Dynamic scheduling system for compound additive production process

Abstract

The invention discloses a compound additive production process dynamic scheduling system, which belongs to the technical field of data processing, and comprises an acquisition and evaluation module for acquiring original data in a compound additive generation process in real time and generating a reaction stage evaluation value; the collaborative optimization module is used for receiving the response stage evaluation value and outputting an adjustment instruction set; the quality rehearsal module is used for receiving the adjustment instruction draft and generating a virtual working condition parameter set; the scheduling generation module is used for generating a first process execution sequence and a second process execution sequence according to the virtual working condition parameter set and the adjustment confidence factor; and the feedback module is used for collecting actual quality data produced by the split charging equipment, comparing the actual quality data with the virtual working condition parameter set, calculating a deviation value, generating an optimization instruction based on the deviation value, and feeding back the optimization instruction to the acquisition and evaluation module and the collaborative optimization module. According to the invention, a future production process is simulated in a virtual environment through the quality rehearsal module, and the production stability and the product consistency are enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a dynamic scheduling system for production process of compounded additives. BACKGROUND

[0002] The compounded additive refers to a composite industrial additive prepared by mixing two or more functional components (such as thickening agent, emulsifier, preservative, etc.) in a specific proportion through chemical reaction or physical modification, which is widely used in food, cosmetics, polymer materials and other fields. Its production process involves key processes such as phase conversion in reaction kettle, temperature and pressure control and material mixing. The physicochemical properties (such as stability, viscosity, activity) of the final product directly affect the application effect of the downstream.

[0003] In the prior art, the scheduling of the production process of the compounded additive is mostly based on the static control method based on fixed proportion and empirical threshold. Its working principle relies on the preset process parameters (such as temperature range, reaction time) and the manually set feeding ratio. The reaction kettle is controlled in stages through PLC or DCS system, and the product quality is verified offline by periodic sampling detection. The whole scheduling process lacks dynamic correlation analysis of real-time process data and historical production batch quality.

[0004] The prior art has the following defects: it cannot capture the influence of the dynamic fluctuations of temperature, pressure and material flow rate in the production process on the reaction progress in real time, resulting in that the scheduling decision lags behind the actual working condition change. Therefore, it is urgent to provide a dynamic scheduling system for production process of compounded additives to solve the above problems. SUMMARY

[0005] The technical problem to be solved by the present application is to overcome the defects of the prior art that it cannot capture the influence of the dynamic fluctuations of temperature, pressure and material flow rate in the production process on the reaction progress in real time, resulting in that the scheduling decision lags behind the actual working condition change. A dynamic scheduling system for production process of compounded additives is provided.

[0006] To solve the above technical problems, one technical solution adopted by the present application is to provide a dynamic scheduling system for production process of compounded additives, comprising a collection and evaluation module, a collaborative optimization module, a quality pre-play module, a scheduling generation module and a feedback module. The collection and evaluation module is deployed near the reaction kettle and the dispensing equipment used for producing the compounded additive, and is used for collecting the original data in the production process of the compounded additive in real time, and generating a reaction stage evaluation value in real time based on the current original data. The collaborative optimization module receives the reaction stage evaluation value, analyzes the historical batch quality data stored in the preset cloud, and outputs an adjustment instruction set, which includes an adjustment confidence factor and an adjustment instruction draft. a quality preview module configured to receive the adjustment instruction draft, simulate a production process of the compounded additive after injection of the adjustment instruction draft in a preset virtual environment for a future time period, and generate a virtual working condition parameter set; a scheduling generation module configured to generate a first execution sequence of a process when the adjustment confidence factor does not exceed a preset threshold value and generate a second execution sequence of the process when the adjustment confidence factor exceeds the preset threshold value according to the virtual working condition parameter set and the adjustment confidence factor; a feedback module configured to collect actual quality data output by the dispensing equipment, compare the actual quality data with the virtual working condition parameter set, calculate a deviation amount, generate an optimization instruction based on the deviation amount, and feed back the optimization instruction to the collection and evaluation module and the collaborative optimization module.

[0007] The application further provides that the original data in the collection and evaluation module include temperature, pressure, and material flow rate. The original data are collected in real time by temperature sensors, pressure sensors, and flow sensors arranged near the reaction kettle and the dispensing equipment.

[0008] The application further provides that the generation step of the reaction stage evaluation value in the collection and evaluation module is: S1, based on the real-time collected original data, calculating dynamic change trends of temperature, pressure, and material flow rate in an adjacent preset time window, inputting the dynamic change trends into a preset material reaction knowledge base for matching, and identifying a deviation degree of a reaction process of the compounded additive from a standard phase transition mode in the material reaction knowledge base; S2, based on the deviation degree, combining a preset weight coefficient, and calculating a reaction stage evaluation value.

[0009] The application further provides that the historical batch quality data in the collaborative optimization module include actual addition proportion data of each component of the compounded additive in a historical production batch, time sequence record data of key process parameters in the historical production batch, physical and chemical quality detection data of a final product of the historical production batch, equipment operation state record data of the historical production batch, and a historical optimal batch target value. The generation step of the adjustment instruction set in the collaborative optimization module is: Q1, based on the reaction stage evaluation value and the time sequence record data of the key process parameters in the historical batch quality data, identifying a deviation of a reaction process of the compounded additive from the historical optimal batch target value, combining physical and chemical quality detection data of a final product of the historical production batch, determining an adjustment item of the key process parameter, and generating a preliminary adjustment strategy; Q2, based on the preliminary adjustment strategy, combining the actual addition ratio data in the historical batch quality data and the device running state record data, calculating the adjustment amplitude and adjustment priority of each key process parameter, and correcting the adjustment amplitude to generate an adjustment instruction draft; Q3, based on the adjustment instruction draft, generating an adjustment confidence factor, integrating the adjustment instruction draft and the adjustment confidence factor into an adjustment instruction set, and outputting to a quality rehearsal module; The generation method of the adjustment confidence factor is: Q301, based on the adjustment instruction draft, calling the historical adjustment strategy record in the historical batch quality data and the physical and chemical quality detection data of the corresponding batch, identifying the success rate of similar adjustment instruction drafts in historical execution, and calculating a preliminary confidence score; Q302, based on the preliminary confidence score, combining the temperature, pressure and material flow rate dynamic change trend in the real-time collected original data, analyzing the deviation of the current production environment of the compounded additive from the successful environment in the historical execution, converting the deviation into a correction coefficient, and integrating the preliminary confidence score and the correction coefficient through a weighted fusion algorithm to generate an adjustment confidence factor.

