Industrial steam flow intelligent adjusting system

Abstract

The invention relates to the technical field of industrial process automatic control, in particular to an industrial steam flow intelligent adjusting system. According to the method, the continuous change trend of the steam state is subjected to time sequence identification, dynamic classification of the steam state can be realized under multivariate combination, so that a steam state switching path is insighted in advance, an information coupling measurement mechanism between enthalpy change and flow is introduced on the basis, and calculation of mutual information between enthalpy and flow is combined, so that the dynamic classification of the steam state is realized. Abnormal sections with adjustment deviation in actual operation are accurately recognized, the adjustment possibility is further judged according to the deviation quantification result of the target flow and the actual output flow, controllable and uncontrollable sections are effectively divided, and the rollback value domain limitation of the uncontrollable sections is set. Therefore, in the subsequent target value adjusting process, target setting of the remaining sections can be reset in a targeted mode, the priority sequence is constructed according to the adjusting time sequence and the adjusting amplitude, and dynamic sorting output of the adjusting strategy is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial process automatic control, and in particular to an industrial steam flow intelligent regulation system. BACKGROUND

[0002] The technical field of industrial process automatic control mainly involves real-time monitoring, regulation and control of various equipment and process parameters in industrial production processes to achieve automation and efficient management of production processes.

[0003] Among them, the traditional industrial steam flow intelligent regulation system refers to a control system for real-time regulation of the flow of steam as a heat source or power medium in the pipeline according to process requirements, so as to solve the problems of steam supply-demand matching and energy utilization efficiency.

[0004] The prior art mainly relies on real-time regulation of steam flow to respond to changes in process requirements, but its control strategy is mostly based on single-point data feedback or fixed-value rule adjustment, lacking the ability to systematically judge the trend of steam state changes, especially when the steam state fluctuates in stages, it cannot identify potential mismatch risks, in addition, such systems usually take the current flow deviation as the only basis for regulation, ignoring the energy conversion relationship between steam enthalpy and flow, resulting in the system still adjusting according to the original target when coupling deviation occurs, which easily leads to local energy waste or system fluctuation, for example, if the state interval switching trend is not identified under high temperature and high pressure, the system will continue to execute the static target flow instruction, causing over-regulation or under-regulation, affecting heat load balance and operation safety, and for the regulation failure history, no closed-loop rollback mechanism is formed, which cannot set response limiting strategies for abnormal sections, resulting in insufficient adaptability and frequent regulation failures of the control system in dynamic complex environments. SUMMARY

[0005] The purpose of the present application is to solve the shortcomings in the prior art and to provide an industrial steam flow intelligent regulation system.

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme: an industrial steam flow intelligent regulation system, the system comprises: a steam state identification module: collecting time series of steam temperature, pressure and density in a specified period in an industrial pipeline, judging steam state changes according to the time series, and obtaining a state change marker sequence; an enthalpy-flow coupling analysis module: referring to the state change marker sequence, performing back substitution calculation on the steam enthalpy change amount of each period, collecting the steam output flow of the corresponding period, comparing the mutual information between the steam enthalpy change amount and the steam output flow, and marking the coupling abnormal section; Adjustment space determination module: compare the deviation between the preset steam target flow and the current steam output flow, determine the adjustment space of each coupling abnormal section, set a forced fallback threshold for the unadjustable section, and obtain an adjustment locking section and a fallback threshold set; Target setting adjustment module: according to the adjustment locking section and the fallback threshold set, adjust the flow target setting value of steam output of each period, and generate a flow adjustment control instruction set.

[0007] The present application improves that the state change marker sequence includes state type markers, change occurrence positions, and continuous marker sections, the coupling abnormal section includes abnormal section numbers, corresponding period starts and ends, and steam flow range influences, the adjustment locking section and the fallback threshold set include locking periods, set threshold value ranges, and execution limitation labels, and the flow adjustment control instruction set includes adjusted target flows, adjustment execution identifications, and period control instructions.

[0008] The present application improves that the steam state recognition module includes: State acquisition submodule: acquire steam temperature, pressure and density data of a specified period in an industrial pipeline, and construct a steam state basic data sequence; Interval classification submodule: according to temperature and pressure combinations in the steam state basic data sequence, judge a current steam corresponding state interval by referring to IAPWS steam reference boundary values, combine density trend information to label a type to which each interval belongs, and generate a state attribution interval label set; Change recognition submodule: based on the state attribution section label set, compare state category change situations in adjacent periods, extract time points and section numbers at which state switching occurs, and record state path changes in time sequence structure, and generate a state change marker sequence.

