Quality detection method and system for propanol rectification process

Abstract

The invention relates to the technical field of propyl alcohol rectification detection, and discloses a quality detection method and system for a propyl alcohol rectification process. According to the method, a data acquisition module acquires initial operation data, counts operation time and task requirements of partitioned tower section equipment, matches available time of the equipment with man-hour distribution of personnel, and generates a tower section equipment distribution result; the quality analysis module calculates key quality parameter detection time and interval based on the result, sorts a man-hour priority calling sequence, and generates a parameter sorting result; the constraint evaluation module is used for calculating material supply and equipment operation time, calibrating resource conflicts, adjusting a detection interval and generating a constraint adjustment parameter set; the progress monitoring module counts time node and parameter distribution difference values and recombines resource distribution to generate a rectification progress plan; and the deviation response module monitors environment and equipment states, calculates deviation, adjusts equipment allocation and time nodes, and generates a deviation correction result. The system realizes precise and dynamic management and control of quality detection in the propyl alcohol rectification process.

Description

Technical Field

[0001] The present invention relates to the technical field of propanol distillation detection, in particular to a quality detection method and system for a propanol distillation process. Background Art

[0002] As an important chemical raw material, propanol is widely used in solvents, organic synthesis, pharmaceutical intermediates, and other fields. Its purity and quality directly impact the performance and application of downstream products. In the propanol production process, distillation is a key step in separating the mixture and improving product purity. This process involves the coordinated operation of multiple tower sections, complex material transfer and energy exchange, and requires extremely high operational precision and quality control.

[0003] Traditional quality monitoring of the propanol distillation process relies heavily on manual sampling and offline analysis, a method with significant limitations. For one thing, fixed sampling intervals make it difficult to capture dynamic changes in key parameters during the distillation process in real time. Quality anomalies are often only discovered through delayed data feedback after they occur, leading to substandard products or delayed production adjustments. Furthermore, offline analysis requires samples to be sent to a laboratory for testing, which is time-consuming and cannot meet the requirements for real-time quality monitoring during continuous production, resulting in a significant waste of production resources.

[0004] The traditional model lacks a scientific approach to equipment and personnel deployment. Equipment operating hours, task requirements, and operator man-hour allocation for each section of the distillation tower often rely on empirical arrangements. This often leads to a mismatch between equipment availability and task requirements, and an imbalance in man-hour allocation. Some sections of the tower may waste resources due to idle equipment, while others may suffer from insufficient equipment or untimely personnel deployment, impacting inspection efficiency and thus disrupting the stability of the entire distillation process.

[0005] The testing schedule for key quality parameters also suffers from deficiencies. Key parameters in the propanol distillation process, such as temperature, pressure, and component concentrations, lack data support for their testing times and intervals, often relying on a fixed, uniform cycle. This approach can lead to insufficient testing of key parameters, preventing the timely detection of minor fluctuations, or excessive testing of secondary parameters, which consumes excessive instrument and personnel resources and reduces overall testing efficiency.

[0006] During the production process, the material supply cycle and equipment operating hours are poorly coordinated, making equipment deployment time conflicts or resource shortages prone to occur. When a tower section experiences a sudden equipment failure or material supply delay, traditional systems lack effective parameter adjustment mechanisms, making it difficult to quickly adapt to changing production conditions. This can trigger a chain reaction, affecting the progress and quality of the entire distillation process. Furthermore, environmental factors such as temperature and humidity changes, as well as natural fluctuations in equipment operating conditions, can also affect distillation results. Traditional detection methods have limited responsiveness to these dynamic deviations, making real-time correction and regulation difficult. Summary of the Invention

[0007] The object of the present invention is to provide a quality detection system for a propanol distillation process to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides a quality detection system for a propanol distillation process, the method comprising: The data acquisition module is used to obtain the initial operating data of the distillation process. Based on the material characteristics and equipment parameters, it calculates the equipment operating time and task requirements of the tower section, matches the equipment available time with the task requirements, calculates the difference in operator man-hour allocation, compares the equipment and personnel availability, and generates the tower section equipment allocation results. A quality analysis module calculates the detection time and parameter interval of key quality parameters based on the tower section equipment allocation result, sorts the instrument working hours and operator working hours priority call order, calculates the parameter resource utilization rate and parameter interval weight, and generates the key quality parameter sorting result; A constraint evaluation module calculates the material supply cycle and equipment operating time based on the ranking results of the key quality parameters, calibrates equipment allocation time conflicts and resource shortage tasks, adjusts parameter detection intervals, and generates a parameter constraint adjustment parameter set; A progress monitoring module adjusts the parameter set based on the parameter constraints, calculates the distribution difference between the distillation process time nodes and parameters, matches resource utilization with equipment available time, calculates the adjustment range of parameters and time nodes, reorganizes the parameter time and equipment resource distribution, and generates a distillation schedule; The deviation response module monitors the environmental changes and equipment operating status of the distillation process based on the distillation schedule, calculates the parameter time and environmental change deviation, counts the equipment working hours and material remaining amount, adjusts the equipment deployment and time nodes, and generates the distillation deviation correction result.

[0009] Preferably, when the data acquisition module obtains the initial operation data of the distillation process, it collects the task requirements and equipment operating time of the partitioned tower sections, comprehensively calculates the available time of the multi-partition equipment and the matching degree with the task requirements, and calculates the deviation between the total equipment time and the task requirements to generate an equipment time deviation set; Analyze the equipment time deviation set, adjust the equipment and operator work time distribution according to the availability of operator resources, and establish an operator and equipment work time adjustment set; According to the operator and equipment working hour adjustment set, equipment and operator allocation comparison is performed to obtain the tower section equipment allocation result.

[0010] Preferably, the quality analysis module analyzes the resource allocation of each key quality parameter based on the tower section equipment allocation result, calculates the starting detection time, and uses a resource optimization algorithm to predict the optimal parameter sequence to generate a start schedule for each key quality parameter; Based on the start-up schedule of each key quality parameter, a priority ranking model is used to prioritize the instrument and operator resources, and the time allocation of the instrument and operator is adjusted according to the urgency of the parameter to establish a parameter priority list; By using the parameter priority list and combining it with the interval requirements in actual parameter detection, the parameter resource utilization and parameter interval weight of the key quality parameters are calculated to generate a key quality parameter ranking result.

