram of the method flow of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0017] Embodiment 1: As Figure 1 - Figure 2 shown, the intelligent manufacturing production quality control system based on big data technology includes a data acquisition module, an edge computing module, a cloud processor, a management control module, and a human - machine interaction module. Among them, the data acquisition module, the edge computing module, the cloud processor, the management control module, and the human - machine interaction module are communicatively connected; The working steps are as follows: S1, the data acquisition module collects production cycle data: The production cycle data includes energy monitoring parameters, pollution monitoring parameters, and resource scheduling parameters; Set the data acquisition cycle Tx to collect the production cycle data, collect the production cycle data through sensors, SCADA systems, and MES systems, and align the timestamps of the production cycle data. The specific parameters are as follows: The SCADA system (Supervisory Control And Data Acquisition) is a data acquisition and monitoring control system, mainly used for monitoring and controlling factory processes; The MES system (Manufacturing Execution System) is a production information management system for the workshop execution layer of manufacturing enterprises; Energy monitoring parameters include electricity consumption information, water consumption information, and gas consumption information; Electricity consumption information includes total power consumption, power consumption of each device, power, voltage, and current; Water consumption information includes total water consumption, water consumption of each process, water pressure, and water flow rate; Gas consumption information includes total gas consumption, gas consumption of each device, gas pressure, and gas flow rate; Pollution monitoring parameters include solid waste information, gaseous waste information, liquid waste information, and noise information; Solid waste information includes the amount of solid waste generated and the amount of hazardous waste generated; Gaseous waste information includes waste gas emissions, SO2 concentration, NOx concentration, and dust concentration; Liquid waste information includes wastewater emissions, COD concentration, BOD concentration, ammonia nitrogen concentration, and total phosphorus concentration; Noise information includes noise intensity and noise frequency; Resource scheduling parameters include raw material information, personnel information, and equipment information; Raw material information includes raw material input and raw material utilization rate; Personnel information includes the number of personnel, personnel working hours, and personnel working hour utilization rate; Equipment information includes equipment operation time, equipment utilization rate, and equipment failure rate.

[0018] S2. The edge computing module standardizes the production cycle data: Through energy monitoring parameters, pollution monitoring parameters, and resource scheduling parameters, it evaluates the energy consumption level, environmental pollution level, and resource utilization level of product manufacturing production in turn, and then comprehensively evaluates the quality of product manufacturing production, and outputs corresponding abnormal warning signals; The specific process of standardization processing is as follows: S2-1. Estimate and fill in missing values: ; Among them, mark the data value corresponding to any time node as , then is the previous time node, and the corresponding data value is , is the next time node, and the corresponding data value is ; S2-2. Screen and process outliers: If , then determine Xi as an outlier; Among them, aveX is the parameter average value, and stdX is the parameter standard deviation; S2-3, determine the linear correlation between variables: Pearson correlation coefficient γxy: ; Among them, is the population covariance of the two variables, and stdX and stdY are the standard deviations of variable X and variable Y respectively; γxy is positively correlated with and negatively correlated with stdX and stdY; the value range of the Pearson correlation coefficient γxy is [-1, 1]; Set the evaluation interval of the population Pearson correlation coefficient γxy, and evaluate the correlation between variable X and variable Y through interval comparison, so as to determine and perform regression analysis; S2-4, output the positive correlation variable set Qa and the negative correlation variable set Qb; When γxy is closer to 1, it indicates that the variables are positively correlated. When γxy is equal to 1, it represents a perfect positive correlation, that is, when one variable increases, the other variable also increases linearly; When γxy is closer to -1, it indicates that the variables are negatively correlated. When γxy is equal to -1, it represents a perfect negative correlation, that is, when one variable increases, the other variable decreases linearly; When γxy is closer to 0, it indicates that the correlation between the variables is weaker. When γxy is equal to 0, it represents no linear correlation; Mark any element in the positive correlation variable set Qa as variable Aq; Mark any element in the negative correlation variable set Qb as variable Bq; S2-5, assign the corresponding weight coefficient φq to variable q according to the correlation strength; When |γxy| ≥ 0.8, it indicates strong correlation; When 0.5 ≤ |γxy| < 0.8, it indicates medium correlation; When 0.3 ≤ |γxy| < 0.5, it indicates weak correlation; When |γxy| < 0.3, it indicates extremely weak correlation or no correlation; Then, mark the weight coefficient of the positive correlation variable Aq as φaq, and mark the weight coefficient of the negative correlation variable Bq as φbq; when |γxy| is higher, the preset value of the weight coefficient φq of variable q is higher; S2-6, comprehensively obtain the standardized parameter value Ψ through n1 positive correlation variables Aq, n2 negative correlation variables Bq and their corresponding weight coefficients: 。