[0010] The application further provides that: the calculation steps of the adjustment amplitude and the adjustment priority of the key process parameters in step Q2 are as follows: Q201, based on the preliminary adjustment strategy, extracting the deviation amount of the current value of the key process parameter and the target value of the historical optimal batch, combining the actual addition ratio data in the historical batch quality data, multiplying the deviation amount and the addition ratio, and multiplying the result by a preset distribution weight to calculate the initial adjustment amplitude of each key process parameter; Q202, combining the device state stability coefficient and the device real-time running feedback data generated by the device running state record data in the historical batch quality data, weighting and fusing the device state stability coefficient and the initial adjustment amplitude to generate the adjustment priority of each key process parameter; Q203, based on the initial adjustment amplitude and the adjustment priority, introducing a preset historical adjustment success rate factor and the device real-time running feedback data to correct the initial adjustment amplitude, generate the corrected adjustment amplitude and adjustment priority, and use them as the adjustment instruction draft.

[0011] The application further provides that: the generation method of the virtual working condition parameter set in the quality rehearsal module is as follows: W1, based on the adjustment instruction draft, injecting the adjustment instruction draft into the preset virtual production environment through a preset virtual simulation engine, simulating the production process of the compounded additive in a future set period, and generating an initial simulation data sequence containing time sequence change data of simulated temperature, pressure and material flow rate; W2, based on the initial simulation data sequence, combining the equipment operation state record data in the historical batch quality data, analyzing the change trend of the key parameters in the simulated compounded additive production process through a dynamic simulation algorithm preloaded in the preset virtual production environment, generating a product quality evolution path of the compounded additive according to the change trend of the key parameters, and generating an intermediate working condition parameter set according to the product quality evolution path; W3, combining the physical and chemical quality detection data of the final product in the historical batch quality data, performing risk assessment on the intermediate working condition parameter set, identifying quality risk points, adding a preset risk identifier to each quality risk point, and integrating the intermediate working condition parameter set with the risk identifier into a virtual working condition parameter set.

[0012] The application further provides that the product quality evolution path generation step in step W2 is: W201, based on the initial simulation data sequence, extracting time sequence change data of simulated temperature, pressure and material flow rate, identifying dynamic deviation characteristics of the key process parameters, and comparing with the historical optimal batch target value to generate a parameter deviation map; W202, based on the parameter deviation map, combining the physical and chemical quality detection data in the historical batch quality data, identifying potential quality risk points in the high deviation area of the parameter deviation map, and generating a quality risk area distribution map; The identification method of the potential risk points in step W202 is: W2021, extracting the area with a deviation value exceeding a preset risk threshold value in the parameter deviation map as a high deviation area; W2022, combining the physical and chemical quality detection data in the historical batch quality data, calculating the matching degree of the deviation value of each high deviation area with the quality abnormal event in the physical and chemical quality detection data, and generating a potential risk score of each high deviation area according to the matching degree; W2023, when the potential risk score exceeds a preset action threshold value, marking the high deviation area as a potential quality risk point; W203, based on the quality risk area distribution map, simulating the continuous evolution trajectory of the quality of the compounded additive in the production process, adjusting the path smoothness of the continuous evolution trajectory combined with the equipment operation state record data, and generating a product quality evolution path.

[0013] The application is further provided that: the generation step of the first process execution sequence in the scheduling generation module is: A1, extract the risk identifier and the adjustment confidence factor carried in the virtual working condition parameter set, analyze the matching degree of the risk identifier level and the adjustment confidence factor, calculate the safety weight and execution urgency of each process step of the compounded additive, and generate a preliminary process sequence draft; A2, based on the preliminary process sequence draft, combining the simulation temperature, pressure and material flow rate time sequence data in the virtual working condition parameter set, a first process execution sequence is generated.

[0014] The application is further provided that: the generation step of the second process execution sequence in the scheduling generation module is: B1, based on the adjustment confidence factor exceeding the preset threshold, triggering the preset real-time rescheduling instruction, calling the risk identifier in the virtual working condition parameter set and the latest values of temperature, pressure and material flow rate in the original data collected in real time, generating a real-time production state snapshot; B2, based on the real-time production state snapshot, combining the adjustment amplitude and the adjustment priority in the adjustment instruction draft, re-evaluating the execution order and resource allocation of each process of the compounded additive production, calculating the optimized process logic connection and parameter adjustment value, and generating a rescheduling scheme; B3, based on the rescheduling scheme, integrating the real-time equipment health index of the equipment running state record data in the historical batch quality data, verifying the feasibility and safety of the rescheduling scheme, and generating a second process execution sequence.

[0015] The application is further provided that: the specific content of the feedback module is: the feedback module collects actual quality data output by the split device in real time, compares the actual quality data with the virtual working condition parameter set generated by the quality rehearsal module item by item, calculates the data deviation between the actual quality data and the virtual working condition parameter set, generates an optimization instruction based on the size and direction of the data deviation, and feeds back the optimization instruction to the collection evaluation module and the collaborative optimization module through the preset communication protocol, adjusts the data collection strategy of the collection evaluation module and the historical data analysis parameters of the collaborative optimization module.

[0016] The beneficial effects of the application are as follows: 1. This invention uses a data acquisition and evaluation module to monitor the dynamic fluctuations of temperature, pressure, and material flow rate in real time, and combines this with a collaborative optimization module to analyze historical batch data and generate adjustment instruction sets. This enables the system to capture subtle changes in the reaction process during production in real time, effectively solving the scheduling decision delay problem caused by data lag in traditional methods, and significantly improving response speed and accuracy. 2. This invention utilizes a quality simulation module to simulate future production processes in a virtual environment, generating a high-precision set of virtual operating condition parameters. Through a feedback module, it compares actual data with simulation results in real time, forming a closed-loop optimization mechanism to ensure that the system can adaptively adjust the process sequence. This overcomes the shortcomings of traditional technologies that cannot dynamically adapt to changes in operating conditions, thereby enhancing production stability and product consistency. Attached Figure Description

[0017] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a flowchart illustrating the steps for generating the adjustment instruction set according to the present invention. Figure 3 This is a flowchart of the method for generating virtual operating condition parameter sets according to the present invention. Detailed Implementation