[0009] The present application improves that the heat enthalpy-flow coupling analysis module includes: Enthalpy calculation submodule: acquire temperature, pressure and density data corresponding to each period in the state change marker sequence, calculate a specific enthalpy value of steam in each period, and a difference value of the specific enthalpy value in continuous periods, and generate a steam enthalpy change amount; Flow extraction submodule: acquire steam flow meter monitoring data in a period corresponding to the steam enthalpy change amount, and generate a steam output flow; Coupling section marking submodule: calculate mutual information values between the steam enthalpy change amount and the steam output flow in a corresponding period, mark periods in which there is a deviated coupling relationship, and generate a coupling abnormal section.

[0010] The present application improves that the adjustment space determination module includes: Target value acquisition submodule: acquire preset steam target flow of each period, and construct a preset steam target flow set; Deviation judgment submodule: compare the preset steam target flow set with the steam output flow of the coupling abnormal section segment by segment, filter out the section segments whose deviation exceeds the steam adjustment allowable threshold, and generate an adjustment abnormal section set; Section marking submodule: mark the abnormal section segments in the adjustment abnormal section set as unadjustable state, set the corresponding flow rollback threshold according to the section number, and generate an adjustment locking section and a rollback threshold set.

[0011] The target setting adjustment module comprises: Locking response submodule: based on the locking section in the adjustment locking section and rollback threshold set, perform a null operation on the steam flow target value in the period, mark the unadjustable state, eliminate the original target value corresponding to the section segment, and generate a locking section elimination result; Threshold comparison submodule: based on the remaining period in the locking section elimination result, extract the corresponding steam output flow, compare it with the same segment rollback threshold, reset part of the target value according to the comparison result, and generate a target value adjustment result; Setting summary submodule: unify the adjustment value of each period in the target value adjustment result and the locking section state identifier, and generate a flow adjustment control instruction set.

[0012] The execution sequence optimization module comprises: The flow adjustment sequence index comprises a sorting number, an adjustment time sequence relationship and a priority execution identifier.

[0013] The execution sequence optimization module comprises: Effectiveness screening submodule: acquire all adjustment setting instructions in the flow adjustment control instruction set, merge effective items according to time labels, and generate an adjustment target time set; Sorting generation submodule: according to the adjustment target time set, combine setting sorting rules according to time sequence and size of two dimensions, and output a sorting priority time sequence list; Index labeling submodule: for the sorting priority time sequence list, establish a sequential execution label from front to back, and record the period and adjustment instruction number corresponding to the sorting position, and generate a flow adjustment sequence index.

[0014] Compared with the prior art, the application has the advantages and positive effects that: In the present application, by time series identification of the continuous change trend of the steam state, dynamic classification of the steam state can be realized under multivariate joint, thereby early insight into the steam state switching path, on this basis, the information coupling degree measurement mechanism between the enthalpy change and the flow is introduced, combined with the calculation of mutual information between enthalpy and flow, the abnormal section existing in the adjustment deviation in actual operation is accurately identified, further through the deviation quantization result of the target flow and the actual output flow to judge the adjustment possibility, effectively divide the controllable and uncontrollable sections, and set the rollback value domain limit for the uncontrollable section, thereby in the subsequent target value adjustment process, the target setting of the remaining section can be reset, according to the adjustment time sequence and the adjustment amplitude to construct the priority order, realize the dynamic sorting output of the adjustment strategy, this processing flow improves the steam system adjustment matching accuracy while considering the implementability, avoids applying invalid instructions to the section without adjustment space, effectively improves the energy utilization efficiency and adjustment response ability of the overall system under complex working conditions. In actual application scene, through field industrial data comparison and simulation verification, the intelligent adjustment system can effectively improve the accuracy and dynamic response ability of steam flow adjustment. Especially under complex working conditions such as frequent switching of steam state, high temperature and high pressure fluctuation and variable load, compared with the traditional control strategy, the average energy utilization efficiency can be improved by about 8% to 15%, the adjustment response time is shortened by about 25%, the waste risk caused by energy over-regulation or under-regulation is significantly reduced, and the stability and economy of the overall system operation are improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a system module diagram of the present application; Figure 2 is a system framework diagram of the present application; Figure 3 is a schematic diagram of the steam state identification module of the present application; Figure 4 is a schematic diagram of the enthalpy-flow coupling analysis module of the present application; Figure 5 is a schematic diagram of the adjustment space determination module of the present application; Figure 6 is a schematic diagram of the target setting adjustment module of the present application; Figure 7 is a schematic diagram of the execution order optimization module of the present application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0017] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0018] Referring to Figure 1 The present application provides a technical solution: an industrial steam flow intelligent regulation system, the system comprising: A steam state recognition module: collecting time series of steam temperature, pressure and density in a specified period in the industrial pipeline, judging the steam state change according to the time series, and obtaining a state change marker sequence; A heat enthalpy-flow coupling analysis module: referring to the state change marker sequence, performing back substitution calculation on the steam heat enthalpy change in each period, collecting the steam output flow in the corresponding period, comparing the mutual information between the steam heat enthalpy change and the steam output flow, and marking the coupling abnormal section; A regulation space determination module: comparing the deviation between the preset steam target flow and the current steam output flow, determining the regulation space of each coupling abnormal section, and setting a forced fallback threshold for the non-adjustable section, to obtain a regulation lock section and a fallback threshold set; A target setting adjustment module: adjusting the flow target setting value of the steam output in each period according to the regulation lock section and the fallback threshold set, and generating a flow regulation control instruction set; The state change marker sequence includes state type markers, change occurrence positions, and continuous marker sections, the coupling abnormal section includes abnormal section numbers, corresponding period start and end, and steam flow range affected, the regulation lock section and the fallback threshold set include lock period, set threshold value range, and execution limitation label, and the flow regulation control instruction set includes adjusted target flow, regulation execution identification, and period control instruction.