[0011] Preferably, the constraint evaluation module extracts the start time and material cycle of each parameter from the key quality parameter ranking results, analyzes equipment usage, applies a resource allocation algorithm to determine the resource requirement of each parameter, and generates a resource requirement analysis result; Based on the resource demand analysis results, mark all time conflicts between resource supply and demand as well as conflicts and resource shortages occurring in equipment allocation, and create a conflict and shortage index table; The conflict and shortage index table is used to recalculate the detection interval of each parameter, optimize the parameter detection plan, and generate a parameter constraint adjustment parameter set.

[0012] Preferably, the progress monitoring module extracts the distribution differences between the time nodes and parameters of the distillation process from the parameter constraint adjustment parameter set, analyzes the parameter detection sequence and resource utilization efficiency based on the comparison of parameter time and resource distribution, and generates a distribution table of time nodes and parameter differences; By using the time node and parameter difference distribution table, resource utilization and device available time are matched, and based on the matching analysis of resource allocation and parameter requirements, time conflict and resource shortage parameters are calibrated to create resource conflict and shortage parameter results; Based on the resource conflicts and insufficient parameter results, the adjustment range and time node correction values ​​of parameter detection are calculated, the parameter time distribution is optimized, and a distillation schedule is generated.

[0013] Preferably, the deviation response module extracts monitoring data from the distillation schedule, including environmental changes and equipment operating status during the distillation process, combines parameter time nodes, uses data analysis methods to determine the impact of the environment and equipment status on the distillation schedule, and generates environmental and equipment status analysis results; Using the results of the environment and equipment status analysis, calculate the deviation between parameter time and environmental changes, and count the working hours and material remaining of each device. Through quantitative analysis methods, calibrate the equipment deployment and time nodes that need to be adjusted, and create a parameter and environment deviation table; Based on the parameter and environmental deviation table, the equipment allocation and time nodes are adjusted to match the actual distillation environment changes, the parameter allocation is optimized, and the distillation offset correction result is generated.

[0014] Preferably, the system further comprises a data storage module for synchronously retrieving the operation data of each storage port through the to-be-stored port of the distillation database for analysis, and performing distributed storage management of the cluster distillation information; The data storage module processes the operating data of each storage port to obtain a storage performance benchmark value of each storage port, wherein the storage performance benchmark value of each storage port is used to comprehensively quantify the storage capacity utilization of each storage port; Counting cluster distillation information data of each edge computing node, wherein the cluster distillation information data of each edge computing node includes a data processing energy efficiency characterization value of each edge computing node and the number of target receiving data collection points; Comprehensively analyze the cluster distillation information data of each edge computing node to obtain a cluster distillation information evaluation value of each edge computing node, and the cluster distillation information evaluation value of each edge computing node is used to comprehensively quantify the comprehensive performance of each edge computing node; The storage performance verification value of each edge computing node is obtained by matching the cluster distillation information evaluation value of each edge computing node; The storage performance benchmark value of each storage port is compared with the storage performance verification value of each edge computing node to obtain the target storage port of each edge computing node, and the cluster distillation information of each edge computing node is transmitted to the corresponding target storage port.

[0015] Preferably, the data storage module processes and obtains a load evaluation value of each data storage node based on the load parameters of each data storage node, and performs load balancing configuration on each data storage node according to the load evaluation value of each data storage node; The load balancing configuration includes counting each data storage node in the distillation database management system and obtaining the load parameters of each data storage node to perform load balancing processing.

[0016] Preferably, the system further comprises a quality verification module for extracting all quality parameter values ​​of the remaining sub-parameter detection areas after deletion, and comparing the quality parameter values ​​with corresponding quality standard parameter values, dividing all quality parameter values ​​according to the comparison results, and constructing a first quality sequence and a second quality sequence according to the division results; numerically sorting all quality parameter values ​​in the first mass number sequence and the second mass number sequence, using every two quality parameter values ​​in the first mass number sequence or the second mass number sequence as quality parameters to be associated, verifying whether the quality parameters to be associated are associated based on a quality evaluation model, and counting the number of associations based on the verification results; The comprehensive quality value of the distillation process is determined according to the associated quantity and the unassociated quality parameter value.

[0017] Preferably, the present invention also includes a quality detection method for a propanol distillation process, including all modules and method flows of the above-mentioned quality detection system for a propanol distillation process.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This quality inspection system for the propanol distillation process effectively optimizes the quality control process through the coordinated operation of multiple modules. The data acquisition module accurately calculates and compares equipment operating time, task requirements, and personnel hours for each tower section based on material characteristics and equipment parameters, generating scientific equipment allocation results for each tower section. This process breaks away from the traditional experience-based allocation model, ensuring a more appropriate match between equipment availability and task requirements, reducing equipment idleness and resource waste. It also balances operator hours and avoids inspection delays caused by imbalanced personnel allocation.

[0019] Based on the tower equipment allocation results, the quality analysis module calculates the key quality parameter testing time and parameter intervals, and prioritizes instrument and operator time. By incorporating parameter resource utilization and parameter interval weights, the generated key quality parameter ranking results are more targeted, adjusting testing priorities based on the impact of different parameters on distillation quality. This ensures that important parameters receive more appropriate testing resources, avoiding the uneven resource allocation problem of traditional fixed-period testing and ensuring that quality testing is more closely aligned with the actual needs of the distillation process.

[0020] The Constraint Assessment Module, based on the ranking results of key quality parameters, comprehensively calculates material supply cycles and equipment operating times, accurately calibrating tasks with equipment scheduling conflicts and resource shortages. It then generates a parameter constraint adjustment parameter set by adjusting parameter detection intervals. This module effectively resolves common resource conflicts in the production process. When material supply and equipment operating times mismatch, it dynamically adjusts detection intervals to allow for equipment scheduling and resource replenishment, reducing production interruptions or quality fluctuations caused by resource conflicts.

[0021] The progress monitoring module adjusts parameter sets based on parameter constraints, collects statistics on the difference between distillation process time nodes and parameter distribution, and calculates the adjustment range of parameters and time nodes by matching resource utilization with equipment available time. Ultimately, it restructures the parameter time and equipment resource distribution to generate a distillation schedule. This plan clearly presents the inspection tasks and resource allocation at each stage, making the entire distillation process more organized. Operators can work in an orderly manner according to the schedule, reducing inefficiencies caused by task confusion.