[0018] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0019] Please see Figure 1 - Figure 3 A dynamic scheduling system for the production process of compound additives includes a data acquisition and evaluation module, a collaborative optimization module, a quality pre-simulation module, a scheduling generation module, and a feedback module. The data acquisition and evaluation module is deployed near the reaction vessel and dispensing equipment used to produce compound additives. It is used to collect raw data in real time during the production process of compound additives and generate reaction stage evaluation values ​​in real time based on the current raw data. The collaborative optimization module receives the evaluation values ​​from the reaction phase, analyzes them in conjunction with the historical batch quality data stored in the cloud, and outputs an adjustment instruction set, which includes an adjustment confidence factor and a draft adjustment instruction. The quality simulation module receives the draft adjustment instructions and simulates the production process of compound additives after injecting the draft adjustment instructions in a preset virtual environment within a future set time period, generating a set of virtual working condition parameters. The scheduling generation module is used to generate the first process execution sequence based on the virtual working condition parameter set and the adjustment confidence factor: when the adjustment confidence factor does not exceed the preset threshold, the first process execution sequence is generated; when the adjustment confidence factor exceeds the preset threshold, the second process execution sequence is generated. The feedback module collects actual quality data output by the sub-packaging device, compares the actual quality data with the virtual working condition parameter set, calculates a deviation amount, generates an optimization instruction based on the deviation amount, and feeds back the optimization instruction to the collection and evaluation module and the collaborative optimization module.

[0020] The system has the beneficial effects that the collection and evaluation module monitors production data in real time and generates a reaction stage evaluation value, the collaborative optimization module outputs an adjustment instruction set in combination with historical data, the quality pre-play module simulates an optimized production process in a virtual environment to generate a virtual working condition parameter set, the scheduling generation module adaptively generates a process execution sequence according to an adjustment confidence factor, and the feedback module realizes closed-loop optimization through comparison of actual and virtual data, thereby significantly improving the accuracy, stability and adaptive control capability of the production of compounded additives.

[0021] In one embodiment of the present application, the original data in the collection and evaluation module includes temperature, pressure and material flow rate; The original data is collected in real time by temperature sensors, pressure sensors and flow sensors deployed near the reaction kettle and the sub-packaging device, and is converted into a unified format signal by a PLC built in the device and transmitted to the collection and evaluation module.

[0022] Preferably, the generation step of the reaction stage evaluation value in the collection and evaluation module is as follows: S1, based on the real-time collected original data, the dynamic change trend of temperature, pressure and material flow rate in the adjacent preset time window is calculated, the dynamic change trend is input into the preset material reaction knowledge base for matching, and the deviation degree of the current reaction process of the compounded additive from the standard phase transition mode in the material reaction knowledge base is identified; The calculation step of the dynamic change trend is as follows: based on the real-time collected temperature, pressure and material flow rate original data, time series data sequences in the adjacent preset time window are extracted by a sliding window analysis algorithm, the change rate of each parameter in the time window (such as temperature slope, pressure gradient and flow rate fluctuation amplitude) is calculated, and the change characteristics of multiple parameters are fused to generate a dynamic change trend vector, which represents the instantaneous fluctuation characteristics of the reaction process; The preset material reaction knowledge base is a database integrating historical production data and standard reaction models, which includes standard temperature ranges, pressure thresholds and material flow rate specifications of the compounded additive at different reaction stages, as well as critical point parameters and ideal reaction path sequences of phase transition, and is used to provide a reference for matching the dynamic change trend; The specific content of identifying the deviation degree of the reaction progress of the current compounded additive from the standard phase transition mode in the material reaction knowledge base is: based on the dynamic change trend vector and the standard phase transition mode in the material reaction knowledge base, the difference value of the current parameter sequence and the standard sequence is calculated through a multi-dimensional similarity matching algorithm, the comprehensive deviation degree of temperature, pressure and flow rate is quantified, and a deviation coefficient is generated, which reflects the deviation amplitude of the actual reaction progress from the ideal mode; S2, based on the deviation degree, a reaction stage evaluation value is calculated by combining a preset weight coefficient.

[0023] The preset weight coefficient is set according to the influence degree of different original data on the reaction progress of the compounded additive, for example, if the change of temperature is more critical to the reaction progress in a certain reaction stage, the weight coefficient corresponding to the temperature will be relatively high.

[0024] The calculation formula of the reaction stage evaluation value is: the temperature deviation coefficient, the pressure deviation coefficient and the material flow rate deviation coefficient are multiplied by the corresponding preset weight coefficient (the weight coefficient is dynamically allocated based on the influence intensity of each parameter on the current reaction stage), the initial evaluation value is obtained through weighted summation algorithm, and then the initial evaluation value is mapped to a normalized interval (such as 0 to 1) by introducing a normalization function, to generate the final reaction stage evaluation value for quantifying the health status of the reaction progress.

[0025] This embodiment can accurately quantify the deviation degree of the reaction progress by real-time acquisition of temperature, pressure and material flow rate data, combination of the material reaction knowledge base matching dynamic change trend, and generation of a standardized evaluation value based on the weight coefficient, which significantly improves the monitoring accuracy and real-time response ability of the compounded additive production process.