[0019] Referring to Figure 2 And Figure 3 The steam state recognition module comprises: A state acquisition sub-module: collecting steam temperature, pressure and density data in a specified period in the industrial pipeline, and constructing a steam state basic data sequence; The steam temperature, pressure and density data of a specified period in an industrial pipeline are collected by installing temperature sensors, pressure sensors and density sensors at key positions of the pipeline. The collection period is usually set by the control system, and the sensor output is automatically read at regular intervals. The data collection module obtains the temperature, pressure and density information at the current time point from each sensor and marks the time stamp uniformly to ensure that each type of data has a time correspondence. In this process, the control system interacts with each sensor through the communication protocol, receives the sampling value in real time and caches it into the edge computing device. Then the cached data is synchronized and uploaded to the database. To avoid the destruction of data sequences by abnormal values, a data preprocessing process is set, including setting and checking the temperature, pressure and density sampling value range, marking and isolating abnormal data after identification. To ensure data stability, baseline calibration of the sensor device is performed daily using standard media or reference values for comparison. The collection system can record continuous data sequences within the entire period according to the configuration strategy, thereby forming a steam state basic data sequence.

[0020] Interval classification submodule: according to the combination of temperature and pressure in the steam state basic data sequence, the IAPWS steam reference boundary value is compared to determine the current steam state interval, and the type of each interval is marked according to the density trend information to generate a state attribution interval label set; When judging the state according to the combination of temperature and pressure in the steam state basic data sequence and comparing the IAPWS steam reference boundary value, the temperature value and pressure value of each sampling point are respectively substituted into the state determination logic for boundary judgment. According to the temperature-pressure critical boundary table provided by IAPWS, the corresponding saturated pressure limit is found at a given temperature, and then the current pressure value is compared with the boundary value to determine whether it is higher than, equal to or lower than the boundary pressure, thereby preliminarily determining the liquid phase, saturated state or superheated state. On this basis, the density value in the continuous period is compared by combining the density data sequence, the density change trend is judged by difference analysis or adjacent multi-point average method, and if the density continuously rises, it is marked as a density rising zone, and if it falls, it is marked as a falling zone. When the trend is stable, it is marked as a stable zone. Through the combination of temperature-pressure state and density change trend, labels such as "superheated-density falling" and "saturated-density stable" can be generated, and each data point is assigned a corresponding interval attribution label. This operation is performed for each sampling point in the entire data sequence, forming a state attribution interval label set.

[0021] Change identification submodule: based on the state attribution interval label set, the state category change in adjacent periods is compared, the time point and interval number where the state switches are extracted, and the state path change is recorded according to the time sequence structure to generate a state change marker sequence; When the state attribution section label set is used for change identification, the entire label sequence is first traversed in chronological order, the state label content of adjacent two time points is compared, if there is inconsistency in the state label content, it can be determined that the state changes at this place, the current time point is recorded as the state switching time, and its position number in the overall data sequence is marked, then the subsequent label content is continuously traversed, all state change nodes are identified, sections with the same continuous label content are grouped and numbered uniformly, the state change nodes are constructed into a state path sequence, the starting time, ending time, starting label, ending label and section number of each section in the path are recorded, whether there is repetition, alternation, transfer and other characteristics in the state path is judged by comparing the state category difference of the continuous sections in the path, a state switching record table should be constructed during execution to assist in analyzing the time sequence characteristics of the data, the table is composed of state path number, previous and next state labels and switching time point, after processing all time periods, a complete state change marking sequence can be obtained.