[0022] The Deviation Response Module monitors environmental changes and equipment operating status during the distillation process in real time. When deviations between parameter timing and environmental changes are detected, the module promptly adjusts equipment deployment and timelines by calculating equipment hours and remaining material, generating distillation deviation correction results. This feature enhances the system's ability to respond to dynamic disturbances. When environmental factors or equipment status fluctuate, it can quickly respond and revise the inspection plan, avoiding the delayed response to abnormal changes often seen with traditional inspection methods. This keeps the distillation process under control and ensures the continuity and accuracy of quality inspections.

[0023] The synergistic effect of each module forms a closed-loop quality inspection system, which comprehensively covers the quality control links of the propanol distillation process from resource allocation, parameter planning, conflict resolution, progress control to deviation correction, making the entire inspection process more scientific and flexible, and adapting to the complex needs of propanol distillation production. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a timing diagram of the quality detection system for the propanol distillation process according to the present invention; Figure 2 This is a flow chart of the working principle of the data acquisition module; Figure 3 A flowchart showing the working principle of the constraint evaluation module; Figure 4 This is a flowchart of the distributed storage management working principle of the data storage module. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] See also Figure 1 The present invention provides a quality detection method and system for a propanol distillation process, the method comprising: Dynamic quality monitoring and optimization of the distillation process are achieved through the collaborative work of multiple modules. The system uses the data acquisition module to obtain initial operating data for the distillation process, including equipment operating time, task requirements, and operator man-hour distribution differences for each tower section. The system then generates tower section equipment allocation results by comparing equipment and personnel availability. Based on these results, the quality analysis module calculates the inspection time and intervals for key quality parameters, prioritizes instruments and operator man-hours, and generates key quality parameter ranking results. The constraint assessment module analyzes material supply cycles and equipment operating times, identifies resource conflicts, adjusts inspection intervals, and generates parameter constraint adjustment parameter sets. The progress monitoring module calculates the differences between time nodes and parameter distributions, reorganizes parameter times and equipment resource distribution, and generates a distillation schedule. The deviation response module monitors environmental changes and equipment status in real time, adjusts equipment allocation and time nodes, and generates distillation offset correction results.

[0027] Example 1: See Figure 2 When the data acquisition module obtains the initial operating data of the distillation process, it first extracts the equipment operation records of the partitioned tower sections from the production control system, including the operating time, start and stop times, and current working status of each tower section. After preliminary screening, these data are matched and analyzed with the task requirements. The task requirements come from the production planning system, including the target output, quality requirements, and estimated completion time of each tower section. The system uses a time series analysis method to compare the equipment operating time with the time window of the task requirements and calculates the degree of match between the two. When the equipment operating time exceeds the task requirements, the system marks it as excess resources; when the equipment operating time is insufficient, it is marked as insufficient resources. This process generates a set of equipment time deviations, which not only contains the resource status of each tower section, but also records the specific value of the deviation.

[0028] Another important function of the data collection module is the analysis of operator resource availability. The system obtains operator scheduling information, skill levels, and historical work hours from the human resources management system. By comparing the task complexity of each tower section and the skill matching of the operators, the system calculates the difference in operator work hours allocation. For example, high-complexity tower sections require highly skilled operators, while low-complexity tower sections can be assigned ordinary operators. The calculation of the work hour allocation difference is based on the difference between task requirements and the actual available work hours of operators. The system optimizes the allocation plan through a dynamic adjustment algorithm. When the operating time of a tower section equipment is insufficient but the operator work hours are sufficient, the system automatically increases the operator allocation ratio for that equipment; otherwise, the allocation ratio is reduced or the task priority is adjusted. This process generates a set of operator and equipment work hour adjustments to ensure a dynamic balance in resource allocation.

[0029] The generation of tower section equipment allocation results relies on a comprehensive analysis of the equipment time deviation set and the working time adjustment set. The system uses a multi-objective optimization algorithm to maximize the utilization of equipment and operators while meeting task requirements. For example, when a tower section has excess equipment resources, the system will prioritize high-priority tasks in other tower sections; when operator hours are insufficient, the system will adjust the task sequence or introduce temporary scheduling. The allocation comparison process not only considers the current resource status, but also predicts resource changes within the future time window to ensure the sustainability of the allocation results. The final tower section equipment allocation results are output in the form of structured data, including the equipment configuration of each tower section, operator allocation, and task execution schedule.

[0030] The quality analysis module further optimizes the detection process of key quality parameters based on the tower equipment allocation results. The system first identifies the key quality parameters of each tower section, such as purity, boiling point, impurity content, etc., and analyzes its detection requirements. The detection requirements include detection frequency, detection accuracy and the type of instrument required. The system predicts the optimal parameter sequence through a resource optimization algorithm that comprehensively considers detection time, resource occupancy and parameter priority. For example, high-priority purity detection parameters will be scheduled in time periods when equipment and operator resources are sufficient, while low-priority impurity detection parameters may be postponed. This process generates a start-up schedule for each key quality parameter, clarifying the detection start time, duration and resource occupancy of each parameter.

[0031] The prioritization model plays a key role in the quality analysis module. The system dynamically prioritizes instrument and operator resources based on parameter urgency, process requirements, and historical data volatility. For example, purity parameter testing typically requires a high-precision gas chromatograph and skilled operators, and the system prioritizes these resources. The prioritization process not only considers the priority of the current task but also incorporates the resource requirements of future tasks to avoid resource conflicts. The parameter priority list is generated based on a weighted score of multiple factors, including parameter importance, test frequency, and resource utilization.

[0032] The calculation of parameter resource utilization and parameter interval weights is one of the core functions of the quality analysis module. Resource utilization reflects the actual occupancy of instruments and operators and is calculated as the ratio of testing time to total available time. For example, if a gas chromatograph has a daily testing time of 6 hours and a total available time of 8 hours, its utilization rate is 75%. Parameter interval weights measure the importance of the testing interval, with higher-weighted parameters requiring more frequent testing. The system optimizes the testing interval through a dynamic adjustment algorithm to ensure that high-weighted parameters receive sufficient testing frequency while avoiding resource waste. The final key quality parameter ranking results are stored as structured data, including the testing time, resource allocation, and priority score of each parameter.