[0026] One of the embodiments of the present application is that the historical batch quality data in the cooperative optimization module includes the actual addition ratio data of each component of the compounded additive in the historical production batch, the time sequence record data of the key process parameters in the historical production batch, the physical and chemical quality detection data of the final product of the historical production batch, the equipment running state record data of the historical production batch and the historical optimal batch target value. The generation steps of the adjustment instruction set in the cooperative optimization module are: Q1, based on the reaction stage evaluation value and the time sequence record data of the key process parameters in the historical batch quality data, the deviation of the reaction progress of the current compounded additive from the historical optimal batch target value is identified, the adjustment item of the key process parameter is determined by combining the physical and chemical quality detection data of the final product of the historical production batch, and a preliminary adjustment strategy is generated. The specific steps for determining the adjustment items of key process parameters are as follows: Based on the reaction stage evaluation value and the time-series record data of key process parameters in the historical batch quality data, the deviation between the current key process parameter value and the historical best batch target value is calculated through a preset deviation identification algorithm. Combined with the quality anomaly event records in the physicochemical quality test data of the final products of historical production batches, the deviation is matched with the preset quality anomaly threshold. When the deviation exceeds the preset quality tolerance, the key process parameter is marked as an adjustment item and an adjustment item list is generated for the construction of subsequent preliminary adjustment strategies. Q2. Based on the initial adjustment strategy, combined with the actual dosage ratio data and equipment operation status record data from historical batch quality data, calculate the adjustment range and priority of each key process parameter, correct the adjustment range, and generate a draft adjustment instruction. The calculation steps for the adjustment range and priority of each key process parameter are as follows: Based on the list of adjustment items, the deviation of each key process parameter is extracted. Combined with the actual addition ratio data in the historical batch quality data, the deviation is multiplied by the addition ratio through a preset weighted calculation model (existing technology), and a preset allocation weight coefficient is introduced to calculate the initial adjustment range of each parameter. Subsequently, the equipment status stability coefficient generated by combining the equipment operation status record data is used to perform weighted fusion with the initial adjustment range through a priority fusion algorithm. Based on the fusion result, the urgency of each parameter adjustment is sorted to generate an adjustment priority sequence. The specific content of correcting the adjustment range is as follows: Based on the initial adjustment range and the adjustment priority sequence, a preset historical adjustment success rate factor (this factor comes from the success rate statistics of similar instructions in the historical adjustment strategy record) is introduced, and combined with the equipment health indicators in the real-time operation feedback data of the equipment, the initial adjustment range is calibrated. When the real-time operation feedback data of the equipment indicates performance fluctuation, the adjustment range is scaled according to the fluctuation ratio to ensure that the corrected adjustment range is adapted to the current production environment. Preferably, the calculation steps for the adjustment range and priority of the key process parameters in step Q2 are as follows: Q201. Based on the initial adjustment strategy, extract the deviation between the current value of the key process parameter and the historical best batch target value. Combine the actual addition ratio data in the historical batch quality data, multiply the deviation by the addition ratio and then multiply by the preset allocation weight to calculate the initial adjustment range of each key process parameter. The initial adjustment range of each key process parameter is calculated as follows: extract the deviation between the current value of the key process parameter and the historical best batch target value, multiply it by the actual addition ratio data in the historical batch quality data, and then multiply it by the preset allocation weight coefficient (the allocation weight coefficient is dynamically set based on the influence weight of the key process parameter on the overall quality). The calculation result is the initial adjustment range of each key process parameter, which quantifies the basic intensity of the parameter adjustment. Q202. Combine the equipment operation status record data in the historical batch quality data to generate the equipment status stability coefficient and the real-time operation feedback data of the equipment. Then, weight and integrate the equipment status stability coefficient with the initial adjustment range to generate the adjustment priority of each key process parameter. The specific content of generating the adjustment priority of each key process parameter by weighted fusion of the equipment stability coefficient and the initial adjustment range is as follows: Based on the initial adjustment range, the equipment stability coefficient (calculated by indicators such as historical equipment failure rate and real-time performance degradation rate) generated by combining the equipment operation status record data is multiplied by the initial adjustment range through a weighted fusion algorithm (existing technology, which will not be elaborated here), and a priority weight factor is introduced (the weight factor is set based on the degree of influence of key process parameters on production continuity). The adjustment range is sorted according to the product result to generate the adjustment priority. The higher priority corresponds to parameters with high stability requirements or large deviations. Q203. Based on the initial adjustment range and adjustment priority, a preset historical adjustment success rate factor and real-time equipment operation feedback data are introduced to correct the initial adjustment range, generate the corrected adjustment range and adjustment priority, and use it as a draft adjustment instruction. The preset historical adjustment success rate factor is a dynamic parameter calculated based on the execution results of similar adjustment instruction drafts in the historical adjustment strategy record and the corresponding batch's physicochemical quality test data. The specific details of correcting the initial adjustment range are as follows: The correction of the initial adjustment range is based on the historical adjustment success rate factor and the real-time operation feedback data of the equipment. The historical adjustment success rate factor is used as a scaling factor and multiplied by the initial adjustment range through the correction algorithm. At the same time, abnormal indicators (such as temperature fluctuations or pressure abnormalities) in the real-time operation feedback data of the equipment are introduced as compensation items. When the equipment feedback is abnormal, the adjustment range is increased to compensate for the risk. Finally, the corrected adjustment range is generated to ensure the robustness of the draft adjustment instruction. Q3. Generate adjustment confidence factors based on the draft adjustment instructions, integrate the draft adjustment instructions and adjustment confidence factors into an adjustment instruction set, and output it to the quality pre-simulation module; The method for generating the adjusted confidence factor is as follows: Q301. Based on the draft adjustment directive, call the historical adjustment strategy records in the historical batch quality data and the corresponding batch physicochemical quality test data to identify the success rate of similar draft adjustment directives in historical execution, calculate the preliminary confidence score, and the preliminary confidence score reflects the degree of matching between the draft adjustment directive and the historical success pattern. The formula for calculating the preliminary confidence score is as follows: Based on the draft adjustment instruction, the historical adjustment strategy records in the historical batch quality data are called up, the ratio of the number of successful times of similar drafts in historical execution to the total number of times is calculated, multiplied by the quality compliance rate of the corresponding batch in the physicochemical quality test data (the compliance rate is calculated by the matching degree between the test results and the standard value), and the result is normalized to the 0-1 range to generate the preliminary confidence score, which reflects the matching degree between the draft and the historical successful pattern; Q302. Based on the preliminary confidence score, combined with the dynamic trends of temperature, pressure and material flow rate in the real-time collected raw data, analyze the deviation between the current production environment of the compound additive and the successful environment in the past, convert the deviation into a correction coefficient, and integrate the preliminary confidence score and the correction coefficient through a weighted fusion algorithm to generate an adjusted confidence factor.

[0027] The specific steps to convert the deviation into a correction coefficient are as follows: Based on the initial confidence score, combined with the dynamic change trends of temperature, pressure and material flow rate in the real-time collected raw data, calculate the root mean square error between the current production environment parameters and the successful environment parameters in the past execution, map the error value to the preset exponential decay function to generate a correction coefficient, and the correction coefficient decreases as the error increases, which is used to adjust the initial confidence score to reflect the real-time environmental differences. By integrating the initial confidence score and the correction coefficient using a weighted fusion algorithm, the adjusted confidence factor is generated as follows: Adjusted Confidence Factor = Initial Confidence Score × +correction factor× ,in and The dynamic weights, summing to 1, are adaptively allocated using machine learning algorithms based on the stability of the real-time production environment and the reliability of historical data. This ultimately generates an adjustment confidence factor, which is used to quantify the real-time reliability of the adjustment instruction set. This embodiment dynamically generates adjustment instruction sets and confidence factors by integrating historical data and real-time feedback, which significantly improves the accuracy and adaptability of compound additive production scheduling, reduces reliance on human experience, ensures coordinated optimization of process parameter adjustments and equipment status and environmental changes, and enhances the system's robustness and decision-making efficiency in fluctuating production scenarios.

