Intelligent control system for polyvinyl chloride polymerization reaction

Abstract

The invention relates to the technical field of industrial automation control, in particular to a polyvinyl chloride polymerization reaction intelligent control system which comprises a trend recognition module, a coupling analysis module, a target generation module, a deviation evaluation module and a control execution module. According to the method, combined analysis is carried out on the change directions of the polyvinyl chloride chain growth rate and the temperature rise amplitude, dependence on a single parameter threshold value is exceeded, early recognition of potential risks such as reaction acceleration or delay is achieved, external meteorological changes are introduced to serve as key variables, the coupling influence of the external meteorological changes and the internal reaction state is quantified, and the recognition accuracy is improved. The temperature control is converted from passive response to predication-based prospective intervention, the regulation and control target constructed on this basis not only dynamically associates the current working condition with the environmental disturbance, but also can be adapted according to the real-time capability of the cooling device, and before the specific regulation is executed, the temperature control is more accurate. The expected response of the control action is deduced by referencing similar historical data, and possible offset between the adjustment result and the target is quantified.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of industrial automation control, and in particular to a polyvinyl chloride polymerization reaction intelligent control system. BACKGROUND

[0002] The technical field of industrial automation control mainly relates to accurate operation and process regulation of production equipment in an industrial production process through automatic technical means, including field signal acquisition, control algorithm design, actuator regulation and system integration.

[0003] The polyvinyl chloride polymerization reaction intelligent control system refers to an operation and regulation system for a polymerization reaction section in a polyvinyl chloride production process, which is used for real-time monitoring and dynamic regulation of key parameters such as temperature, pressure, stirring speed, feeding rate and reaction time in a polymerization kettle.

[0004] The prior art mainly relies on real-time monitoring and dynamic regulation of independent key parameters such as temperature and pressure in the polymerization kettle, and the regulation logic is based on threshold judgment of the absolute value of a single parameter. This method has inherent hysteresis when the reaction trend is abnormal at the beginning, for example, the control system will only respond when the kettle temperature has deviated significantly from the set value, at which time the best intervention opportunity has been missed. At the same time, since the running model does not consider external environmental factors, when facing external disturbances such as drastic changes in environmental temperature, the system cannot predict the superimposed influence on the reaction heat release process, so that its regulation action can only be used as a post-repair, and its control strategy is usually preset and fixed, lacking a mechanism for dynamically adjusting the regulation strength and method according to the subtle differences of the current reaction state and the future trend prediction, which is easy to cause control overshoot or insufficient response, and further affects the uniformity of the molecular weight distribution of the polyvinyl chloride product and the overall stability of the production process. SUMMARY

[0005] The purpose of the application is to solve the shortcomings in the prior art and provide a polyvinyl chloride polymerization reaction intelligent control system.

[0006] In order to achieve the above purpose, the application adopts the following technical scheme: a polyvinyl chloride polymerization reaction intelligent control system comprises: A trend identification module obtains the polyvinyl chloride chain growth rate and the corresponding reaction region temperature rise amplitude in a specified period during the polyvinyl chloride polymerization reaction process, judges the change direction of the two in the continuous time and performs cross analysis as a combined feature, and assigns a polymerization reaction state label according to the analysis result. A coupling analysis module obtains the polymerization reaction state label and meteorological data of the corresponding time period, analyzes the response relationship between the reaction process temperature change and the external meteorological factors, and outputs a temperature trend analysis result. a target generation module, configured to construct a temperature control target based on the polymerization reaction state label and the temperature trend analysis result, with a PVC chain growth rate and a corresponding reaction region temperature rise amplitude as a control subject; a deviation evaluation module, configured to take a chain growth rate response record corresponding to the temperature control target in all running data as a comparison benchmark, to estimate a growth rate generated by the current cooling control, to compare the target growth rate, and to output a chain growth deviation prediction result; a control execution module, configured to determine an injection cooling and heat exchange adjustment action in combination with the temperature control target and the chain growth deviation prediction result, and to generate a temperature control result.

[0007] As a further scheme of the present application, the polymerization reaction state label is specifically a chain growth rate change direction, a temperature rise amplitude change direction, and a trend combination feature, the temperature trend analysis result includes a temperature change sensitive section, a meteorological factor response strength, and a stage temperature rise probability, the temperature control target is specifically a target chain growth rate interval, a target temperature rise amplitude range, and an adjustment parameter boundary, the chain growth deviation prediction result includes a growth rate prediction value and a target deviation degree, and the temperature control result is specifically an injection cooling adjustment instruction, a heat exchange adjustment instruction, and an execution adjustment sequence.