[0033] The collaborative work of the data acquisition and quality analysis modules ensures efficient and reliable quality inspection processes for the distillation process. The system adapts to changes in the production environment through real-time data acquisition and dynamic resource allocation. For example, if a tower section experiences a sudden equipment failure, the system immediately adjusts the inspection plan, migrating tasks to backup equipment or reassigning operators. This dynamic adjustment capability not only improves inspection efficiency but also reduces the risk of production interruptions. Data exchange between modules is achieved through a unified interface, ensuring accurate and timely information transmission. The entire implementation process requires no human intervention; the system automatically completes data collection, analysis, and decision-making, providing intelligent support for quality control of the distillation process.

[0034] Example 2: See Figure 3The constraint assessment module extracts the start time and material cycle information for each parameter from the key quality parameter ranking results. Material cycles are derived from real-time data from the material supply system, including raw material feed rates, intermediate product inventory, and finished product output cycles. When analyzing equipment usage, the system integrates equipment operation logs from each section of the distillation column to record the current load status, maintenance schedule, and historical failure rates. The resource allocation algorithm performs calculations based on multi-dimensional constraints, including maximum equipment capacity, operator skill matching, and material supply continuity. The algorithm first establishes a resource requirement model, breaking down each parameter testing task into three core elements: equipment occupancy time, operator man-hours, and material consumption. Through an iterative optimization process, the system assigns precise resource allocation values ​​to each parameter. For example, a purity test task may require 2 hours of gas chromatograph time, 1.5 hours of senior operator time, and 200 ml of a specific material sample. The resource requirement analysis results are output in a structured table, clearly indicating the specific resource requirements and time windows for each parameter for the three resource types.

[0035] Based on the results of resource demand analysis, the system automatically scans the time overlapping areas of all parameter tasks. When multiple parameters apply for the same equipment resource in the same time period, the conflict detection engine marks it as an equipment allocation time conflict. For example, a mass spectrometer is required for both tower top temperature monitoring and tower bottom component analysis. The system identifies the conflict and records the conflict parameter number, conflicting equipment, and time interval. Resource shortage tasks are determined by comparing resource requirements with actual available resources: if the total required operating hours for a certain period of time is 10 hours, but only 8 hours are actually available, it is marked as a personnel resource shortage; if the material inventory is lower than the total inspection demand, it is marked as a material shortage. The conflict and shortage index table is stored in a matrix structure, with rows representing time segments, columns representing resource types, and cells recording conflict levels (such as high / medium / low) and shortage quantities (such as a material shortage of 50ml). This table provides a visual basis for subsequent parameter adjustments.

[0036] The dynamic relaxation algorithm is used to recalculate the parameter detection interval. The system first identifies the high-conflict parameters in the conflict index table and analyzes the adjustable space of their detection frequency. For non-critical parameters or low-weight parameters, the system automatically extends the detection interval. For example, a certain impurity content test was originally scheduled to be conducted every 2 hours. After analysis, the maximum interval allowed by the process is 4 hours. The system adjusts it to 3.5 hours to alleviate equipment conflicts. For resource-shortage tasks, the system starts the material reallocation program: when there is a shortage of materials in a certain period of time, high-priority parameter detection is prioritized and low-priority tasks are delayed; shortage of operating personnel triggers cross-regional scheduling, and personnel are temporarily deployed from idle tower sections. The parameter detection plan optimization process introduces a buffer time mechanism, inserting a 5-15 minute flexible interval between adjacent detection tasks to deal with emergencies. The final generated parameter constraint adjustment parameter set contains three core data: the updated detection schedule, the resource reallocation plan, and the conflict resolution record.

[0037] After receiving the parameter constraint adjustment parameter set, the progress monitoring module first parses the time node data of the distillation process. The time nodes are derived from the process flow chart, including time markers of key stages such as preheating start-up, full reflux operation, and product extraction. When the system calculates the distribution difference between parameters, it uses time axis mapping technology: each parameter detection task is marked on the distillation process timeline, and the deviation between the interval length of adjacent detection tasks and the process requirements is calculated. For example, the process requires tower temperature monitoring every 30 minutes, but it is actually adjusted to 45 minutes due to resource conflicts, then a negative difference of 15 minutes is recorded. The time node and parameter difference distribution table is presented in a dual-axis coordinate system, with the horizontal axis being the distillation process timeline and the vertical axis being the difference magnitude, forming a visual distribution map.

[0038] A sliding window algorithm is used to pair resource utilization with equipment availability. The system divides the entire day's work schedule into 15-minute time units and compiles three statistics within each time window: the average resource utilization, equipment idleness, and cumulative parameter differences for all inspection tasks during that period. When resource utilization exceeds 85% and a negative parameter difference exists within a time window, the system marks it as a high-conflict risk zone. When the equipment idleness exceeds 40% but the parameter difference is positive, it is marked as a resource-idle zone. The resource conflict and insufficient parameter result table is sorted by time window number, noting the conflict type (equipment / personnel / materials), conflict parameter number, and recommended adjustment direction for each window.

[0039] Parameter adjustment ranges are calculated based on quantitative analysis of the conflict results table. For high-conflict risk areas, the system calculates three adjustment values: time node shift (maximum 30 minutes), test time compression (maximum 15%), and resource substitution options (e.g., replacing some mass spectrometry with a UV spectrometer). Time node corrections are generated using a genetic algorithm. The system simulates hundreds of time adjustment scenarios and selects the one with the smallest sum of parameter differences as the optimal solution. Parameter time distribution optimization involves a three-stage reorganization: first, high-conflict parameters are moved to adjacent low-utilization periods; second, long-running test tasks are split into multiple subtasks and arranged interspersed; and finally, test buffers are established before critical process nodes. After consistency verification, the reorganized time distribution plan generates a distillation schedule consisting of timescales, parameter sequences, and a resource allocation matrix. This plan is presented as a Gantt chart, with color-coded areas for normal execution, adjustment risk areas, and buffer zones, providing operators with intuitive progress guidance.

[0040] A closed-loop feedback mechanism was established during system implementation. When the progress monitoring module detects a continued widening of parameter discrepancies, it automatically triggers re-optimization in the constraint evaluation module. When resource conflicts cannot be resolved within the current parameter set, the data acquisition module is contacted to request additional resources or adjust task requirements. All optimization decisions are recorded in the version control system, forming a traceable progress optimization chain.