[0028] One embodiment of the present invention is as follows: the method for generating the virtual working condition parameter set in the quality pre-simulation module is as follows: W1. Based on the draft adjustment instructions, the draft adjustment instructions are injected into the preset virtual production environment through a preset virtual simulation engine to simulate the production process of compound additives within a future set time period and generate an initial simulation data sequence. The initial simulation data sequence includes time-series changes in simulated temperature, pressure and material flow rate. The default definition of the virtual simulation engine is: a software platform that integrates multiphysics modeling and real-time computing. It receives draft adjustment instructions as input, and through the built-in fluid dynamics and chemical reaction coupling algorithm, it simulates the interactive evolution of temperature, pressure and material flow rate in a virtual production environment, dynamically renders the continuous changes in the production process, and outputs a high-fidelity initial simulation data sequence. The optimal value of the future time period is based on the historical cycle data of the compound additive production process and the real-time equipment operation rhythm. The duration distribution of key process stages is calculated through statistical analysis models, and the minimum time window (such as 30 to 60 minutes) covering the complete cycle of phase transformation in the reactor is selected to ensure that the simulation can capture the dynamics of the entire process from feeding to maturation, while avoiding excessive calculation of resource consumption. W2. Based on the initial simulation data sequence and combined with the equipment operation status record data in the historical batch quality data, the dynamic simulation algorithm preset in the virtual production environment is used to analyze the changing trend of key parameters in the simulated compound additive production process. The product quality evolution path of the compound additive is generated according to the changing trend of key parameters, and the intermediate working condition parameter set is generated according to the product quality evolution path. Preferably, the steps for generating the product quality evolution path in step W2 are as follows: W201. Based on the initial simulation data sequence, extract the time-series change data of simulated temperature, pressure and material flow rate, identify the dynamic deviation characteristics of key process parameters, and compare them with the historical best batch target values ​​to generate parameter deviation maps. The steps for generating the parameter deviation map are as follows: Based on the time-series changes in simulated temperature, pressure, and material flow rate in the initial simulation data sequence, the dynamic deviation characteristics of key process parameters (such as fluctuation frequency and amplitude) are identified. The actual value of each key process parameter is compared with the historical best batch target value point by point to calculate the relative deviation. Then, the discrete deviation points are fused into a continuous surface through spatial interpolation technology to finally generate the parameter deviation map. This map visualizes the spatial distribution and intensity gradient of the deviation within the production area in the form of a heat map. W202. Based on the parameter deviation map and combined with the physicochemical quality test data in the historical batch quality data, identify potential quality risk points in the high deviation area of ​​the parameter deviation map and generate a quality risk area distribution map. The method for identifying potential risk points in step W202 is as follows: W2021. Extract the region in the parameter deviation map where the deviation value exceeds the preset risk threshold as the high deviation region. The preset risk threshold is obtained by calculating the deviation distribution of physicochemical quality test data in historical batch quality data through statistical analysis methods, and is used to define the critical deviation level that may cause quality problems. W2022. Combining the physicochemical quality testing data in the historical batch quality data, for each high deviation area, calculate the matching degree between its deviation value and the quality abnormality events in the physicochemical quality testing data, and generate a potential risk score for each high deviation area based on the matching degree. The matching degree is calculated by weighted fusion of the deviation amplitude, duration, and correlation coefficient of historical quality data in the high deviation area. The specific formula is as follows: the matching degree is equal to the deviation amplitude weight factor multiplied by the standardized deviation value, plus the duration weight factor multiplied by the deviation duration ratio, and then multiplied by the correlation coefficient between the quality abnormality event and the current deviation pattern in the historical physicochemical quality testing data. Finally, the calculation result is mapped to the interval of 0 to 1 through the normalization function to quantify the similarity between the high deviation area and the historical quality problem. W2023. When the potential risk score exceeds the preset action threshold, the high deviation area is marked as a potential quality risk point, and the location coordinates, risk intensity and impact range parameters of the potential quality risk point are recorded. The optimal value of the preset action threshold is based on the statistical analysis of abnormal events in the physicochemical quality testing data of historical batches. The upper limit of the risk score distribution (such as the 95th percentile) is calculated and dynamically calibrated in combination with the stability index in the equipment operation status record data. This ensures that the threshold can sensitively capture real risks while avoiding false triggering due to random fluctuations. The typical value is a normalized score between 0.75 and 0.90. The steps for generating the location coordinates, risk intensity, and impact range parameters of potential quality risk points are as follows: Based on the marked high deviation areas, the spatial coordinates of the risk points in the virtual production environment (such as a 3D mesh index) are calculated using a preset geometric centroid algorithm. The risk intensity is directly mapped to the potential risk score and corrected by multiplying it by the environmental disturbance coefficient in the equipment operation status record data. The impact range parameter is generated by calibrating the impact radius by simulating the propagation distance of the deviation along the material flow path and combining it with the ripple range data of historical quality anomaly events. W203. Based on the distribution map of quality risk areas, simulate the continuous evolution trajectory of the quality of compound additives during the production process, and adjust the path smoothness of the continuous evolution trajectory by combining the equipment operation status record data to generate the product quality evolution path. Based on the quality risk area distribution map, the specific content of simulating the continuous evolution trajectory of compound additive quality during the production process is as follows: Based on the quality risk area distribution map, the intensity change of each risk point is integrated along the production time axis to simulate the dynamic evolution curve of compound additive quality parameters (such as viscosity and active ingredient concentration). The smoothness of the trajectory is adjusted by the real-time stability index in the equipment operation status record data (such as using low-pass filtering to suppress noise). Finally, a continuous product quality evolution path that conforms to physical laws is generated, reflecting the entire process of risk accumulation and dissipation. W3. Combine the physicochemical quality testing data of the final product in the historical batch quality data to conduct a risk assessment on the intermediate operating condition parameter set, identify possible quality risk points, add a preset risk label to each quality risk point, and integrate the intermediate operating condition parameter set with risk labels into a virtual operating condition parameter set.