[0008] As a further scheme of the present application, the trend identification module includes: a trend extraction submodule, configured to obtain a chain growth rate and a corresponding reaction region temperature rise amplitude in a specified period in a PVC polymerization reaction process, to calculate a change direction in each adjacent time period, to respectively assign a chain growth rate change direction label and a temperature rise amplitude change direction label according to a positive or negative value of a difference, and to generate a trend direction label; a feature pairing submodule, configured to combine the chain growth rate change direction label and the temperature rise amplitude change direction label into a trend pairing feature based on the trend direction label, to obtain a polymerization stage temperature control target state corresponding to a time period of the trend pairing feature, to judge a deviation direction between the pairing feature and the temperature control state, to screen an associated feature pairing, and to generate a state deviation trend combination; a state assignment submodule, configured to divide a polymerization reaction state type corresponding to the combination trend based on the state deviation trend combination and to assign a corresponding label, and to generate a polymerization reaction state label.

[0009] As a further scheme of the present application, the coupling analysis module includes: a label reading submodule, configured to obtain the polymerization reaction state label and to collect meteorological data of a corresponding time period, to extract a temperature change parameter from the meteorological data, and to correspondingly pair the state label in a time synchronization mode, and to generate a state-meteorological pairing group; The meteorological coupling submodule is configured to, based on the state-meteorological pair group, construct a feature pair of a state label and a temperature change parameter, input into a Bayesian regression model, calculate a joint probability distribution of a temperature change value under the state label, establish a response relationship of meteorological change on temperature evolution of a reaction process, and generate temperature response probability distribution data; The trend identification submodule is configured to, based on the temperature response probability distribution data, judge whether a temperature value in a current polymerization reaction process is in a probability region sensitive to temperature change, identify a trend region attribution according to a probability distribution threshold, and output a temperature trend analysis result.

[0010] As a further scheme of the present application, the target generation module comprises: The parameter adaptation submodule is configured to acquire the polymerization reaction state label and the temperature trend analysis result, call polymerization vinyl chloride chain growth rate and reaction region temperature rise amplitude data in a corresponding time period, extract a set of regulation input parameters according to a state type and a trend region attribution, and output the set of regulation input parameters. The scheme screening submodule is configured to call a preset cooling regulation plan in a parameter library, match a response feature corresponding to a state label and trend region information in a corresponding stage, judge an adaptation degree between each plan and the set of regulation input parameters, and generate a matching cooling plan group. The target construction submodule is configured to combine real-time cooling device running capability parameters, limit a chain growth rate regulation value and a temperature rise amplitude regulation amplitude in a target interval according to the matching cooling plan group, form a next period regulation index combination, and output a temperature control target.

[0011] As a further scheme of the present application, the deviation evaluation module comprises: The response recording submodule is configured to call the temperature control target, refer to cooling control data corresponding to a current regulation type and the temperature control target in all running data, extract a chain growth rate response record in the cooling control data as a comparison data source, and generate a chain growth rate benchmark record set. The growth estimation submodule is configured to, based on the chain growth rate benchmark record set, call a regulation content of the current temperature control target as an input, deduce a chain growth rate change trend in a data sequence, estimate a response value generated by the current cooling control as a chain growth rate prediction value, and output the chain growth rate prediction value. The offset evaluation submodule is configured to call a target growth rate set in the temperature control target, calculate a residual sequence distribution of the chain growth rate prediction value, quantify an offset degree, and output a chain growth rate offset prediction result.

[0012] As a further scheme of the present application, the control execution module comprises: The action setting sub-module obtains the temperature control target and the chain growth offset prediction result, calls the regulation interval and the predicted offset direction, calculates the injection cooling and heat exchange adjustment amplitude and the sorting order for each cooling control point, determines the sequence and target amplitude of the adjustment operation, and generates an adjustment action execution sequence; The instruction generation sub-module constructs the injection cooling instruction and the heat exchange instruction of the cooling control point according to the sequence of the adjustment action execution sequence, encodes the execution position and the adjustment amplitude of the instruction respectively, forms a format recognizable by the control end, and generates an injection cooling and heat exchange operation instruction set; The control feedback sub-module synchronously transmits the injection cooling and heat exchange operation instruction set to the corresponding execution unit of the control end, collects and records the real-time response result of the regulation and control process, and outputs the temperature control result.

[0013] Compared with the prior art, the application has the advantages and positive effects that: In the application, the change direction of the polyvinyl chloride chain growth rate and the temperature rise amplitude is combined and analyzed, which transcends the dependence on a single parameter threshold, realizes early identification of potential risks such as reaction acceleration or retardation, further introduces external meteorological changes as a key variable, quantifies the coupling influence of the internal reaction state, changes the temperature control from passive response to prediction-based proactive intervention, and builds a regulation target that not only dynamically associates the current working condition and environmental disturbance but also can be adapted according to the real-time capacity of the cooling device. Before the specific adjustment is executed, the expected response of the control action is deduced by referring to historical similar data, the possible deviation between the adjustment result and the target is estimated and quantified, and finally the execution sequence and the adjustment amplitude of the injection cooling and heat exchange and other adjustment methods are finely set according to the deviation prediction, which effectively avoids the problems of excessive or insufficient adjustment, ensures the smooth operation of the reaction process and the consistency of the product quality. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The system flowchart of the application; Figure 2 The flowchart of the trend identification module of the application; Figure 3 The flowchart of the coupling analysis module of the application; Figure 4 The flowchart of the target generation module of the application; Figure 5 The flowchart of the deviation evaluation module of the application; Figure 6 The flowchart of the control execution module of the application. DETAILED DESCRIPTION