[0041] Example 3: The deviation response module continuously receives the time node data in the distillation schedule and the real-time stream of the environmental monitoring system. Environmental change data include tower section temperature gradient, pressure fluctuation, cooling water flow anomaly, etc., with a sampling frequency of 5 times per second; the equipment operation status covers motor current, valve opening, pump vibration spectrum, etc., which are uploaded every 200 milliseconds through the industrial Internet of Things node. The system establishes a three-dimensional time mapping model: the first dimension is the process timeline (distillation stage progress), the second dimension is the environment timeline (sensor timestamp), and the third dimension is the equipment timeline (controller response timing). When the three-axis time deviation exceeds the process tolerance threshold, the deviation analysis engine is triggered. The environment and equipment status analysis uses multi-source data fusion technology, such as aligning the temperature sensor data with the heat exchanger valve opening in time and space, and calculating the environmental time-varying factor by the following formula:

[0042] in: represents the environmental time-varying factor, is the weight coefficient of the i-th type environmental parameter (temperature 0.4, pressure 0.3, flow 0.3), is the actual monitoring value, is the process expectation value, The parameter's allowable fluctuation range. A calculated value greater than 1 generates a yellow alert, while a value greater than 1.5 triggers a red alert. Equipment status analysis incorporates a health index model, converting vibration spectrum characteristics (e.g., an excessive 2kHz amplitude) into an equipment reliability score. Equipment scores below 80 are automatically marked as high-risk.

[0043] The calculation of parameter time deviation adopts a dual-track analysis framework. The main track modifies the detection time node based on the environmental time-varying factor: When the value is between 1.0 and 1.2, the system compresses the inspection time by a proportional factor of 0.8. When it exceeds 1.5, a backup inspection plan is activated. Auxiliary tracks equipment chain reactions. For example, when abnormal reboiler temperature causes tower pressure to rise, the system automatically increases the interval between adjacent inspection tasks by 20%-40%. The deviation quantification matrix is ​​updated every 15 minutes and records three core data types: the number of minutes of inspection delay caused by environmental factors, the loss of effective inspection time due to equipment degradation, and parameter correlation deviations caused by chain reactions.

[0044] The equipment working hour statistics adopts a dynamic accumulation algorithm. The system collects real-time data from the running counters of each device (such as the cumulative running hours of a centrifugal pump), and combines it with the current task load to predict the remaining available working hours. The remaining material quantity is monitored through the combination of weighing sensors and flow meters, and the material balance equation is updated every 30 seconds:

[0045] Wherein: is the real-time material quantity, is the initial material quantity, is the output flow, is the replenishment flow, is the sampling interval. When the predicted remaining working hours are insufficient for the current detected task demand by 110%, the system automatically marks the device as overloaded; when the material quantity is less than 120% of the total demand of the next three detection tasks, a material warning is triggered. The quantitative analysis engine correlates the device status, material quantity, and environmental bias in multiple dimensions, such as the risk of detection sample distortion caused by accelerated material evaporation in high temperature environments.

[0046] The equipment deployment adjustment implements a hierarchical response strategy. First-level adjustment for local conflicts: migrate affected detection tasks to backup devices of the same type, with a time node delay of no more than 15% of the original plan. Second-level adjustment for regional resource shortages: start cross-tower segment device sharing, such as temporarily deploying a gas chromatograph from the B tower segment to the A tower segment, and synchronously adjusting the operator scheduling. Third-level adjustment for systemic crisis: dynamically reduce low-weight detection items, reducing the detection frequency of fourth-level priority parameters by 50%. The time node correction uses an elastic stretching algorithm, inserting a buffer detection window during the transition period of the rectification stage, allowing a ±8 minute floating time for key parameter detection.

[0047] The parameter and environmental bias table is constructed as a three-dimensional matrix data structure. The row dimension records the affected 56 quality parameter numbers, the column dimension labels 7 types of environmental interference factors (temperature, pressure, flow, component, liquid level, voltage, vibration), and the depth dimension stores time slice data (every 5 minutes a data page). The matrix cells fill in three types of values: environmental bias influence coefficient (0-1 standardized value), device state correction amount (time adjustment value in minutes), and material compensation factor (percentage of material to be replenished). For example, the cell [PUR-03, temperature, 11:30] may store the data group (0.75, +6.2, 12%), indicating that purity parameter 3 needs to be delayed for 6.2 minutes of detection due to temperature deviation and 12% of material needs to be replenished.

[0048] Generating distillation offset correction results involves four layers of logic. The base layer implements equipment reallocation: automatically issuing control instructions based on the deviation table, such as switching test task T007 from the faulty mass spectrometer MS-02 to the backup instrument MS-05. The scheduling layer adjusts time nodes: recalculates test timings and marks 34 key time correction points in the schedule (e.g., adjusting the boiling point test from 14:20 to 14:27). The compensation layer controls material replenishment: In conjunction with the DCS system, the feed valve opening is adjusted to ensure quantitative replenishment of shortfalls with an accuracy of ±0.5%. The fault tolerance layer activates emergency response plans: when tower pressure fluctuations exceed safety thresholds, rapid testing mode is automatically enabled (compressing the typical 20-minute test time to 12 minutes). The final correction results are output as incremental update packages, consisting of an equipment allocation change set, a time adjustment mapping table, and a material compensation list. These packages are updated to the central control system every 15 minutes.

[0049] The system establishes a deviation response traceability mechanism. Each correction operation records the decision-making basis data package, including the environmental snapshot at the time of triggering (instantaneous values ​​of 12 parameters such as temperature and pressure), equipment health status score, material remainder curve segment and intermediate results of deviation calculation. Historical correction cases are classified and stored by distillation stage, forming a decision knowledge base containing 2,700 typical scenarios. When a new deviation event occurs, the system automatically matches the correction plan of similar historical scenarios, and recommends the plan with a matching degree of more than 85% as the priority strategy. The real-time monitoring interface uses a three-color warning light to display the offset status of each tower section: the green light indicates that the deviation is controllable ( ), yellow light prompts attention ( ), red light requires immediate intervention ( or device health score < 60).