[0029] The specific steps for identifying potential quality risk points are as follows: By combining the intermediate operating condition parameter set with the physicochemical quality test data of the final products in historical batches, and scanning the abnormal fluctuation characteristics (such as sudden jumps or continuous deviations) in the intermediate operating condition parameter set, when the fluctuation amplitude exceeds the historical normal fluctuation range and has a significant statistical correlation with the abnormal records in the quality test data (such as excessive components or performance failure), the fluctuation area is marked as a quality risk point, and the risk level is assigned according to the fluctuation intensity and duration. Risk identifiers include risk type codes (such as abnormal temperature and pressure fluctuations), severity levels (divided into low, medium and high based on risk scores), impact time windows (the duration of the risk), and recommended response strategies (such as adjusting heating power or suspending feeding). These risk identifiers are generated by querying the historical processing case library and are bound to the virtual operating condition parameter set to guide the decision priority of the scheduling module. This embodiment uses a virtual simulation engine to accurately simulate the entire production process. By combining parameter deviation maps and risk matching degree calculations, it proactively identifies quality risk points and generates quantitative risk labels, achieving closed-loop control from prediction to intervention. This significantly improves the quality consistency and fault prediction capabilities of compound additive production, and reduces scrap rate and operation and maintenance costs.

[0030] In the production process of compound additives, based on the draft adjustment instructions, a virtual simulation engine is used to simulate the production process over the next 45 minutes, generating an initial simulation data sequence containing the temporal changes in temperature, pressure, and material flow rate. Based on this sequence, the dynamic deviation characteristics of key process parameters are extracted and compared with the historical best batch target values ​​to generate a parameter deviation map. High deviation areas with deviation values ​​exceeding a preset risk threshold of 0.75 are identified. The matching degree is calculated by combining historical physicochemical quality testing data. When the potential risk score exceeds the action threshold of 0.85, potential quality risk points are marked, and their coordinates, risk intensity, and impact range are recorded. Based on the quality risk area distribution map, the continuous evolution trajectory of quality is simulated to generate a product quality evolution path, thereby obtaining an intermediate operating condition parameter set. Finally, the intermediate operating condition parameter set is risk-assessed by combining historical quality data. Risk labels such as risk type codes, severity levels, impact time windows, and recommended response strategies are added to the identified quality risk points, integrating them into a virtual operating condition parameter set.

[0031] This example uses a virtual simulation engine to accurately simulate the entire process. By combining parameter deviation maps and risk matching degree calculations, it proactively identifies quality risk points and adds quantitative labels, achieving closed-loop control from prediction to intervention. This significantly improves the quality consistency, fault prediction capability, and decision-making efficiency of compound additive production, while reducing scrap rate and operation and maintenance costs.

[0032] One embodiment of the present invention is as follows: the generation step of the first process execution sequence in the scheduling generation module is as follows: A1. Extract risk labels and adjusted confidence factors from the virtual working condition parameter set, analyze the matching degree between the risk label level and the adjusted confidence factor, calculate the safety weight and execution urgency of each process step of the compound additive, and generate a preliminary process sequence draft. The preliminary process sequence draft includes process order, execution conditions and risk response strategies. The formulas for calculating the safety weight and execution urgency of each process step in the compound additive are as follows: The safety weight is calculated by extracting the risk level value from the risk identifier and combining it with the adjusted confidence factor through a weighted fusion algorithm, where the risk level weight accounts for 70% and the adjusted confidence factor weight accounts for 30%. When the adjusted confidence factor is lower than the preset threshold of 0.6, a safety compensation coefficient is introduced (the compensation coefficient is dynamically calculated based on the frequency of historical safety events) to generate a safety weight value. The execution urgency is calculated by calculating the remaining ratio of the current time to the planned completion time based on the impact time window parameter in the risk identifier, combining it with the risk level value through a linear normalization model to convert it into an urgency score, and then multiplying it by the reciprocal of the adjusted confidence factor for calibration to generate an execution urgency value, which is used to quantify the urgency of the process execution. A2. Based on the preliminary process sequence draft, and combined with the simulated temperature, pressure and material flow rate time sequence data in the virtual working condition parameter set, the first process execution sequence is generated. The first process execution sequence ensures that the production process operates efficiently within the safety threshold and supports the deviation comparison and optimization instruction generation of the feedback module.

[0033] Step A2 specifically involves: based on the preliminary draft of the process sequence, importing the simulated temperature, pressure, and material flow rate time-series data from the virtual operating condition parameter set, analyzing the matching degree between the fluctuation of the time-series data and the process sequence in the draft, and dynamically adjusting the process execution time point and resource allocation priority when the simulated temperature or pressure exceeds the preset safety range of the draft, reordering the process sequence based on the execution urgency value, and generating the first process execution sequence. This sequence integrates real-time simulation data and risk response strategies to ensure that the production process operates efficiently within the safety threshold and supports deviation comparison by the feedback module.

[0034] This embodiment dynamically generates process sequences by quantifying safety weights and execution urgency, thereby improving the accuracy and adaptability of compound additive production scheduling, ensuring process safety and efficiency, and reducing risk response delays.

[0035] One embodiment of the present invention is as follows: the generation step of the second process execution sequence in the scheduling generation module is as follows: B1. Based on the adjustment confidence factor exceeding the preset threshold, a preset real-time rescheduling instruction is triggered, which calls the risk identifier in the virtual working condition parameter set and the latest values ​​of temperature, pressure and material flow rate in the real-time collected raw data to generate a real-time production status snapshot. The real-time production status snapshot integrates the deviation information between the actual parameters of the current production environment and the virtual data for rescheduling decision-making. The preset real-time rescheduling command is an automated decision-making mechanism based on a rules engine. When the adjusted confidence factor exceeds a preset threshold (e.g., 0.8), the system automatically triggers this command. This command calls the risk indicators in the virtual operating condition parameter set and simultaneously extracts the latest values ​​of temperature, pressure, and material flow rate from the real-time collected raw data. B2. Based on real-time production status snapshots, and combined with the adjustment range and priority in the draft adjustment instructions, re-evaluate the execution sequence and resource allocation of each process in the production of compound additives, calculate the optimized process logic connection and parameter adjustment values, and generate a rescheduling plan. The optimized process logic connection and parameter adjustment value calculation steps are as follows: First, based on the deviation information in the real-time production status snapshot, clarify the current execution status and risks of each process. For each process, analyze its logical relationship with the preceding and following processes, and determine whether the execution order of the processes needs to be changed, based on the adjustment range and priority in the draft adjustment instructions. Then, for each process, consider its required resources, such as manpower, materials, and time. Based on the data in the real-time production status snapshot and the requirements of the draft adjustment instructions, reallocate resources to ensure their rational use. When allocating resources, prioritize the needs of high-priority processes, while also considering the total resource constraints and resource sharing between processes.