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] Please see Figure 1 A smart control system for polyvinyl chloride polymerization reaction includes: The trend recognition module obtains the PVC chain growth rate and the corresponding temperature rise of the reaction region during a specified period in the PVC polymerization reaction, determines the direction of change of the two over a continuous period of time and uses them as combined features for cross-analysis, and assigns polymerization reaction state labels based on the analysis results. The coupling analysis module acquires the polymerization reaction status label and the corresponding meteorological data for the time period, analyzes the response relationship between temperature changes during the reaction process and external meteorological factors, and outputs temperature trend analysis results. The target generation module, based on the polymerization reaction status label and temperature trend analysis results, uses the PVC chain growth rate and the corresponding temperature rise in the reaction region as the main control factors to construct a temperature control target. The deviation assessment module uses the chain growth rate response record corresponding to the temperature control target in all operating data as a comparison benchmark to estimate the growth rate generated by this cooling control, compare it with the target growth rate, and output the chain growth deviation prediction result. The control execution module, combining the temperature control target and the chain growth offset prediction results, determines the adjustment actions for cooling injection and heat exchange, and generates temperature control results; The polymerization reaction status labels specifically include the direction of chain growth rate change, the direction of temperature rise change, and trend combination characteristics. The temperature trend analysis results include temperature change sensitive sections, meteorological factor response intensity, and staged temperature rise probability. The temperature control targets specifically include the target chain growth rate range, the target temperature rise range, and the adjustment parameter boundaries. The chain growth offset prediction results include the predicted growth rate value and the target offset degree. The temperature control results specifically indicate the cold adjustment command, heat exchange adjustment command, and the order of execution of the adjustment.

[0017] Please see Figure 2 The trend recognition module includes: The trend extraction submodule obtains the chain growth rate and the corresponding temperature rise of the reaction region within a specified period during the polyvinyl chloride polymerization reaction, calculates the direction of change of each time period, and assigns a label to the direction of change of chain growth rate and the direction of change of temperature rise according to the positive or negative value of the difference, thereby generating trend direction labels. Obtaining the chain growth rate and the corresponding temperature rise amplitude in a specified period during the polyvinyl chloride polymerization process, this process starts from the online Raman spectrum analyzer deployed in the polymerization kettle to retrieve the continuous monitoring record of the last one hour, and divide this time window into six time periods at intervals of ten minutes, obtain the numerical sequence of the chain growth rate [0.48, 0.52, 0.51, 0.58, 0.60, 0.59], the unit is kilogram per mole per second, at the same time, the temperature rise amplitude in the same period is collected from the distributed optical fiber temperature sensor array covering the jacket of the reaction kettle, after calculating the arithmetic mean value of the data of three key monitoring points, the corresponding temperature rise amplitude numerical sequence is obtained [0.19, 0.23, 0.22, 0.29, 0.30, 0.28], the unit is Celsius, then calculate the change direction of each adjacent time period, this calculation is completed by subtracting the value of the next time period from the value of the previous time period, then according to the positive and negative of the difference result, respectively, the chain growth rate change direction and the temperature rise amplitude change direction are labeled, here the labeling standard is: the difference result is positive, labeled as "positive"; the difference result is negative, labeled as "negative"; the difference result is zero, labeled as "stable", based on this standard, the change direction label sequence of the chain growth rate is determined as [positive, negative, positive, positive, negative], the change direction label sequence of the temperature rise amplitude is determined as [positive, negative, positive, positive, negative], and finally the trend direction label composed of the two groups of direction labels is generated.

[0018] The feature pairing submodule combines the chain growth rate change direction label and the temperature rise amplitude change direction label into a trend pairing feature based on the trend direction label, obtains the polymerization stage temperature control target state corresponding to the time period of the trend pairing feature, judges the deviation direction between the pairing feature and the temperature control state, selects the associated feature pair, and generates a state deviation trend combination. Based on the trend direction label, that is, the chain growth rate change direction label sequence and the temperature rise amplitude change direction label sequence generated in the previous step, the labels at the same time position in the two sequences are combined one by one, thereby forming a trend pairing feature of five consecutive time periods, the specific content of which is [(positive, positive), (negative, negative), (positive, positive), (positive, positive), (negative, negative)], and during this period, the temperature control target state of the polymerization stage corresponding to the five time periods in the production process procedure is retrieved from the manufacturing execution information archive, and the state sequence is [constant temperature maintenance, constant temperature maintenance, programmed temperature rise, programmed temperature rise, constant temperature maintenance]. Next, the deviation direction between the pairing feature and the temperature control state is judged, and the basis for the judgment is a set of solidified logic rules, the core content of which is that in the "constant temperature maintenance" state, the theoretically ideal pairing feature is (stable, stable), and any combination other than "stable" is judged as deviating, wherein the combination of (positive, positive) is defined as "advance deviation", and the combination of (negative, negative) is defined as "lag deviation"; in the "programmed temperature rise" state, the theoretically ideal pairing feature is (positive, positive), and other combinations are judged as deviating, and then the associated feature pairing with deviation is screened according to this basis, and finally the state offset trend combination is generated.