[0050] Example 4: See Figure 4The data storage module manages the massive amounts of data generated during the distillation process through a distributed architecture. Its core functions are storage performance optimization and load balancing. During the storage performance optimization implementation process, the system first establishes a storage port performance evaluation system. Each storage port (such as the SSD array on node A or the NVMe storage pool on node B) must regularly report six operational metrics: data throughput (MB / s), IOPS (input and output operations per second), response latency (ms), storage capacity utilization (%), error correction code trigger frequency (times / hour), and device temperature (°C). These metrics are normalized and synthesized into a storage performance baseline using a weighted ratio of 0.2:0.3:0.15:0.15:0.1:0.1. For example, during the monitoring period, a port records a throughput of 580 MB / s (normalized value 0.87), IOPS of 45,000 (0.92), latency of 1.8 ms (0.95), utilization of 65% (0.35), ECC triggers twice (0.9), and a temperature of 42°C (0.88). The baseline value is calculated as 0.87 × 0.2 + 0.92 × 0.3 + 0.95 × 0.15 + 0.35 × 0.15 + 0.9 × 0.1 + 0.88 × 0.1 = 0.826.

[0051] Cluster distillation information evaluation for edge computing nodes utilizes a multi-dimensional feature analysis method. The data reported by each node includes three core features: data processing energy efficiency (KB / W of data processed per watt of power consumption), the number of target data collection points, and the average packet size (KB). The system assigns weights of 0.5, 0.3, and 0.2 to these three features, respectively, and calculates the node evaluation value through weighted summation. For example, if node X records an energy efficiency value of 125KB / W (normalized to 0.89), 38 data collection points (0.76), and an average packet size of 24KB (0.82), its evaluation value is 0.89 × 0.5 + 0.76 × 0.3 + 0.82 × 0.2 = 0.837. The storage performance verification value is determined using an evaluation value mapping table with five predefined verification levels (A for values ​​above 0.9, B for values ​​between 0.8 and 0.9, and so on). For example, a node X evaluation value of 0.837 corresponds to a B verification value of 0.85.

[0052] The storage allocation decision-making process uses a matching priority algorithm. The system maintains a dynamic allocation matrix that records the matching status of each storage port and edge node. When new data needs to be stored, the matrix is ​​searched for available ports that meet a validation value ≥ a reference value, and candidate ports are sorted in ascending order of response latency. For example, when new data generated by node Y needs to be stored, the system detects three candidate ports: port C (validation value 0.88 / reference value 0.82 / latency 1.2ms), port D (0.85 / 0.80 / 1.5ms), and port E (0.90 / 0.85 / 2.0ms). Port C is prioritized.

[0053] Table 1: A fragment of a data allocation record.

[0054]

[0055] Load balancing is implemented using a dynamic partition rebalancing strategy. The system scans the load parameters of each data storage node every minute, including CPU core utilization (%), memory utilization (%), network bandwidth utilization (%), and disk queue depth (number of disks). These parameters are calculated as moving averages using a sliding window algorithm to eliminate interference from instantaneous fluctuations. When a node experiences CPU utilization > 85% and memory utilization > 80% for three consecutive cycles, a load migration process is triggered. The migration process follows a three-step strategy: first, non-real-time analytical data (such as historical process curves) is migrated to cold storage nodes; second, real-time monitoring data is transferred in shards, with each shard size controlled within the 50-100MB range; and finally, mirror copies are created for frequently accessed metadata. For example, if node Z is overloaded, the system migrates its 1200MB of historical detection data to the archive node, splits the 300MB of real-time data collected between nodes M and N, and creates a metadata mirror on node P.

[0056] The load balancing system of the distillation database management system includes a special scenario processing mechanism. When a tower section is detected to have entered the product extraction stage, it is predicted that the data volume in this area will surge, and resource reservation for adjacent nodes will be initiated in advance. The system maintains a capacity buffer pool, always keeping 15-20% of idle storage space for emergency deployment. The data distribution heat map is updated every 30 minutes, using a color gradient to display the load status of each node: blue (<50% load), green (50-75%), yellow (75-90%), and red (>90%). Load migration decisions are based on the trend of the heat map. For example, when a node continuously changes from green to yellow, preventive data diversion is initiated even if the threshold is not reached.

[0057] The data storage module implements a full-link monitoring system. Each data operation (write, read, migration) generates a tracking log, recording 12 metadata items, including operation type, data characteristics, source and target nodes, and duration. The anomaly detection engine analyzes log streams in real time, identifying unusual patterns such as frequent spikes in small data packets (which may indicate sensor failure) and regular latency fluctuations (which may reflect network congestion). If a port's response latency increases by 30% year-over-year, its verification value is automatically temporarily lowered by 0.05-0.1, reducing the amount of new data allocated until performance recovers. The storage health dashboard centrally displays key metrics: current storage cluster utilization (maintained within the ideal range of 65-75%), data distribution balance (variance value <0.1), and abnormal operation ratio (<0.5%).

[0058] The disaster recovery backup mechanism utilizes a three-tiered storage architecture. Real-time detection data is stored in a high-performance primary storage node (SSD / NVMe), retaining the last 72 hours of data. Interim data is transferred to a standard-performance secondary node (SAS hard disk array) for 30 days. Long-term archived data is compressed and stored in a tertiary node (tape library) for five years. Data migration tasks are automatically executed according to pre-set policies. For example, expired data is migrated from primary storage to the secondary node at 02:00 daily. Data verification is implemented during the backup process, using a CRC32 cyclic redundancy check to ensure integrity. If the verification fails, the backup automatically retries three times before issuing an alarm.

[0059] The storage module collaborates with other systems through standard interfaces. When the quality analysis module queries historical data, the storage module prioritizes returning results from the memory cache, and triggers a fast disk retrieval if the cache misses. The progress monitoring module uses an asynchronous buffering mechanism to write real-time data, first storing it in a high-speed buffer and then writing it in batches to persistent storage to balance IO pressure. All interface communications are encrypted and transmitted using the AES-256 algorithm to protect process data security. Communication keys are rotated and updated every 24 hours. Storage operation audit logs are retained for 365 days, recording the initiating module, operator, timestamp, and data processing volume of each access request to meet the GMP compliance requirements of the pharmaceutical industry.