[0022] Referring to Figure 2 and Figure 4 , the enthalpy-flow coupling analysis module comprises: An enthalpy calculation submodule: obtains the temperature, pressure and density data corresponding to each time period in the state change marking sequence, calculates the specific enthalpy value of the steam in each time period and the difference of the specific enthalpy value in the continuous time period, and generates the steam enthalpy change amount; After obtaining the temperature, pressure and density data corresponding to each period in the state change marker sequence, the state judgment classification needs to be determined to decide which enthalpy calculation path to use. For steam in the superheated state, the specific enthalpy calculation expression in region three of the IAPWS-IF97 industrial steam property formula needs to be combined. Region three is suitable for high pressure and high temperature conditions and can be used as an input parameter by looking up the specific enthalpy calculation expression in region three of the IAPWS industrial steam standard table. The specific enthalpy calculation expression based on the state equation is selected, the temperature is 260, the pressure is 3, and the specific enthalpy is 2800 after substituting into the specific enthalpy calculation formula. If the temperature of the next period is 255 and the pressure is 3, the specific enthalpy is 2750, and the difference between the two is -50, which is recorded as the enthalpy change of this period. For steam in the saturated state, the saturated steam specific enthalpy data in Table 2 of IAPWS-IF97 needs to be consulted. The specific enthalpy of the saturated liquid is 1080 and the specific enthalpy of the saturated vapor is 2800 at a temperature of 250. By comparing the current density with the theoretical saturated liquid and vapor density, if it is closer to the vapor density, the steam specific enthalpy is taken as 2800. The density of the next period is compared, and if the state changes to saturated liquid, the specific enthalpy changes to 1080, and the specific enthalpy change is -1720. If the steam is in the liquid phase, the specific enthalpy formula in region one of IAPWS-IF97 needs to be used. This region is suitable for low temperature and high pressure conditions, and the specific enthalpy calculation can be given by an empirical polynomial expression. At a temperature of 230 and a pressure of 3.5, the specific enthalpy is 1080 after substituting into the empirical expression. The enthalpy change of this period is obtained by subtracting the specific enthalpy of the next period. The above calculation is repeated for all state periods to obtain the steam enthalpy change sequence.

[0023] Flow extraction sub-module: obtain the steam flow meter monitoring data in the period corresponding to the steam enthalpy change to generate the steam output flow; When obtaining the steam flow meter monitoring data in the period corresponding to the steam enthalpy change, the pairing of the enthalpy change sequence and the flow data needs to be completed through a time synchronization mechanism. First, the start and end time points of each enthalpy change are extracted from the record. For example, the enthalpy change corresponding time is from 10:00 to 10:00:10. Then, all valid sampling values in this time period are extracted from the flow meter data. Assuming that the flow meter records data every second, ten data points can be extracted from this period. The average of the sum of these ten data points is taken as the average flow value representing this period. If the flow changes little in some periods, the flow difference between the start and end time points can be directly taken as the representative output flow. Then, the average flow value is paired with the enthalpy change value of the period to form a unified data structure. If there is missing or sudden change in the flow data during the process, the data can be interpolated to fill in the abnormal values to ensure data continuity. Finally, the enthalpy change and the steam output flow of all periods are paired and arranged in sequence.

[0024] Coupling section marking sub-module: calculate the mutual information value between the steam enthalpy change and the steam output flow in the corresponding period, mark the period with deviated coupling relationship, and generate the coupling abnormal section; Calculate the statistical dependence between the steam enthalpy change and the steam output flow, and use mutual information as a measurement index. Mutual information reflects how much information one variable contains in another variable. When applied to continuous period data, it can reveal the coupling strength of enthalpy change and flow change. The core steps of this process include: constructing joint probability distribution, calculating mutual information value, developing threshold strategy and identifying coupling deviation.

[0025] The enthalpy change and the steam output flow are paired in each corresponding period to form an observation sample set. Assuming there are sample points, normalize the two variables (linear transformation to the interval) respectively, and then divide and into equal intervals, for example , that is, form 5 enthalpy intervals and 5 flow intervals. Then, for each group data points, count the enthalpy interval and the flow interval to which they belong, count the frequency of all combinations to form a joint frequency matrix, and then convert it to a joint probability distribution .

[0026] The formula for calculating mutual information is: ; Where, : the mutual information value between the enthalpy change and the steam flow . Compare it with the mutual information threshold obtained by subtracting the standard deviation from the total mutual information mean. If the mutual information value of a period is lower than the mutual information threshold, it is marked as an abnormal period with deviated coupling relationship; : the joint probability of the enthalpy data falling into the interval and the flow data falling into the interval; : the marginal probability of the enthalpy data falling into the interval; : the marginal probability of the flow data falling into the interval; : the total number of intervals of variable discretization.

[0027] Mutual information essentially measures the difference between the joint distribution and the independent distribution of variables, if and are completely independent, then mutual information is zero; if there is a statistical dependence, mutual information is positive, the higher the value indicates the greater the coupling strength.