[0019] The state allocation submodule divides the type of the polymerization reaction state corresponding to the combination trend and assigns a corresponding label based on the state offset trend combination, and generates a polymerization reaction state label. Based on the state offset trend combination, which is [(time period one, (positive, positive), advance deviation), (time period two, (negative, negative), lag deviation), (time period five, (negative, negative), lag deviation)], the type of the polymerization reaction state corresponding to the combination trend is divided and a corresponding label is assigned, and the division here is based on a rule knowledge base, which is based on statistical clustering analysis of more than ten thousand batches of historical polymerization reaction data. For example, one of the rules clearly states that when the temperature control target is "constant temperature maintenance" and the deviation direction is "advance deviation", the type of the polymerization reaction state is defined as "reaction acceleration risk"; when the temperature control target is "constant temperature maintenance" and the deviation direction is "lag deviation", the type of the state is defined as "reaction retardation risk". Based on these established rules, the state offset trend combination screened out in the previous step is matched and assigned one by one, and for the combination of time period one, it is assigned a "reaction acceleration risk" label, and for the combinations of time period two and time period five, they are both assigned a "reaction retardation risk" label, thereby generating a clear polymerization reaction state label for each deviating time period.

[0020] Please refer to Figure 3 , the coupling analysis module includes: The label reading submodule acquires the polymerization state label and collects meteorological data of the corresponding time period, extracts the air temperature change parameter from the meteorological data, and generates a state-meteorological pairing group in a time synchronization manner. The label reading submodule acquires the polymerization state label and collects meteorological data of the corresponding time period, extracts the air temperature change parameter from the meteorological data, and generates a state-meteorological pairing group in a time synchronization manner.

[0021] The weather coupling submodule constructs a feature pair of the state label and the air temperature change parameter based on the state-meteorological pairing group, inputs it into a Bayesian regression model, calculates the joint probability distribution of the air temperature change on the reaction temperature change value under the condition of the state label, establishes the response relationship of the meteorological change on the temperature evolution of the reaction process, and generates temperature response probability distribution data. Based on the state-meteorological pairing group, a feature pair of the state label and the air temperature change parameter is constructed. This process first performs quantification and normalization processing on the non-numerical state label "reaction acceleration risk". The quantification standard is pre-set according to the statistical correlation strength of different state labels with the reaction heat release rate in the historical database. For example, "reaction acceleration risk" is quantified as an original score S equal to 1.5 because it has the highest positive correlation, and then normalized to eliminate the dimension. The normalization method is to divide the original score by a reference value , which is set to 1.5. The setting basis is to select the maximum absolute value of all possible state label quantization values, so that the value range of the normalized state variable is constrained between [-1, 1]. The of the current state is calculated as 1.5 divided by 1.5, which is 1.0. This feature pair is input into an optimized Bayesian regression model to calculate the joint probability distribution of the state label and the air temperature change on the reaction temperature change value. The model expression is: ; In this expression, each term has a clear physical meaning and setting basis: The expected value of the change in the temperature of the next cycle, in degrees Celsius (°C), is the final prediction target of this module. is the base temperature change intercept, with a value of 0.08, in degrees Celsius (°C), which represents the natural rate of temperature increase in the reactor caused only by the basic exothermicity of the polymerization reaction itself under ideal reference conditions where the reaction state is most stable (i.e. ) and the external air temperature does not change (i.e. ). The setting basis is the statistical average calculation of the actual temperature rise rate under this ideal reference state in more than 5,000 batches of historical data. is the state influence coefficient, with a value of 0.55, in degrees Celsius (°C), which represents the temperature change caused by a unit of normalized state variable. Since is dimensionless, has the same dimension as temperature, which quantifies the independent contribution of the internal reaction state (such as reaction acceleration or retardation) to the temperature change. The setting basis is the result of historical data regression analysis. The more significant the influence of the internal reaction state on temperature loss of control in history, the larger the absolute value of this coefficient. is the normalized state variable calculated in the previous paragraph, with a value of 1.0, which is dimensionless. It converts the qualitative label "reaction acceleration risk" from paragraph 3 into a standardized numerical value for calculation in the model. is the meteorological main effect coefficient, with a value of 1.10, which is dimensionless. It represents the multiple of the change in the reactor temperature caused by a change of one degree Celsius in the external air temperature, reflecting the overall sensitivity of the reactor to environmental temperature and the heat insulation performance. The setting basis is the result of historical data regression analysis. The larger the overall heat transfer coefficient of the reactor and the worse the heat insulation performance, the larger the value of this coefficient. is the meteorological data from paragraph 4, i.e., the measured change in the external air temperature, with a value of 0.4, in degrees Celsius (°C). is the interaction coefficient, with a value of 0.50, which is dimensionless. It is used to quantify the coupling effect between the internal reaction state and the external temperature change, i.e., whether the influence of the external temperature change on the reactor temperature will be amplified or reduced when the reaction state itself is unstable. The setting basis also comes from historical data regression analysis. If cases of simultaneous deterioration of the reaction state and extreme weather leading to dramatic temperature changes frequently occur in history, the absolute value of this coefficient will deviate significantly from zero. The calculation process of the current value into the formula is: ; the calculation result shows that the expected value of the predicted temperature change is 1.27 degrees Celsius, and the model gives that this expected value follows a normal distribution with a mean of 1.27 degrees Celsius and a standard deviation of 0.16 degrees Celsius, which is the generated temperature response probability distribution data.