[0060] Example 5: The quality verification module is activated during the final stage of the distillation process quality inspection, processing a filtered dataset of sub-parameter detection areas. This dataset contains the raw values ​​of quality parameters collected from each section of the distillation column during a specific inspection cycle, such as the purity percentage of the overhead extract, the heavy component content of the intermediate section, and the boiling point of the bottom residue. The system first performs a parameter classification operation: each measured parameter value is compared item by item with the process standard value, which is derived from the quality specification table in the process database. All qualified parameters are sorted in ascending numerical order to form a first quality sequence, and unqualified parameters are sorted in chronological order to form a second quality sequence.

[0061] After the sequence is constructed, the parameter association verification process is initiated. The system numerically rearranges the first mass sequence, repositioning parameters such as purity, boiling point, and density from smallest to largest. The second mass sequence maintains its timestamp-ordered structure. Association verification is performed on a parameter pair basis: two adjacent parameter values ​​are extracted from the first sequence as pairs to be verified, such as the combination of 93.1% purity and 89°C boiling point; similarly, two parameter values ​​with adjacent time periods are extracted from the second sequence, such as the combination of 1.8% heavy component content and 2.4 cP viscosity. The quality evaluation model uses association rules based on the process knowledge graph. For example, in a propanol distillation system, purity and boiling point are positively correlated: an increase in purity is generally accompanied by a decrease in boiling point. When the trend of the parameter pair to be verified conforms to the process association rules, the system marks the parameter pair as associated; if the direction of change is contradictory, it is marked as unassociated.

[0062] The sliding window counting method is used to count the number of associations. During the verification process, the system dynamically records two types of data: the number of parameter pairs that meet the process association rules, and the number of parameter pairs that violate them. For the first quality series, a periodic association rate is generated after every ten parameter pairs are verified. For the second quality series, statistics are calculated by time block, with an association index output every 30-minute testing period. Independent parameters not involved in the association are recorded separately. For example, if valid data for the water content parameter of a tower section is not obtained due to a sensor failure, the system will list it as a non-association parameter set and annotate the cause of the anomaly.

[0063] The calculation of the comprehensive quality value incorporates multiple assessment factors. The core calculation is based on three dimensions: the first is the strength of parameter correlation, measured by the proportion of correlated parameter pairs to the total verified pairs. For example, if 68 out of 80 parameter pairs verified in a given testing cycle are successfully correlated, the correlation strength is 85%. The second dimension is the impact factor of uncorrelated parameters, which is weighted differently depending on the parameter type: a weight of 0.8 for uncorrelated key parameters like purity, and 0.2 for uncorrelated minor parameters like chromaticity. The third dimension is the dispersion of parameter values, calculating the difference between the maximum and minimum values ​​in the first series. The final comprehensive quality value is generated using a three-tiered weighting algorithm: 60% for correlation strength, 30% for the influence of uncorrelated parameters, and 10% for dispersion. The result is converted to a 100-point quality score. For example, a correlation strength of 85%, a non-correlated influence factor of 0.15, and a dispersion of 12% result in a comprehensive quality score of 78.3.

[0064] The quality verification process includes an exception handling mechanism. If a parameter is detected as not correlated for three consecutive verifications, the system automatically triggers a traceability process: it retrieves the parameter's historical data curve, analyzes the calibration records of the testing equipment, and verifies the environmental sensor values ​​from the same period. If the tower temperature sensor fluctuates by ±3°C during testing, the system marks this non-correlation as due to environmental interference, reducing its weight in the overall quality calculation. For parameter anomalies caused by equipment failure, the module automatically initiates an equipment maintenance work order and freezes the verification results for that parameter.

[0065] The verification results are output in the form of graded reports. The basic report contains the comprehensive quality value and zone ranking of each sub-parameter detection area; the detailed report records the verification status of all parameter pairs, the list of non-related parameters and abnormal descriptions; the trend report shows the quality value change curve for six consecutive detection cycles. All reports are transmitted to the central quality dashboard through a secure interface and displayed in real time on the large screen in the distillation control room using a three-color light system: a green light indicates a comprehensive quality value ≥85 points, a yellow light indicates 70-85 points, and a red light indicates less than 70 points. When a red light warning appears, the system automatically associates the deviation response module to trigger the distillation offset correction process.

[0066] A data traceability system runs throughout the entire verification process. The complete path of each parameter, from initial collection to final verification, is recorded in a blockchain log, including timestamps, operator (or automated system identification), processing stage markers, and other elements. Historical verification data is archived and stored by distillation batch, supporting multi-dimensional retrieval by column section number, test time, parameter type, and other factors. If an operator questions a verification result, a playback function allows them to step through the application of parameter classification rules, the execution of associated verification logic, and the weighted calculation process.

[0067] The quality verification module establishes a feedback loop with the front-end detection system. When the overall quality value continuously falls below the threshold, the module sends parameter weight adjustment suggestions to the quality analysis module, such as increasing the priority weight of the purity parameter. When a specific type of parameter is frequently found to be uncorrelated, the module proposes to the data acquisition module to increase the detection frequency of that parameter. All feedback suggestions are accompanied by detailed analysis, forming a continuously optimized quality monitoring ecosystem. The verification results database generates a quality characteristic map every quarter, noting the parameter correlation patterns and abnormal high-incidence areas of each tower section.

[0068] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0069] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A quality detection system for a propanol distillation process, characterized in that: The system comprises: The data acquisition module is used to obtain the initial operating data of the distillation process. Based on the material characteristics and equipment parameters, it calculates the equipment operating time and task requirements of the tower section, matches the equipment available time with the task requirements, calculates the difference in operator man-hour allocation, compares the equipment and personnel availability, and generates the tower section equipment allocation results. A quality analysis module calculates the detection time and parameter interval of key quality parameters based on the tower section equipment allocation result, sorts the instrument working hours and operator working hours priority call order, calculates the parameter resource utilization rate and parameter interval weight, and generates the key quality parameter sorting result; A constraint evaluation module calculates the material supply cycle and equipment operating time based on the ranking results of the key quality parameters, calibrates equipment allocation time conflicts and resource shortage tasks, adjusts parameter detection intervals, and generates a parameter constraint adjustment parameter set; A progress monitoring module adjusts the parameter set based on the parameter constraints, calculates the distribution difference between the distillation process time nodes and parameters, matches resource utilization with equipment available time, calculates the adjustment range of parameters and time nodes, reorganizes the parameter time and equipment resource distribution, and generates a distillation schedule; The deviation response module monitors the environmental changes and equipment operating status of the distillation process based on the distillation schedule, calculates the parameter time and environmental change deviation, counts the equipment working hours and material remaining amount, adjusts the equipment deployment and time nodes, and generates the distillation deviation correction result.