[0028] Suppose a sample of 10 time periods is analyzed, and the normalized and interval mapping processing is completed, the heat enthalpy change and steam flow are each divided into 5 intervals. After statistics, the following joint probability and marginal probability (unit: relative frequency) are obtained: where the heat enthalpy is in the 3rd interval, and the flow is also in the 3rd interval; , the heat enthalpy is in the 4th interval, and the flow is in the 4th interval; , the heat enthalpy is in the 5th interval, and the flow is in the 5th interval, and the corresponding marginal probability is: , ; , ; , .

[0029] Project 1: Calculate the mutual information item of : ; Project 2: Calculate the mutual information item of : ; Project 3: Calculate the mutual information item of : .

[0030] The total mutual information value is obtained by adding the three items: .

[0031] This value represents the total mutual information contributed by the three typical combinations, and if other combinations are zero, this is the total mutual information value; in real calculation, all non-zero items should be calculated and then summed to obtain the total mutual information value .

[0032] To identify the coupling deviation section, it is necessary to establish evaluation criteria according to the calculation results of mutual information values of all time periods. First, the mutual information values corresponding to all time periods are counted to form a complete mutual information value sequence. Second, the average value of the sequence is calculated as the baseline of the overall coupling level between the heat enthalpy change and the steam flow under the current operating state. Third, the standard deviation of the mutual information value sequence is calculated to reflect the volatility of the coupling strength. Fourth, the determination threshold is set: the average value minus one standard deviation is taken as the lower limit of coupling abnormality determination. Fifth, the mutual information value of each time period is traversed, and if the mutual information of a certain time period is lower than the threshold, it is determined that there is a deviation coupling relationship between the heat enthalpy and the flow in the abnormal section.

[0033] Suppose a system is continuously observed for 30 time periods, and the mutual information value of each period is calculated, the average value is 1.8, and the standard deviation is 0.4. The determination threshold is 1.4. If the mutual information of the 12th period is 1.2 and the 19th period is 1.1, both of them are lower than 1.4, so they are marked as coupling abnormal sections. These sections will be further extracted to form a coupling abnormal section marking table, including abnormal time period number, start and end time, corresponding mutual information value, heat enthalpy interval number and flow interval number combination information, which is used for subsequent state analysis and fault backtracking processing.

[0034] Please refer to Figure 2 and Figure 5 , the adjustment space determination module comprises: Target value acquisition submodule: acquire preset steam target flow of each time period, and construct a preset steam target flow set; To acquire the preset steam target flow of each time period, first, the target values corresponding to each time period need to be extracted from the dispatching plan, process setting or control parameters. These target values may come from the daily production dispatching plan in the process control system, the prediction results of the historical operation model or the manually set operation curve. The system matches the target values according to the time stamp structure, for example, the target flow from 9:00 to 9:05 is 1500 cubic meters per hour, so the preset flow of all time points in this time period is uniformly set to this value. If there are multiple speed requirements in some time periods, the target values are updated according to the segmented setting rules. In addition, in order to facilitate comparison with actual measurement data, the target value structure needs to be arranged as a time period index and target flow key-value pair structure, stored in the system cache, and kept aligned with the time period output by the coupling analysis module. All target values are arranged in order to form a complete preset steam target flow set.

[0035] Deviation judgment submodule: compare the preset steam target flow set with the steam output flow of the coupling abnormal section segment by segment, filter out the sections with deviation exceeding the steam adjustment allowable threshold, and generate an adjustment abnormal section set; When comparing the preset steam target flow rate set with the steam output flow rate of the coupled abnormal section segment by segment, the target flow rate and the actual output flow rate are first matched one by one according to the time series. The difference between the target flow rate and the actual value is calculated in each time period. Then, the difference is compared and judged according to the steam regulation allowable threshold set by the system. For example, if the set allowable deviation is 5%, the percentage error of the actual flow rate relative to the target value is calculated for each time period. When the error is greater than 5%, it is judged as a deviation exceeding the limit segment. After comparing all the data segment by segment, the segment numbers of all error exceeding the limit are extracted. These segments are arranged by time and recorded as a unified set of regulation abnormal segments. This set contains the start and end time, target value, actual value and exceedance range of each abnormal segment, which is convenient for subsequent regulation limit labeling.