[0019] The specific steps for standardizing production cycle data and evaluating the energy consumption level, environmental pollution level, and resource utilization level of product manufacturing production are as follows: By standardizing the energy monitoring parameters, obtain the standardized parameter values of electricity consumption information, water consumption information, and gas consumption information, and then comprehensively obtain the energy consumption evaluation index ECindex. By setting the evaluation interval of the energy consumption evaluation index ECinde and conducting interval comparison, evaluate the energy consumption level of product manufacturing production; The specific steps for standardizing the energy monitoring parameters are as follows: The energy monitoring parameters include electricity consumption information, water consumption information, and gas consumption information; Standardize the electricity consumption information. Analyze the total electricity consumption, electricity consumption of each device, power, voltage, and current as variables to obtain the standardized parameter value Ψpw of the electricity consumption information; Standardize the water consumption information. Analyze the total water consumption, water consumption of each process, water pressure, and water flow as variables to obtain the standardized parameter value Ψwt of the water consumption information; Standardize the gas consumption information. Analyze the total gas consumption, gas consumption of each device, gas pressure, and gas flow as variables to obtain the standardized parameter value Ψgs of the gas consumption information; Standardize the energy monitoring parameters. Combine the standardized parameter values of electricity consumption information, water consumption information, and gas consumption information to obtain the energy consumption evaluation index ECindex: ; Among them, e1, e2, and e3 are the weight coefficients of the standardized parameter value Ψpw of the electricity consumption information, the standardized parameter value Ψwt of the water consumption information, and the standardized parameter value Ψgs of the gas consumption information respectively. The weight coefficients are preset after being calculated through a large amount of experimental data, and e1, e2, and e3 are all greater than 0; Set the evaluation interval of the energy consumption evaluation index ECinde, and evaluate the energy consumption level of product manufacturing production through interval comparison.

[0020] By standardizing the pollution monitoring parameters, obtain the standardized parameter values of solid waste information, gaseous waste information, liquid waste information, and noise information, and then comprehensively obtain the environmental pollution evaluation index EPindex. By setting the evaluation interval of the environmental pollution evaluation index EPindex and conducting interval comparison, evaluate the environmental pollution level of product manufacturing production; The specific steps for standardizing the pollution monitoring parameters are as follows: The pollution monitoring parameters include solid waste information, gaseous waste information, liquid waste information, and noise information; Standardize the solid waste information, analyze the solid waste generation amount and hazardous waste generation amount as variables, and obtain the standardized parameter value Ψgf of the solid waste information; Standardize the gaseous waste information, analyze the waste gas emission amount, SO2 concentration, NOx concentration, and dust concentration as variables, and obtain the standardized parameter value Ψqf of the gaseous waste information; Standardize the liquid waste information, analyze the wastewater emission amount, COD concentration, BOD concentration, ammonia nitrogen concentration, and total phosphorus concentration as variables, and obtain the standardized parameter value Ψyf of the liquid waste information; Standardize the noise information, analyze the noise intensity and noise frequency as variables, and obtain the standardized parameter value Ψzs of the noise information; Standardize the pollution monitoring parameters, combine the standardized parameter values of the solid waste information, gaseous waste information, liquid waste information, and noise information, and obtain the environmental pollution assessment index EPindex: ; Among them, f1, f2, f3, and f4 are the weight coefficients of the standardized parameter value Ψgf of the solid waste information, the standardized parameter value Ψqf of the gaseous waste information, the standardized parameter value Ψyf of the liquid waste information, and the standardized parameter value Ψzs of the noise information respectively, and f1, f2, f3, and f4 are all greater than 0; Set the evaluation interval of the environmental pollution assessment index EPindex, and evaluate the environmental pollution degree of product manufacturing production through interval comparison.