[0022] The trend identification submodule judges whether the temperature value in the current polymerization reaction process is in a probability region sensitive to air temperature change based on temperature response probability distribution data, identifies trend region attribution based on a probability distribution threshold, and outputs a temperature trend analysis result. Based on temperature response probability distribution data, i.e., a normal distribution with a mean of 1.27 degrees Celsius and a standard deviation of 0.16 degrees Celsius, it is judged whether the temperature value in the current polymerization reaction process is in a probability region sensitive to air temperature change. The judgment here is based on a probability distribution threshold, which is set to be a predicted temperature change expectation value of more than 0.80 degrees Celsius. The setting of this threshold is based on a retrospective analysis of historical data. The statistical results show that when the predicted temperature rise value is more than 0.80 degrees Celsius, there are 98% of cases that eventually trigger cooling regulation action within the next thirty minutes. The current calculated expectation value of 1.27 degrees Celsius is greater than the threshold of 0.80 degrees Celsius. Therefore, the current polymerization reaction process is judged to be in a probability region highly sensitive to air temperature change. Based on this judgment result, the trend region attribution of the current reaction is identified as "high-risk external disturbance dominated warming", and this temperature trend analysis result is output.

[0023] Please refer to Figure 4 The target generation module includes: The parameter adaptation submodule acquires the polymerization reaction state label and the temperature trend analysis result, calls the polyvinyl chloride chain growth rate and the reaction region temperature rise amplitude data of the corresponding time period, extracts the set of regulation and control input parameters based on the state type and the trend region attribution; The polymerization reaction state label "reaction acceleration risk" and the temperature trend analysis result "high-risk external disturbance dominated warming" are acquired, and the measured values of the polyvinyl chloride chain growth rate and the reaction region temperature rise amplitude that completely correspond to the current time period are called from the real-time database. The values are 0.68 kg per mole per second and 0.39 degrees Celsius, respectively. Based on the two composite conditions of state label and trend attribution, an accurate query is performed in the historical production database to retrieve all historical working condition records that simultaneously satisfy the two conditions. From these records, the numerical values of all regulation and control input parameters used within fifteen minutes after their occurrence and before the successful implementation of control intervention are extracted to form a set containing dozens of historical parameter states before successful intervention. This set is the set of regulation and control input parameters.

[0024] The scheme screening submodule calls the preset cooling regulation plan in the parameter library, matches the response characteristics of the state label corresponding stage and the trend region information, judges the adaptation degree between each plan and the set of regulation and control input parameters, and generates a matched cooling plan group. The preset cooling adjustment plan in the parameter library is called, and more than 20 standard operation sequences marked as "plan-A level", "plan-B level", etc. are stored in the library. Then, the state label "reaction acceleration risk" and the trend area information "high risk external disturbance dominant type of temperature rise" are matched as indexes to evaluate the adaptation degree of each pre-designed plan. The adaptation degree evaluation process is as follows: first, the arithmetic mean vector of all parameters in the control input parameter set obtained in the previous step is calculated; then, for each pre-designed plan, the theoretical control effect vector is calculated, and the Euclidean distance between the theoretical effect vector and the parameter set average vector is calculated. The smaller the distance, the higher the adaptation degree. The filtering standard is that the distance value must be less than the preset adaptation threshold 1.25. The threshold is set according to the 80th percentile of the distance between the selected scheme and the current process parameter set in all historical successful control cases. All plans with a calculated distance less than 1.25 are selected to generate a matching cooling plan group.