[0036] Section Marking Submodule: Marks abnormal sections in the set of abnormal adjustment sections as unadjustable, and sets the corresponding flow back-off threshold according to the section number, generating adjustment lock-up sections and back-off threshold sets; When marking abnormal segments in the set of abnormal adjustment segments as unadjustable, it is first necessary to identify which segments have lost effective adjustment response capability during operation. This identification is based on the following criteria: the flow deviation within the segment has not been corrected by adjustment commands for multiple consecutive time periods, and the deviation trend continues to expand or remains stable in the high deviation range for a long period. Combined with the system's adjustment response records for similar historical operating conditions, if such segments have experienced multiple adjustment failures, they are determined to be unadjustable. Subsequently, a flow rollback threshold needs to be set for each segment marked as unadjustable. This threshold serves as a subsequent operational limit benchmark for segments already determined to be unadjustable, used to determine whether there is a serious risk of loss of control or the need for load reduction. The flow rollback threshold is not used to determine whether the segment has entered an unadjustable state, but rather, based on the established unadjustable state, it sets the next stage control lower limit benchmark for the segment. The specific calculation method comprehensively considers the target flow of the segment, the maximum flow deviation value that has occurred within the segment, and the residual deviation amount from adjustment failures in the segment or similar historical segments, using the following formula: ; in, It is the flow rate reduction threshold for this section, used to reduce or limit the steam flow rate; The target steam flow rate set for this section of the dispatch; The maximum actual deviation value observed in this section is taken from all sampling points. The maximum value; This represents the mean residual deviation after historical adjustment failures, indicating the system's persistent error when adjustments failed to take effect under similar historical conditions. , Respectively, the weight coefficient of two factors, the value is based on the following: when a certain section deviation amplitude is significantly greater than a certain proportion of the target value (such as more than 10%), it indicates that there is a significant gap between itself and the system setting, and the weight of the maximum deviation item should be increased Make the rollback threshold more stringent, for example, set to 0.6; and when the residual error is large after adjustment failure, it indicates that the control system response is insufficient, so increase The weight makes the system more sensitive to reflect its historical adjustment failure characteristics, for example, set to 0.3.

[0037] Suppose the target flow of section No. 21 is 1600, the maximum deviation is 220, and the historical residual error is 50. The formula is calculated as follows: Therefore, the non-adjustable section No. 21 has a corresponding flow rollback threshold of 1483, and the system writes this value into the "adjustment lock section and rollback threshold set". In subsequent operation, once the flow of this section is lower than 1483, it can trigger load limiting, alarm or control logic switching strategy, thereby completing the limit binding and state structure output of each non-adjustable section.

[0038] Please refer to Figure 2 and Figure 6 , the target setting adjustment module includes: Lock response submodule: based on the lock section in the adjustment lock section and rollback threshold set, the steam flow target value in the time period is executed empty operation, and the non-adjustable state is marked, the original target value corresponding to the section is excluded, and the lock section exclusion result is generated; Based on the lock section in the adjustment lock section and rollback threshold set, all section numbers and corresponding time period start and end range need to be extracted from the list of sections determined as non-adjustable state, and then all target flow records of the time period involved are searched in the original target flow table. The target flow field of these records is set to null value, and the time period is marked as non-adjustable state at the same time. This marking method can use Boolean marking field record in the target flow table structure, and the coverage range of the lock section is set at the same time to avoid its participation in the target value adjustment operation in the later stage. If a section crosses multiple sampling time points during operation, the empty operation should be executed point by point to ensure consistency. For example, the lock section No. 18 covers the time period from 10:15:00 to 10:15:30, and the target value field of all sampling points in the time period in the target flow table is emptied, and the lock identification is true at the same time. All target flow data items that have completed the empty operation are excluded from the original target value set structure to form the lock section exclusion result.

[0039] Threshold comparison submodule: based on the remaining time period in the lock section exclusion result, the corresponding steam output flow is extracted and compared with the same section rollback threshold. According to the comparison result, part of the target value is reset, and the target value adjustment result is generated; Based on the remaining period of the lockout segment rejection result, the corresponding steam output flow is extracted, which needs to match the corresponding timestamp data point from the actual flow record table, and compare it with the rollback threshold corresponding to the period. In the comparison process, it is judged segment by segment. If the current actual flow value is higher than the set rollback threshold of the segment, it means that the current flow has deviated from the non-adjustable critical value and has adjustment space. Therefore, the target value of this period can be restored or appropriately adjusted. If the current actual flow value is lower than the rollback threshold, the current target value remains unchanged or the target is further reduced to prevent the adjustment system from misjudging and issuing invalid instructions. The adjustment range of the target value can be set according to the actual flow exceeding the threshold ratio, for example, if the actual flow of a certain segment is 1520 and the rollback threshold is 1480, the exceeding ratio is about 2.7%, and the adjustment target value is set as the lower limit of the actual flow value, such as 1490 as the corrected target. Execute the judgment and adjustment operation on each remaining period to arrange the new target value adjustment result.