[0021] By standardizing the resource scheduling parameters, obtain the standardized parameter values of the raw material information, personnel information, and equipment information, and then comprehensively obtain the resource utilization evaluation index RUindex. By setting the evaluation interval of the resource utilization evaluation index RUindex and performing interval comparison, evaluate the resource utilization degree of product manufacturing production; The specific steps for standardizing the resource scheduling parameters are as follows: The resource scheduling parameters include raw material information, personnel information, and equipment information; Standardize the raw material information, analyze the raw material input amount and raw material utilization rate as variables, and obtain the standardized parameter value Ψcl of the raw material information; Standardize the personnel information, analyze the number of personnel, personnel working hours, and personnel working hour utilization rate as variables, and obtain the standardized parameter value Ψry of the personnel information; Standardize the equipment information, analyze the equipment operation time, equipment utilization rate, and equipment failure rate as variables, and obtain the standardized parameter value Ψsb of the equipment information; Standardize the resource scheduling parameters, combine the standardized parameter values of raw material information, personnel information, and equipment information, and obtain the resource utilization evaluation index RUindex: ; Among them, g1, g2, and g3 are the weight coefficients of the standardized parameter values Ψcl of raw material information, Ψry of personnel information, and Ψsb of equipment information respectively, and g1, g2, and g3 are all greater than 0; Set the evaluation interval of the resource utilization evaluation index RUindex, and evaluate the resource utilization degree of product manufacturing production through interval comparison.

[0022] Comprehensively evaluate the product manufacturing production quality coefficient PQ by combining the energy consumption evaluation index ECindex, environmental pollution evaluation index EPindex, and resource utilization evaluation index RUindex: ; Among them, , , are the weight factors of the energy consumption evaluation index ECindex, environmental pollution evaluation index EPindex, and resource utilization evaluation index RUindex respectively, and , , are all greater than 0; when the energy consumption evaluation index ECindex and environmental pollution evaluation index EPindex are lower, and the resource utilization evaluation index RUindex is higher, the product manufacturing production quality coefficient PQ is higher, and the evaluation of product manufacturing production quality is better; By setting the evaluation interval of the product manufacturing production quality coefficient PQ and performing interval comparison to determine the risk level of product manufacturing production quality, and then output an abnormal warning signal of the corresponding level.

[0023] S3. The cloud processor performs in-depth analysis and optimization on the production cycle data: By in-depth analyzing the energy consumption degree, environmental pollution degree, and resource utilization degree of product manufacturing production, obtain a global process parameter optimization plan, thereby constructing a product full-life cycle data chain to achieve product quality traceability and prompt; The specific process of the global process parameter optimization plan is: Obtain N0 groups of data samples through the data collection cycle Tx, and mark any group of energy consumption evaluation index, environmental pollution evaluation index, and resource utilization evaluation index in the samples as ECi, EPi, and RUi respectively; Mark the product manufacturing production quality coefficient PQ as the true value , build a deep learning model for time series analysis. By deeply analyzing the sample data of product manufacturing production, obtain the predicted value of the product manufacturing production quality coefficient : ; Furthermore, obtain the predicted value of the product manufacturing production quality coefficient and the true value The sum of squared errors SSe between them is: ; Obtain the parameters by minimizing SSe , , , ; Then, , where V is the sample data matrix and y is the target vector; Then calculate the SHAP value for model feature impact analysis: For any feature j, its SHAP value Φj is: ; Among them, F is the set of all features, S is any feature subset of F, is the number of elements in the subset S, is the predicted value of the model when only using the feature subset S; Mark the SHAP values of the features of the energy consumption evaluation index ECi, the environmental pollution evaluation index EPi, and the resource utilization evaluation index RUi as Φ1, Φ2, and Φ3 respectively.