[0025] The target construction submodule limits the target interval of the chain growth rate adjustment value and the temperature rise amplitude adjustment amplitude according to the matching cooling plan group, forms a next cycle control index combination, and outputs a temperature control target in combination with the real-time cooling device running capability parameters. The real-time cooling device running capability parameters obtained from the device monitoring service are combined, for example, the actual maximum output flow of the main cooling water pump is obtained by reading the ultrasonic flowmeter data, which is 96% of the rated value. According to the matching cooling plan group generated in the previous step (assuming that "plan-A level" is selected), the target interval of the chain growth rate adjustment value and the temperature rise amplitude adjustment amplitude is limited. The ideal chain growth rate adjustment target set in "plan-A level" is -0.12 kg per mole per second, and the ideal temperature rise amplitude adjustment target is -0.60 degrees Celsius. Considering that the real-time running capability is 96% of the rated value, the two ideal adjustment target values are multiplied by 0.96 respectively to serve as the lower limit of the adjustment target interval, i.e. the chain growth rate adjustment lower limit is -0.1152, and the temperature rise amplitude adjustment lower limit is -0.576. At the same time, the upper limit of the adjustment is set, which is based on the maximum allowed adjustment amount specified in the process specification to prevent excessive inhibition of the reaction, i.e. no more than 110% of the ideal value. Thus, the chain growth rate adjustment upper limit is calculated as -0.132, and the temperature rise amplitude adjustment upper limit is calculated as -0.66. Finally, the next cycle control index combination is formed, and the temperature control target is output.

[0026] Please refer to Figure 5 The deviation evaluation module includes: The response record submodule calls the temperature control target, references the cooling control data corresponding to the current control type and the temperature control target in all running data, extracts the chain growth rate response record in the cooling control data as the comparison data source, and generates a chain growth rate benchmark record set. The temperature control target, i.e., the chain growth rate adjustment target interval [-0.132, -0.1152] and the temperature rise amplitude adjustment target interval [-0.66, -0.576], is called, and all historical cooling control data corresponding to the current control type and the current temperature control target in the entire historical running data archive is referenced as the search condition, and the specific screening criteria are that the minimum value and the maximum value of the control target interval in the historical record fall within the range of plus or minus 5% of the minimum value and the maximum value of the current target interval. From all the historical cooling control events screened out through this standard, the actual response change data of the chain growth rate within 15 minutes after the execution of the control instruction is specially extracted, and these data sequences collectively constitute a comparison data source, and finally a chain growth rate benchmark record set is generated.

[0027] The growth estimation submodule calls the adjustment content of the current temperature control target as input, deduces the chain growth rate change trend in the data sequence, estimates the response value generated by this cooling control as the chain growth rate prediction value based on the chain growth rate benchmark record set. Based on the chain growth rate benchmark record set containing, for example, 30 historical chain growth rate response data sequences under similar conditions, the adjustment content of the current temperature control target is called as the deduction input, and the adjustment content of the input is the median value of the current target adjustment interval, i.e., the chain growth rate adjustment target is -0.1236 kg per mole per second. The process of deducing the chain growth rate change trend in the data sequence is as follows: first, the arithmetic mean value of the 30 historical response data sequences in the benchmark record set is calculated at each time point to construct an average response curve, then the current input adjustment target value is compared with the average adjustment target value of the 30 historical events to calculate a scaling factor, for example, the historical average adjustment target is -0.1200, and the scaling factor is 1.03, then each data point on the average response curve is multiplied by the scaling factor 1.03 to obtain a scaled prediction response curve, and finally the value corresponding to the end point of the control period on the prediction response curve is taken to estimate the final response value generated by this cooling control, which is the chain growth rate prediction value.

[0028] The offset evaluation submodule calls the target growth rate set in the temperature control target, calculates the residual sequence distribution of the chain growth rate prediction value, quantifies the offset degree, and outputs the chain growth offset prediction result. The target growth rate set in the temperature control target, i.e. the interval median value -0.1236 kg per mole per second, is called, and the residual between the target value and the chain growth rate prediction value output by the growth estimation module (assumed to be -0.1250) is calculated, which is -0.0014. To more comprehensively quantify the degree of deviation, the residual is calculated to the entire corresponding time point between the predicted response curve and the target response curve, obtaining a complete residual sequence distribution. Then, the root mean square error of the residual sequence is calculated, which is used as the final quantitative indicator of the degree of deviation. The evaluation criteria for the degree of deviation are divided into three intervals: a root mean square error less than 0.002 is defined as "low deviation", an error between 0.002 and 0.005 is defined as "moderate deviation", and an error greater than 0.005 is defined as "high deviation". The boundary values of these intervals are set according to the statistical distribution of the deviation between historical predictions and actual results, corresponding to the 30th and 80th percentiles of the distribution, respectively. According to the calculated root mean square error value and its corresponding interval, the chain growth deviation prediction result is output.