[0036] Next, based on the reassessed process execution sequence and resource allocation, adjustment values ​​for key parameters of each process are calculated. These parameters include, but are not limited to, temperature, pressure, and material flow rate, to ensure that each process can operate normally under the new execution sequence and resource allocation. For temperature parameters, the required temperature adjustment value is calculated based on the temperature deviation in the real-time production status snapshot and the requirements of the draft adjustment instructions, which may involve raising or lowering the temperature. For pressure parameters, similarly, appropriate pressure adjustment values ​​are determined based on the deviation and adjustment requirements to ensure the stability of the production process. For material flow rate, a reasonable flow rate adjustment value is calculated in conjunction with the process execution sequence and resource allocation to ensure that the material supply matches the process requirements. When calculating parameter adjustment values, the mutual influence and constraints between parameters are fully considered to ensure that the adjusted parameter combination meets the requirements of the production process.

[0037] B3. Based on the rescheduling scheme, integrate the real-time equipment health indicators of equipment operation status records from historical batch quality data, verify the feasibility and security of the rescheduling scheme, and generate the second process execution sequence.

[0038] This embodiment triggers a real-time rescheduling command, combines a virtual operating condition parameter set with real-time collected data to generate a production status snapshot, reassesses the process execution sequence and resource allocation based on this snapshot, generates a rescheduling plan, and verifies its feasibility and safety using equipment operating status records, generating a second process execution sequence. This process can quickly respond to changes in the production process and dynamically optimize production scheduling when the confidence factor is adjusted to meet the conditions. On the one hand, it can effectively deal with potential risks and anomalies in the production process, adjust the process sequence and resource allocation in a timely manner, avoid product quality degradation or equipment failure due to risk accumulation, and improve the stability and reliability of production.

[0039] One embodiment of the present invention is as follows: The feedback module specifically includes: collecting the actual quality data produced by the packaging equipment in real time, comparing the actual quality data with the virtual working condition parameter set generated by the quality pre-simulation module item by item, calculating the data deviation between the actual quality data and the virtual working condition parameter set, generating optimization instructions based on the magnitude and direction of the data deviation, and feeding the optimization instructions back to the acquisition and evaluation module and the collaborative optimization module through a preset communication protocol, thereby adjusting the data acquisition strategy of the acquisition and evaluation module and the historical data analysis parameters of the collaborative optimization module.

[0040] The beneficial effects of this embodiment are as follows: By collecting the actual quality data produced by the packaging equipment in real time and comparing it with the virtual operating condition parameter set, quality deviations in the production process can be detected in a timely manner. Optimization instructions generated based on data deviations can specifically adjust the data collection strategy of the data acquisition and evaluation module, making data collection more accurate and effective, and better reflecting the actual situation of the production process. At the same time, adjusting the historical data analysis parameters of the collaborative optimization module helps to more deeply explore the patterns and information in the historical data.

[0041] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A dynamic scheduling system for the production process of compound additives, characterized in that: It includes a data acquisition and evaluation module, a collaborative optimization module, a quality simulation module, a scheduling generation module, and a feedback module; The data acquisition and evaluation module is deployed near the reaction vessel and packaging equipment used to produce compound additives. It is used to collect raw data in real time during the production process of compound additives and generate reaction stage evaluation values ​​in real time based on the raw data. The collaborative optimization module receives the evaluation value of the reaction stage, analyzes it in conjunction with the historical batch quality data stored in the cloud, and outputs an adjustment instruction set, which includes an adjustment confidence factor and a draft adjustment instruction. The quality simulation module receives the draft adjustment instruction and simulates the production process of compound additives after injecting the draft adjustment instruction in a preset virtual environment within a future set time period, generating a set of virtual working condition parameters. The scheduling generation module is used to perform the following actions based on the virtual operating condition parameter set and the adjustment confidence factor: when the adjustment confidence factor does not exceed a preset threshold, a first process execution sequence is generated; when the adjustment confidence factor exceeds the preset threshold, a second process execution sequence is generated. The feedback module collects the actual quality data produced by the packaging equipment, compares it with the virtual working condition parameter set, calculates the deviation, generates optimization instructions based on the deviation, and feeds the optimization instructions back to the acquisition and evaluation module and the collaborative optimization module.

2. The dynamic scheduling system for the production process of compound additives according to claim 1, characterized in that: The raw data in the acquisition and evaluation module includes temperature, pressure, and material flow rate; The raw data is collected in real time by temperature sensors, pressure sensors, and flow sensors deployed near the reactor and dispensing equipment.

3. The dynamic scheduling system for the production process of compound additives according to claim 2, characterized in that: The steps for generating the reaction stage evaluation value in the acquisition and evaluation module are as follows: S1. Based on the real-time collected raw data, calculate the dynamic change trend of temperature, pressure and material flow rate within adjacent preset time windows, input the dynamic change trend into a preset material reaction knowledge base for matching, and identify the degree of deviation between the reaction process of the current compound additive and the standard phase transition mode in the material reaction knowledge base. S2. Based on the degree of deviation and combined with the preset weighting coefficient, the evaluation value of the reaction stage is calculated.

4. The dynamic scheduling system for the production process of compound additives according to claim 3, characterized in that: The historical batch quality data in the collaborative optimization module includes the actual addition ratio data of each component of the compound additive in the historical production batch, the time sequence record data of key process parameters in the historical production batch, the physicochemical quality test data of the final product of the historical production batch, the equipment operation status record data of the historical production batch, and the historical optimal batch target value. The steps for generating the adjustment instruction set in the collaborative optimization module are as follows: Q1. Based on the reaction stage evaluation value and the time-series record data of key process parameters in the historical batch quality data, identify the deviation between the current reaction process of the compound additive and the historical optimal batch target value. Combined with the physicochemical quality test data of the final products of historical production batches, determine the adjustment items of key process parameters and generate a preliminary adjustment strategy. Q2. Based on the preliminary adjustment strategy, and combining the actual addition ratio data in the historical batch quality data with the equipment operation status record data, calculate the adjustment range and adjustment priority of each key process parameter, and correct the adjustment range to generate a draft adjustment instruction. Q3. Generate adjustment confidence factors based on the aforementioned adjustment instruction draft, integrate the adjustment instruction draft and adjustment confidence factors into an adjustment instruction set, and output it to the quality pre-simulation module; The method for generating the adjusted confidence factor is as follows: Q301. Based on the draft adjustment instruction, call the historical adjustment strategy records and corresponding batch physicochemical quality test data in the historical batch quality data, identify the success rate of similar draft adjustment instructions in historical execution, and calculate the preliminary confidence score. Q302. Based on the preliminary confidence score, combined with the dynamic trends of temperature, pressure and material flow rate in the real-time collected raw data, analyze the deviation between the current production environment of the compound additive and the successful environment in the past, convert the deviation into a correction coefficient, and integrate the preliminary confidence score and the correction coefficient through a weighted fusion algorithm to generate an adjusted confidence factor.