[0024] S4, the management control module receives the global process parameter optimization plan and generates corresponding management control instructions to achieve the adaptive adjustment and optimization of the system; Thus, analyze the influence of each feature on the model prediction, and thereby obtain the global process parameter optimization plan. Among them, the global process parameter optimization plan is designed by professional and technical personnel based on the influence of each feature. Specifically, it is necessary to adjust and optimize the local parameters and conduct targeted design in combination with the actual product manufacturing situation, so as to build a product full-life cycle data chain, achieve product quality traceability and prompt, and ensure the simplicity and usability of human-computer interaction.

[0025] S5, the human-computer interaction module receives the abnormal warning signal and performs corresponding warning prompt operations to prompt the management personnel to perform corresponding processing; For example, for abnormal aspects such as energy monitoring, pollution monitoring, and resource scheduling, corresponding processing is carried out on energy management, pollution control, or resource allocation in the product production process, and it is refined and positioned to factors such as electricity consumption, water consumption, and gas consumption of energy, solid waste, liquid waste, gas waste, noise, etc. of pollution, and raw materials, personnel allocation, equipment operation, etc. of resources for refined processing, so as to improve production management efficiency and optimize product production quality.

[0026] In summary, the present invention comprehensively constructs a product full - life - cycle data platform from multiple perspectives of energy monitoring, pollution monitoring, and resource scheduling, realizes the energy - efficiency evaluation of product manufacturing production quality, and through the data acquisition module, conducts multi - source data fusion, constructs a unified data platform, realizes the integration and sharing of multi - source heterogeneous data, and then combines the edge computing module and the cloud processor to realize cloud - edge collaborative computing, ensuring real - time data processing and analysis, improving efficiency and real - time performance. Furthermore, through the management control module and the human - machine interaction module, global process parameter optimization and human - machine collaboration early - warning processing are carried out to achieve the simple and easy - to - use effect of human - machine interaction, and improve the intelligent application level of the product production management system, and finally realize the optimized transformation of energy conservation, greenness, pollution reduction, and sustainable development in the product manufacturing industry.

[0027] The setting of the size of the interval and threshold is for the convenience of comparison. Regarding the size of the threshold, it depends on the amount of sample data and the number of base numbers set by those skilled in the art for each group of sample data; as long as the proportional relationship between the parameter and the quantified value is not affected.

[0028] The above - mentioned formulas are all dimensionless and take their numerical calculations. The formula is a formula obtained by software simulation of collecting a large amount of data to approximate the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation; As described above, only the preferred specific embodiments of the present invention are given, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. Intelligent manufacturing production quality control system based on big data technology, characterized by: It includes a data acquisition module, an edge computing module, a cloud processor, a management control module and a human-computer interaction module, wherein the data acquisition module, the edge computing module, the cloud processor, the management control module and the human-computer interaction module are communicatively connected; The data acquisition module is used to collect production cycle data: production cycle data includes energy monitoring parameters, pollution monitoring parameters and resource scheduling parameters; The edge computing module is used to standardize the production cycle data: through energy monitoring parameters, pollution monitoring parameters and resource scheduling parameters, the energy consumption, environmental pollution and resource utilization of product manufacturing production are evaluated in turn, and then the product manufacturing production quality is comprehensively evaluated, and the corresponding abnormal warning signals are output; The cloud processor is used to conduct in-depth analysis and optimization of production cycle data: by deeply analyzing the energy consumption, environmental pollution and resource utilization of product manufacturing and production, the global process parameter optimization plan is obtained, thereby building a product life cycle data chain and achieving product quality traceability and prompts; The management and control module is used to receive the global process parameter optimization plan and generate corresponding management and control instructions to achieve adaptive adjustment and optimization of the system; The human-computer interaction module is used to receive abnormal warning signals and perform corresponding warning prompt operations, thereby prompting management personnel to take corresponding measures.