[0029] Please refer to Figure 6 , the control execution module comprises: The action setting submodule obtains the temperature control target and the chain growth deviation prediction result, calls the control interval and the predicted deviation direction, calculates the adjustment amplitude and the sorting order of the cooling and heat exchange for each cooling control point, determines the sequence and target amplitude of the adjustment operation, and generates an adjustment action execution sequence. The temperature control target interval and the chain growth deviation prediction result are obtained. It is assumed that the deviation prediction result is "negative moderate deviation", which indicates that the estimated cooling effect may be slightly stronger than expected, so the original plan needs to be fine-tuned when setting specific adjustment actions. The control interval and the predicted deviation direction are called, and the adjustment amplitude and the sorting order of the cooling and heat exchange for each cooling control point are calculated. Based on the prediction of "negative moderate deviation", the adjustment strategy will preferentially use jacket heat exchange adjustment with higher adjustment accuracy and relatively flat response, and reduce the injection amount of the cooling agent which responds quickly but is difficult to control accurately. First, determine the sequence of the adjustment operation as adjusting the jacket circulating water first, and then according to the deviation value quantified by the deviation prediction result, slightly revise the total adjustment target, and split the revised total target according to the proportion of 85% for heat exchange adjustment and 15% for cooling adjustment. Finally, a detailed adjustment action execution sequence containing the sequence and specific adjustment amplitude is generated.

[0030] The instruction generation submodule constructs the cooling and heat exchange instructions for the cooling control points in the order of the adjustment action execution sequence, encodes the execution position and adjustment amplitude of the instructions respectively, forms a format recognizable by the control end, and generates a set of cooling and heat exchange operation instructions. According to the sequence of the adjusting action execution sequence, the cooling injection instruction and the heat exchange instruction corresponding to the cooling control point are constructed, first, the adjusting range allocated to the heat exchange adjustment is converted into a specific valve opening increment instruction by querying the characteristic curve of the heat exchanger valve, for example, the opening of the jacket A zone circulating water outlet valve is increased by 18%, and the instruction "HWCV-A: SETPOS: INC: 18" is generated, then, the adjusting range allocated to the cooling injection adjustment is converted into a precise pump starting time instruction by querying the physical property parameters of the cooling injection agent and the flow calibration data of the injection pump, for example, the B port cooling injection pump starts for 0.8 seconds, and the instruction "INJ-B: RUNTIME: SET: 0.8" is generated, the three elements of the execution position, the operation type and the adjusting range of the instruction are collectively coded into a standardized string format which can be directly recognized and executed by the control terminal hardware, and finally the cooling injection and heat exchange operation instruction set is formed.

[0031] The control feedback submodule synchronously transmits the cooling injection and heat exchange operation instruction set to the corresponding execution unit of the control terminal, collects and records the real-time response results of the regulation process, and outputs the temperature control result; The cooling injection and heat exchange operation instruction set is synchronously transmitted to the corresponding execution unit of the field programmable logic controller which controls the jacket valve and the cooling injection pump respectively through the field bus network, the execution unit receives and analyzes the instruction, drives the valve positioner and pump relay in the field to complete the specified operation, within a preset observation period after the instruction is issued, executed and completed, the entire real-time response results of the regulation process are continuously collected and recorded at a frequency not less than 1 Hz, the collected data includes the temperature, pressure, stirring power in the reaction kettle and the chain growth rate calculated in real time by the Raman spectrometer, and these high-frequency response data are bound and associated with the executed instruction content and the millisecond timestamp, and stored in the long-term historical database, finally a complete detailed data log containing the instruction, process response and final state is output as the result of this temperature control, by implementing the temperature control result, the temperature rise abnormality caused by weather changes, thermal load fluctuations and the like can be effectively responded to, the side reaction or explosion caused by local overheating of the polymerization reaction is prevented, the chain growth rate deviates from the target, the molecular weight distribution is affected, and the heat exchange imbalance brings energy consumption increase and equipment loss.

[0032] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms, any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the technical solution content of the present application still belongs to the protection scope of the present application.

Claims

1. A polyvinyl chloride polymerization reaction intelligent control system, characterized in that, The system comprises: A trend identification module obtains the PVC chain growth rate and the corresponding temperature rise amplitude of the reaction zone in a specified period during the PVC polymerization process, judges the change direction of the two in continuous time, and performs cross analysis as a combined feature, and assigns a polymerization reaction state label according to the analysis result; A coupling analysis module obtains the polymerization reaction state label and the meteorological data of the corresponding time period, analyzes the response relationship between the temperature change of the reaction process and the external meteorological factors, and outputs the temperature trend analysis result; A target generation module takes the PVC chain growth rate and the corresponding temperature rise amplitude of the reaction zone as the control subject based on the polymerization reaction state label and the temperature trend analysis result, and constructs a temperature control target; A deviation evaluation module takes the chain growth rate response record corresponding to the temperature control target in all running data as a comparison benchmark, estimates the growth rate generated by the current cooling control, compares it with the target growth rate, and outputs the chain growth deviation prediction result; A control execution module determines the adjustment action of cold injection and heat exchange by combining the temperature control target and the chain growth deviation prediction result, and generates a temperature control result.