2. The quality detection system for propanol distillation process according to claim 1, characterized in that: When the data acquisition module obtains the initial operation data of the distillation process, it collects the task requirements and equipment operating time of the partitioned tower sections, comprehensively calculates the available time of the multi-partition equipment and the matching degree with the task requirements, and calculates the deviation between the total equipment time and the task requirements to generate a set of equipment time deviations; Analyze the equipment time deviation set, adjust the equipment and operator work time distribution according to the availability of operator resources, and establish an operator and equipment work time adjustment set; According to the operator and equipment working hour adjustment set, equipment and operator allocation comparison is performed to obtain the tower section equipment allocation result.

3. The quality detection system for propanol distillation process according to claim 2, characterized in that: The quality analysis module analyzes the resource allocation of each key quality parameter based on the tower section equipment allocation results, calculates the starting detection time, and uses the resource optimization algorithm to predict the optimal parameter sequence to generate a start schedule for each key quality parameter; Based on the start-up schedule of each key quality parameter, a priority ranking model is used to prioritize the instrument and operator resources, and the time allocation of the instrument and operator is adjusted according to the urgency of the parameter to establish a parameter priority list; By using the parameter priority list and combining it with the interval requirements in actual parameter detection, the parameter resource utilization and parameter interval weight of the key quality parameters are calculated to generate a key quality parameter ranking result.

4. The quality detection system for propanol distillation process according to claim 3, characterized in that: The constraint evaluation module extracts the start time and material cycle of each parameter from the key quality parameter sorting results, analyzes equipment usage, applies a resource allocation algorithm to determine the resource requirements of each parameter, and generates a resource requirement analysis result; Based on the resource demand analysis results, mark all time conflicts between resource supply and demand as well as conflicts and resource shortages occurring in equipment allocation, and create a conflict and shortage index table; Using the conflict and shortage index table, the detection interval of each parameter is recalculated, the parameter detection plan is optimized, and a parameter constraint adjustment parameter set is generated.

5. The quality detection system for propanol distillation process according to claim 4, characterized in that: The progress monitoring module extracts the distribution differences between the time nodes and parameters of the distillation process from the parameter constraint adjustment parameter set, analyzes the parameter detection sequence and resource utilization efficiency based on the comparison of parameter time and resource distribution, and generates a distribution table of time nodes and parameter differences; By using the time node and parameter difference distribution table, resource utilization and device available time are matched, and based on the matching analysis of resource allocation and parameter requirements, time conflict and resource shortage parameters are calibrated to create resource conflict and shortage parameter results; Based on the resource conflicts and insufficient parameter results, the adjustment range and time node correction values ​​of parameter detection are calculated, the parameter time distribution is optimized, and a distillation schedule is generated.

6. The quality detection system for propanol distillation process according to claim 5, characterized in that: The deviation response module extracts monitoring data from the distillation schedule, including changes in the distillation process environment and equipment operating status, combines parameter time nodes, uses data analysis methods to determine the impact of the environment and equipment status on the distillation schedule, and generates environmental and equipment status analysis results; Using the results of the environment and equipment status analysis, calculate the deviation between parameter time and environmental changes, and count the working hours and material remaining of each device. Through quantitative analysis methods, calibrate the equipment deployment and time nodes that need to be adjusted, and create a parameter and environment deviation table; Based on the parameter and environmental deviation table, the equipment allocation and time nodes are adjusted to match the actual distillation environment changes, the parameter allocation is optimized, and the distillation offset correction result is generated.

7. The quality detection system for propanol distillation process according to claim 6, characterized in that: The system also includes a data storage module for synchronously retrieving the operating data of each storage port through the distillation database storage port for analysis, and performing distributed storage management of the cluster distillation information; The data storage module processes the operating data of each storage port to obtain a storage performance benchmark value of each storage port, wherein the storage performance benchmark value of each storage port is used to comprehensively quantify the storage capacity utilization of each storage port; Counting cluster distillation information data of each edge computing node, wherein the cluster distillation information data of each edge computing node includes a data processing energy efficiency characterization value of each edge computing node and the number of target receiving data collection points; Comprehensively analyze the cluster distillation information data of each edge computing node to obtain a cluster distillation information evaluation value of each edge computing node, and the cluster distillation information evaluation value of each edge computing node is used to comprehensively quantify the comprehensive performance of each edge computing node; The storage performance verification value of each edge computing node is obtained by matching the cluster distillation information evaluation value of each edge computing node; The storage performance benchmark value of each storage port is compared with the storage performance verification value of each edge computing node to obtain the target storage port of each edge computing node, and the cluster distillation information of each edge computing node is transmitted to the corresponding target storage port.

8. The quality detection system for propanol distillation process according to claim 7, characterized in that: The data storage module processes and obtains a load evaluation value of each data storage node based on the load parameters of each data storage node, and performs load balancing configuration on each data storage node according to the load evaluation value of each data storage node; The load balancing configuration includes counting each data storage node in the distillation database management system and obtaining the load parameters of each data storage node to perform load balancing processing.

9. The quality detection system for propanol distillation process according to claim 8, characterized in that: The system further includes a quality verification module for extracting all quality parameter values ​​of the remaining sub-parameter detection areas after deletion, comparing the quality parameter values ​​with corresponding quality standard parameter values, dividing all quality parameter values ​​according to the comparison results, and constructing a first quality sequence and a second quality sequence according to the division results; numerically sorting all quality parameter values ​​in the first mass number sequence and the second mass number sequence, using every two quality parameter values ​​in the first mass number sequence or the second mass number sequence as quality parameters to be associated, verifying whether the quality parameters to be associated are associated based on a quality evaluation model, and counting the number of associations based on the verification results; The comprehensive quality value of the distillation process is determined according to the associated quantity and the unassociated quality parameter value.

10. A quality detection method for a propanol distillation process, characterized in that: The invention comprises all modules and method processes of the quality detection system for the propanol distillation process as described in any one of claims 1 to 9.

Images