5. The dynamic scheduling system for the production process of compound additives according to claim 4, characterized in that: The calculation steps for the adjustment range and priority of the key process parameters in step Q2 are as follows: Q201. Based on the preliminary adjustment strategy, extract the deviation between the current value of the key process parameter and the historical optimal batch target value, combine the actual addition ratio data in the historical batch quality data, multiply the deviation by the addition ratio and then multiply by the preset allocation weight to calculate the initial adjustment range of each key process parameter. Q202. Combine the equipment operating status record data in the historical batch quality data to generate the equipment status stability coefficient and the equipment real-time operation feedback data. Then, weight and fuse the equipment status stability coefficient with the initial adjustment range to generate the adjustment priority of each of the key process parameters. Q203. Based on the initial adjustment range and the adjustment priority, a preset historical adjustment success rate factor and the real-time operation feedback data of the device are introduced to correct the initial adjustment range, generate the corrected adjustment range and adjustment priority, and use them as a draft adjustment instruction.

6. The dynamic scheduling system for the production process of compound additives according to claim 5, characterized in that: The method for generating the virtual operating condition parameter set in the quality pre-simulation module is as follows: W1. Based on the aforementioned draft adjustment instructions, the draft adjustment instructions are injected into the preset virtual production environment through a preset virtual simulation engine to simulate the production process of compound additives within a future set time period, generating an initial simulation data sequence. The initial simulation data sequence includes time-series changes in simulated temperature, pressure, and material flow rate. W2. Based on the initial simulation data sequence and combined with the equipment operation status record data in the historical batch quality data, the dynamic simulation algorithm preset in the preset virtual production environment is used to analyze the changing trend of key parameters in the simulated compound additive production process, generate the product quality evolution path of the compound additive according to the changing trend of key parameters, and generate an intermediate working condition parameter set according to the product quality evolution path. W3. Combining the physicochemical quality testing data of the final product in the historical batch quality data, conduct a risk assessment on the intermediate operating condition parameter set, identify quality risk points, add a preset risk label to each quality risk point, and integrate the intermediate operating condition parameter set with the risk label into a virtual operating condition parameter set.

7. The dynamic scheduling system for the production process of compound additives according to claim 6, characterized in that: The steps for generating the product quality evolution path in step W2 are as follows: W201. Based on the initial simulation data sequence, extract the time-series change data of simulated temperature, pressure and material flow rate, identify the dynamic deviation characteristics of the key process parameters, and compare them with the historical best batch target values ​​to generate a parameter deviation map. W202. Based on the parameter deviation map and combined with the physicochemical quality testing data in the historical batch quality data, identify potential quality risk points in the high deviation area of ​​the parameter deviation map and generate a quality risk area distribution map. The method for identifying the potential risk points in step W202 is as follows: W2021. Extract regions in the parameter deviation map whose deviation values ​​exceed a preset risk threshold as high deviation regions; W2022. Combining the physicochemical quality testing data in the historical batch quality data, for each high deviation area, calculate the matching degree between its deviation value and the quality abnormality event in the physicochemical quality testing data, and generate a potential risk score for each high deviation area based on the matching degree. W2023. When the potential risk score exceeds the preset action threshold, the high deviation area is marked as a potential quality risk point. W203. Based on the quality risk area distribution map, simulate the continuous evolution trajectory of the quality of compound additives during the production process, and adjust the path smoothness of the continuous evolution trajectory in combination with the equipment operation status record data to generate a product quality evolution path.

8. The dynamic scheduling system for the production process of compound additives according to claim 7, characterized in that: The steps for generating the first process execution sequence in the scheduling generation module are as follows: A1. Extract the risk identifiers and adjusted confidence factors from the virtual working condition parameter set, analyze the matching degree between the risk identifier level and the adjusted confidence factor, calculate the safety weight and execution urgency of each process step of the compound additive, and generate a preliminary process sequence draft. A2. Based on the preliminary process sequence draft, and combined with the simulated temperature, pressure and material flow rate time series data in the virtual working condition parameter set, generate the first process execution sequence.

9. The dynamic scheduling system for the production process of compound additives according to claim 8, characterized in that: The steps for generating the second process execution sequence in the scheduling generation module are as follows: B1. Based on the fact that the adjustment confidence factor exceeds the preset threshold, a preset real-time rescheduling instruction is triggered, which calls the risk identifier in the virtual working condition parameter set and the latest values ​​of temperature, pressure and material flow rate in the real-time collected raw data to generate a real-time production status snapshot. B2. Based on the real-time production status snapshot, and in conjunction with the adjustment range and adjustment priority in the draft adjustment instruction, re-evaluate the execution order and resource allocation of each process in the compound additive production, calculate the optimized process logic connection and parameter adjustment value, and generate a rescheduling scheme. B3. Based on the rescheduling scheme, integrate the real-time equipment health indicators of the equipment operation status record data in the historical batch quality data, verify the feasibility and security of the rescheduling scheme, and generate the second process execution sequence.

10. The dynamic scheduling system for the production process of compound additives according to claim 9, characterized in that: The specific content of the feedback module is as follows: The feedback module collects the actual quality data produced by the packaging equipment in real time, compares the actual quality data with the virtual working condition parameter set generated by the quality pre-simulation module item by item, calculates the data deviation between the actual quality data and the virtual working condition parameter set, generates optimization instructions based on the magnitude and direction of the data deviation, and feeds the optimization instructions back to the acquisition and evaluation module and the collaborative optimization module through a preset communication protocol, so as to adjust the data acquisition strategy of the acquisition and evaluation module and the historical data analysis parameters of the collaborative optimization module.

Images