2. The intelligent control system for polyvinyl chloride polymerization reaction according to claim 1, wherein, The polymerization reaction state label is specifically the change direction of the chain growth rate, the change direction of the temperature rise amplitude, and the trend combination feature. The temperature trend analysis result includes the temperature change sensitive section, the meteorological factor response strength, and the stage temperature rise probability. The temperature control target is specifically the target chain growth rate interval, the target temperature rise amplitude range, and the adjustment parameter boundary. The chain growth deviation prediction result includes the growth rate prediction value and the target deviation degree. The temperature control result is specifically the cold injection adjustment instruction, the heat exchange adjustment instruction, and the execution adjustment sequence.

3. The intelligent control system for polyvinyl chloride polymerization reaction according to claim 1, wherein, The trend identification module comprises: A trend extraction submodule obtains the chain growth rate and the corresponding temperature rise amplitude of the reaction zone in a specified period during the PVC polymerization process, calculates the change direction of each in adjacent time periods, assigns a chain growth rate change direction label and a temperature rise amplitude change direction label according to the positive and negative of the difference value, and generates a trend direction label; A feature pairing submodule combines the chain growth rate change direction label and the temperature rise amplitude change direction label into a trend pairing feature based on the trend direction label, obtains the polymerization stage temperature control target state of the corresponding time period of the trend pairing feature, judges the deviation direction between the pairing feature and the temperature control state, filters the associated feature pairing, and generates a state deviation trend combination; A state assignment submodule divides the polymerization reaction state type corresponding to the combined trend based on the state deviation trend combination and assigns a corresponding label to generate a polymerization reaction state label.

4. The intelligent control system for polyvinyl chloride polymerization reaction according to claim 3, wherein, The coupling analysis module comprises: A label reading submodule obtains the polymerization reaction state label and acquires the meteorological data of the corresponding time period, extracts the air temperature change parameter from the meteorological data, and pairs the state label in a time-synchronous manner to generate a state-meteorological pairing group. The meteorological coupling submodule is configured to, based on the state-meteorological pair group, construct a feature pair of a state label and a temperature change parameter, input the feature pair into a Bayesian regression model, calculate a joint probability distribution of a temperature change value under the state label, establish a response relationship of meteorological change on temperature evolution of a reaction process, and generate temperature response probability distribution data; The trend identification submodule is configured to, based on the temperature response probability distribution data, determine whether a temperature value in a current polymerization reaction process is in a probability region sensitive to temperature change, identify a trend region attribution according to a probability distribution threshold, and output a temperature trend analysis result.

5. The intelligent control system for polyvinyl chloride polymerization reaction according to claim 4, wherein, The target generation module includes: The parameter adaptation submodule is configured to acquire the polymerization reaction state label and the temperature trend analysis result, call polymerization vinyl chloride chain growth rate and reaction region temperature rise amplitude data of a corresponding time period, extract a set of regulation input parameters according to a state type and a trend region attribution, and output a matching cooling plan group. The target construction submodule is configured to combine real-time cooling device operation capability parameters, limit a chain growth rate adjustment value and a temperature rise amplitude adjustment range according to the matching cooling plan group, form a next period regulation index combination, and output a temperature control target. The bias evaluation module includes:

6. The intelligent control system for polyvinyl chloride polymerization reaction according to claim 5, wherein, The response record submodule is configured to call the temperature control target, refer to cooling control data corresponding to a current regulation type and the temperature control target in all operation data, extract chain growth rate response records in the cooling control data as a comparison data source, and generate a chain growth rate benchmark record set. The growth estimation submodule is configured to, based on the chain growth rate benchmark record set, call adjustment content of the current temperature control target as input, deduce a chain growth rate change trend in a data sequence, estimate a response value generated by the current cooling control as a chain growth rate prediction value, and output a chain growth rate offset prediction result. The control execution module includes: The action setting submodule is configured to acquire the temperature control target and the chain growth rate offset prediction result, call a regulation interval and a predicted offset direction, calculate an injection cooling and heat exchange adjustment range and a sorting order for each cooling control point, determine a sequence relationship and a target range of adjustment operations, generate an adjustment action execution sequence, and output an injection cooling and heat exchange operation instruction set.

7. The intelligent control system for polyvinyl chloride polymerization reaction according to claim 6, wherein, The instruction generation submodule is configured to construct injection cooling and heat exchange instructions of a cooling control point in a sequence of the adjustment action execution sequence, encode an execution position and an adjustment range of each instruction, form a control terminal recognizable format, and generate the injection cooling and heat exchange operation instruction set. The control feedback submodule is configured to synchronously transmit the injection cooling and heat exchange operation instruction set to a corresponding execution unit of a control terminal, collect and record real-time response results of a regulation process, and output a temperature control result. ​ ​

Images