2. The intelligent manufacturing production quality control system based on big data technology according to claim 1 is characterized by: The specific process of standardization is as follows: S2-1, estimate and fill missing values: ; Among them, any time node The corresponding data values ​​are marked as ,but is the previous time node, and the corresponding data value is , is the next time node, and the corresponding data value is ; S2-2, filter outliers: like , then Xi is determined to be an outlier; Among them, aveX is the parameter mean, stdX is the parameter standard deviation; S2-3, determine the linear correlation between variables: Pearson correlation coefficient γxy: ; in, is the population covariance of the two variables, stdX and stdY are the standard deviations of variables X and Y respectively; γxy and Positively correlated, negatively correlated with stdX and stdY; the value range of Pearson correlation coefficient γxy is [-1,1]; Set the evaluation interval of the overall Pearson correlation coefficient γxy, and evaluate the correlation between variables X and Y through interval comparison to determine and perform regression analysis; S2-4, output the positively correlated variable set Qa and the negatively correlated variable set Qb; Mark any element of the positively correlated variable set Qa as variable Aq; Mark any element of the negatively correlated variable set Qb as variable Bq; S2-5, assign corresponding weight coefficient φq to variable q according to the strength of correlation; The weight coefficient of the positively correlated variable Aq is marked as φaq, and the weight coefficient of the negatively correlated variable Bq is marked as φbq; S2-6, the standardized parameter value Ψ is obtained comprehensively through n1 positively correlated variables Aq, n2 negatively correlated variables Bq and their corresponding weight coefficients.

3. The intelligent manufacturing production quality control system based on big data technology according to claim 2 is characterized by: Set the data collection cycle Tx to collect production cycle data, collect production cycle data through sensors, SCADA system and MES system, and align the timestamps of production cycle data. The specific parameters are as follows: Energy monitoring parameters include electricity usage information, water usage information, and gas usage information; Power consumption information includes total power consumption, power consumption of each device, power, voltage and current; Water use information includes total water consumption, water consumption of each process, water pressure and water flow; Gas consumption information includes total gas consumption, gas consumption of each device, gas pressure and gas flow; Pollution monitoring parameters include solid waste information, gas waste information, liquid waste information and noise information; Solid waste information includes the amount of solid waste generated and the amount of hazardous waste generated; Gas and waste information includes waste gas emissions, SO2 concentration, NOx concentration and dust concentration; Liquid waste information includes wastewater discharge volume, COD concentration, BOD concentration, ammonia nitrogen concentration, and total phosphorus concentration; Noise information includes noise intensity and noise frequency; Resource scheduling parameters include raw material information, personnel information, and equipment information; Raw material information includes raw material input and raw material utilization rate; Personnel information includes the number of personnel, personnel hours and personnel hour utilization rate; Equipment information includes equipment operating time, equipment utilization and equipment failure rate.