[0040] Set summary submodule: unify the adjustment value of each period in the target value adjustment result and the lockout segment state identifier to generate a set of flow adjustment control instructions; When unifying the adjustment value of each period in the target value adjustment result and the lockout segment state identifier, a complete data structure needs to be established, which includes the timestamp field, the lockout identifier field, and the target flow field. First, parse the target value adjustment result line by line, merge the target value of each time period and whether it belongs to the lockout segment, then set the target value in the lockout segment to an invalid value or a default marker, and update the remaining non-locked sections to new target values according to the adjustment result. The entire arrangement process needs to maintain consistency with the interface data format of the adjustment control system, such as aligning data output every five seconds, forming a structured instruction set, and each record in the instruction set including time, target value, adjustment flag, lockout state, etc.

[0041] Please refer to Figure 2 and Figure 7 It also includes an execution sequence optimization module, which sorts the flow adjustment control instruction set according to the execution priority to generate a flow adjustment sequence index. The flow adjustment sequence index includes sorting number, adjustment time sequence relationship, and priority execution identifier. The execution sequence optimization module includes: The effectiveness screening submodule: obtains all adjustment setting instructions in the flow adjustment control instruction set, merges valid entries according to time labels, and generates a set of adjustment target times. When obtaining all the adjustment setting instructions in the flow regulation control instruction set, the entire instruction set data structure needs to be traversed, the entries containing specific numerical settings in each record are extracted, the records with null value identifier, placeholder field or marked as locked state are removed, and only the instruction entries with real adjustment significance are retained. Each instruction needs to at least include the timestamp field and the target flow setting field, and then all valid records are merged by time period according to the time label, that is, the case of multiple setting instructions at the same time point is de-duplicated or merged. If multiple adjustment instructions appear at a time point, the median or the latest record of the setting value is used as the representative value of the time point. For example, three setting instructions appear at 10:01:00 am, which are 1520, 1530 and 1510, and 1520 can be merged as the effective target. All the valid adjustment records merged by time constitute a new adjustment target time set.

[0042] The sorting generation submodule: according to the adjustment target time set, the setting sorting rule is combined according to the time sequence and the size of the setting value, and the sorted priority time sequence list is output; When sorting according to the adjustment target time set, the sorting rule needs to be constructed based on two dimensions. The first dimension is the time sequence, that is, all target time points are sorted according to the timestamp from early to late. The second dimension is the size order of the adjustment target value. In some running strategies that need load balancing, instructions with larger or smaller target values may need to be processed first. Therefore, the system needs to set a combined sorting logic, for example, the target value is sorted in descending order, and when the target value is the same, the time is sorted in ascending order. Specifically, the primary key is the size of the setting value, and the secondary key is the time label.

[0043] The index labeling submodule: for the sorted priority time sequence list, the sequential execution label is established from front to back, and the time period and the adjustment instruction number corresponding to the sorting position are recorded, and the flow regulation sequence index is generated; When the sequential execution label is established for the sorted priority time sequence list, the first time point after sorting needs to be numbered in sequence from the first time point, and the number is the priority order identifier of the flow regulation system scheduling execution. At the same time, in the process of generating the number, the original time period timestamp and the adjustment instruction number in the flow regulation control instruction set corresponding to the number also need to be recorded. For example, the record sorting number is 1, the time is 10:06:00, and the instruction number is 35. This information needs to be written into the flow regulation sequence index structure. The structure fields include: execution serial number, time period, instruction number, target value, etc. After all the sorting results are numbered and information bound, a complete sequence index table is formed.

[0044] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application in other forms. Any person skilled in the art can make changes or modifications to the above disclosed technical contents into equivalent embodiments with equivalent changes, and apply them to other fields. However, 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 technical solution of the present application.

Claims

1. An intelligent industrial steam flow regulation system, characterized in that: The system comprises: A steam state recognition module: collect time series of steam temperature, pressure and density in a specified period in an industrial pipeline, judge steam state change according to the time series, and obtain a state change marker sequence; A enthalpy-flow coupling analysis module: refer to the state change marker sequence, back substitution calculation of steam enthalpy change in each period, collect steam output flow in the corresponding period, compare the mutual information between steam enthalpy change and steam output flow, and mark the coupling abnormal section; An adjustment space determination module: compare the deviation between the preset steam target flow and the current steam output flow, determine the adjustment space of each coupling abnormal section, set a forced fallback threshold for the non-adjustable section, and obtain an adjustment locking section and a fallback threshold set; A target setting adjustment module: adjust the flow target setting value of steam output in each period according to the adjustment locking section and the fallback threshold set, and generate a flow adjustment control instruction set.

2. The industrial steam flow intelligent regulation system of claim 1, wherein: The state change marker sequence includes state type markers, change occurrence positions, and continuous marker sections, the coupling abnormal section includes abnormal section numbers, corresponding period start and end, and steam flow range, the adjustment locking section and the fallback threshold set include locking period, setting threshold value range, and execution limitation label, and the flow adjustment control instruction set includes adjusted target flow, adjustment execution identification, and period control instruction.

3. The industrial steam flow intelligent regulation system of claim 1, wherein: The steam state recognition module comprises: A state collection submodule: collect steam temperature, pressure and density data in a specified period in an industrial pipeline, and construct a steam state basic data sequence; An interval classification submodule: according to the temperature and pressure combination in the steam state basic data sequence, refer to the IAPWS steam reference boundary value, judge the current steam corresponding state interval, combine the density trend information to label the type of each interval, and generate a state attribution interval label set; A change recognition submodule: based on the state attribution section label set, compare the state category change in adjacent periods, extract the time point and section number of state switching, and record the state path change according to the time sequence structure, and generate a state change marker sequence.

4. The industrial steam flow intelligent regulation system of claim 1, wherein: The enthalpy-flow coupling analysis module comprises: An enthalpy calculation submodule: obtain the temperature, pressure and density data corresponding to each period in the state change marker sequence, calculate the specific enthalpy value of steam in each period, and the difference value of the specific enthalpy value in the continuous period, and generate the steam enthalpy change; A flow extraction submodule: obtain the steam flow meter monitoring data in the period corresponding to the steam enthalpy change, and generate the steam output flow; A coupling section marking submodule: calculate the mutual information value between the steam enthalpy change and the steam output flow in the corresponding period, mark the period with deviated coupling relationship, and generate a coupling abnormal section.

5. The industrial steam flow intelligent regulation system of claim 4, wherein: For calculating the mutual information value between the steam enthalpy change and the steam output flow in the corresponding period, the formula is: ； wherein, is the amount of change in the enthalpy is the mutual information value between the steam flow rate is compared with the mutual information threshold value obtained by subtracting the standard deviation from the total mutual information average, and if the mutual information value of a certain period is lower than the mutual information threshold value, it is marked as an abnormal period in which the coupling relationship deviates, is the joint probability that the enthalpy data falls into the first interval and the flow rate data falls into the first interval, is the marginal probability that the enthalpy data falls into the first interval, is the marginal probability that the flow rate data falls into the first interval, is the total number of intervals of the variable discrete division.

6. The industrial steam flow intelligent regulation system of claim 1, wherein: The adjustment space determination module comprises: A target value acquisition submodule: obtain the preset steam target flow in each period, and construct a preset steam target flow set; The deviation judgment submodule: compare the preset steam target flow set with the steam output flow of the coupled abnormal section segment by segment, filter out the segments with deviation exceeding the steam regulation allowed threshold, and generate a regulation abnormal section set; The section marking submodule: mark the abnormal sections of the regulation abnormal section set as unregulatable state, set the corresponding flow rollback threshold according to the section number, and generate a regulation lock section and rollback threshold set.

7. The industrial steam flow intelligent regulation system of claim 6, wherein: For setting the corresponding flow rollback threshold, the formula is: ； wherein, is a flow back-off threshold of the target section for load shedding or flow limiting of the steam flow, is a target steam flow set for the target section schedule, is a maximum observed actual deviation in the target section, is a mean residual deviation after historical regulation failures, , are weight coefficients of and respectively.

8. The industrial steam flow intelligent regulation system of claim 1, wherein: The target setting adjustment module comprises: The lock response submodule: based on the lock section in the regulation lock section and rollback threshold set, perform null operation on the steam flow target value in the involved time period, mark the unregulatable state, eliminate the original target value corresponding to the section, and generate a lock section elimination result; The threshold comparison submodule: based on the remaining time period in the lock section elimination result, extract the corresponding steam output flow, compare it with the same section rollback threshold, reset part of the target value according to the comparison result, and generate a target value adjustment result; The setting summary submodule: unify the adjustment value and lock section state identification of each time period in the target value adjustment result, and generate a flow regulation control instruction set.

9. The industrial steam flow intelligent regulation system of claim 1, wherein: It also includes an execution sequence optimization module, which: sorts the flow regulation control instruction set according to the execution priority, and generates a flow regulation sequence index; The flow regulation sequence index includes sorting number, regulation time sequence relationship and priority execution identification.

10. The industrial steam flow intelligent regulation system of claim 9, wherein: The execution sequence optimization module comprises: The effectiveness screening submodule: obtains all regulation setting instructions in the flow regulation control instruction set, merges effective items according to time label, and generates a regulation target time set; The sorting generation submodule: according to the regulation target time set, set the sorting rule according to the two dimensions of time sequence and setting value size, and output a sorting priority time sequence list; The index labeling submodule: for the sorting priority time sequence list, establish the order execution label from front to back, and record the time period and regulation instruction number corresponding to the sorting position, and generate a flow regulation sequence index.

Images