4. The intelligent manufacturing production quality control system based on big data technology according to claim 3 is characterized by: The specific steps for standardizing production cycle data and evaluating the energy consumption, environmental pollution and resource utilization of product manufacturing are as follows: By standardizing the energy monitoring parameters, the standardized parameter values ​​of electricity, water and gas information are obtained, and then the energy consumption evaluation index ECindex is comprehensively obtained. By setting the evaluation interval of the energy consumption evaluation index ECindex and performing interval comparison, the energy consumption level of product manufacturing and production is evaluated; By standardizing the pollution monitoring parameters, the standardized parameter values ​​of solid waste information, gas waste information, liquid waste information and noise information are obtained, and then the environmental pollution assessment index EPindex is comprehensively obtained. By setting the evaluation interval of the environmental pollution assessment index EPindex and performing interval comparison, the environmental pollution degree of product manufacturing and production is evaluated; By standardizing resource scheduling parameters, standardized parameter values ​​of raw material information, personnel information and equipment information are obtained, and then the resource utilization evaluation index RUindex is comprehensively obtained. By setting the evaluation interval of the resource utilization evaluation index RUindex and performing interval comparison, the resource utilization degree of product manufacturing production is evaluated; By combining the energy consumption assessment index ECindex, the environmental pollution assessment index EPindex and the resource utilization assessment index RUindex, the product manufacturing production quality coefficient PQ is comprehensively assessed; By setting the evaluation interval of the product manufacturing quality coefficient PQ and comparing the intervals to determine the risk level of the product manufacturing quality, an abnormal warning signal of the corresponding level can be output.

5. The intelligent manufacturing production quality control system based on big data technology according to claim 4 is characterized by: The specific process of the global process parameter optimization solution is as follows: Obtain N0 groups of data samples through the data collection period Tx, and mark any group of energy consumption assessment index, environmental pollution assessment index and resource utilization assessment index in the samples as ECi, EPi and RUi respectively; Mark the product manufacturing quality factor PQ as the true value , build a deep learning model for time series analysis, and obtain the predicted value of the product manufacturing quality coefficient by deeply analyzing the sample data of product manufacturing production : ; Then obtain the predicted value of the product manufacturing production quality coefficient With the true value The sum of squared errors SSe between them is: ; Obtain parameters by minimizing SSe , , , ; but, , where V is the sample data matrix and y is the target vector; Then calculate the SHAP value to analyze the impact of model features: For any feature j, its SHAP value Φj is: ； Among them, F is the set of all features, S is any feature subset of F, is the number of elements in the subset S, is the predicted value of the model when only the feature subset S is used; The characteristic SHAP values ​​of the energy consumption assessment index ECi, the environmental pollution assessment index EPi and the resource utilization assessment index RUi are marked as Φ1, Φ2 and Φ3 respectively.

6. The intelligent manufacturing production quality control system based on big data technology according to claim 4 is characterized by: The specific steps for standardizing energy monitoring parameters are as follows: Energy monitoring parameters include electricity usage information, water usage information, and gas usage information; The electricity consumption information, the water consumption information and the gas consumption information are respectively standardized to obtain a standardized parameter value Ψpw of the electricity consumption information, a standardized parameter value Ψwt of the water consumption information and a standardized parameter value Ψgs of the gas consumption information; Then, the energy consumption assessment index ECindex is comprehensively obtained to evaluate the energy consumption level of product manufacturing and production.

7. The intelligent manufacturing production quality control system based on big data technology according to claim 4 is characterized by: The specific steps for standardizing pollution monitoring parameters are as follows: Pollution monitoring parameters include solid waste information, gas waste information, liquid waste information and noise information; The solid waste information, gas waste information, liquid waste information and noise information are standardized respectively to obtain the standardized parameter value Ψgf of the solid waste information, the standardized parameter value Ψqf of the gas waste information, the standardized parameter value Ψyf of the liquid waste information and the standardized parameter value Ψzs of the noise information; Then, the environmental pollution assessment index EPindex is comprehensively obtained to evaluate the degree of environmental pollution caused by product manufacturing and production.

8. The intelligent manufacturing production quality control system based on big data technology according to claim 4 is characterized by: The specific steps for standardizing resource scheduling parameters are as follows: Resource scheduling parameters include raw material information, personnel information, and equipment information; Standardize the raw material information, personnel information and equipment information respectively to obtain the standardized parameter value Ψcl of the raw material information, the standardized parameter value Ψry of the personnel information and the standardized parameter value Ψsb of the equipment information; Then, the resource utilization evaluation index RUindex is comprehensively obtained to evaluate the resource utilization degree of product manufacturing and production.

Images