Energy efficiency analysis and dynamic optimization management platform for acid separation system

Abstract

The invention relates to the technical field of industrial intelligent management, in particular to an acid separation system energy efficiency analysis and dynamic optimization management platform which comprises an electric energy monitoring module, a density difference triggering module, a variable layering response module, a flow dissipation recognition module and a synchronous suppression adjustment module. According to the method, the node response trend is constructed through the electric power difference and the electric power direction sequence, the sensitive area with significant electric power fluctuation in the system is excavated, and the dynamic hierarchical binding of key variables is realized by combining the upstream electric power conduction logic and the concentration change direction, so that the aging identification and intervention precision of response variables is greatly enhanced. By means of the consistency evaluation of the change trend of the node electric power difference and the concentration fluctuation direction, the variable with the continuous response capability can be accurately screened out, so that the control strategy has dynamic plasticity in the variable level. And analyzing the flow change ratio of the variable binding flow section in the excitation state to identify the flow intervention point with the collaborative disturbance characteristic.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial intelligent management, and particularly to an energy efficiency analysis and dynamic optimization management platform for an acid separation system. Background Art

[0002] The technical field of industrial intelligent management includes the digital, information-based, and intelligent analysis and scheduling of aspects such as energy utilization efficiency, equipment operating status, process control parameters, and resource allocation in the industrial production process. The core content of this technical field is to combine embedded sensing devices, data acquisition equipment, edge computing capabilities, and data modeling methods to monitor and record key parameters in the industrial process in real time, and then support decision-making and management optimization through the calculation results of the data.

[0003] Among them, the energy efficiency analysis and dynamic optimization management platform for the acid separation system refers to a management method for analyzing and adjusting the operating efficiency, energy consumption, and system load changes of the separation device during the treatment process of acid-containing liquids. It includes technical matters such as acid liquid state identification based on thermodynamic calculations, energy flow path tracking and measurement, separation efficiency time series evaluation, identification of changes in material characteristics during the separation stage, and dynamic switching of operating strategies based on the energy efficiency ratio. The method is to set multiple temperature and flow measurement nodes, combine the acid concentration estimation model and the electric work balance calculation results, extract the key energy efficiency indicators in the current system operating state, and then determine the target operating configuration parameters based on the comparison with the rule database and historical operating data, and implement the adjustment and optimization of the separation process.

[0004] The existing technology mainly relies on static thermodynamic calculations and comparison with historical operating data in the evaluation of the system operating state, lacking dynamic monitoring of the characteristics of electric work fluctuations and their continuous trends, resulting in a time delay problem in the identification of key variable responses. Although the input and output data of node electric power are collected, the electric work difference sequence and direction trend sequence cannot be constructed, making it difficult to form a complete perception of the internal disturbance trend of the system. For example, in the case of a sharp change in the input of node electric power, due to the failure to synchronously capture the change direction of its concentration and the peak value of the electric work deviation, potential high-risk process segments are often overlooked. The existing method fails to identify the linkage characteristics between control variables, and the adjustment of variable excitation states relies on manual experience, resulting in unstable intervention action frequencies and even over-regulation phenomena. In addition, the amplitude control of variable adjustment for equipment classification is relatively rough, and the concentration fluctuation amplitude is not used to quantify the variable excitation degree, resulting in a problem of insufficient adjustment accuracy. When the system load suddenly changes, the existing scheme is difficult to freeze high-frequency excitation variables in time, easily causing system oscillation and reducing the separation efficiency and energy consumption control ability. Generally speaking, the existing methods have obvious shortcomings in the dynamic identification and rhythm control under the multi-variable coupling state, and are difficult to meet the real-time optimization requirements in complex operating scenarios. Summary of the Invention

[0005] The object of the present invention is to solve the disadvantages existing in the prior art, and a platform for energy efficiency analysis and dynamic optimization management of an acid separation system is proposed.

[0006] To achieve the above object, the present invention adopts the following technical solutions: The platform for energy efficiency analysis and dynamic optimization management of an acid separation system includes: The electric energy monitoring module obtains the electric power input, electric power output and temperature data of each node in the acid separation system, and establishes a node electric work fluctuation sequence set; The density difference trigger module extracts the electric work difference and peak deviation records based on the node electric work fluctuation sequence set, marks the first-level response items and screens the candidate nodes with continuous offset trends to obtain a node response trend list; The variable hierarchical response module judges whether the electric work difference and the fluid concentration change direction of the candidate nodes and the upstream nodes are consistent based on the candidate nodes in the node response trend list, binds the second-level excitation item nodes, and generates a dynamic excitation table of control variables; The flow dissipation identification module identifies the flow change ratio of the flow segment bound by the excitation variable according to the control variables that are already in the excitation state in the dynamic excitation table of control variables, combines the temperature fluctuation, concentration direction trend and excitation frequency consistency, and establishes a list of fluctuation active intervention points; The synchronous suppression adjustment module classifies the disturbance variables by equipment based on the control variables bound to the marked disturbance points in the list of fluctuation active intervention points, sorts the concentration adjustment amplitudes, controls the excitation frequency and marks the frozen state, and generates the energy efficiency optimization result of the acid separation system.

[0007] As a further solution of the present invention, the node electric work fluctuation sequence set includes an electric work difference sequence, an electric power direction sequence, and an electric power fluctuation amplitude sequence; the node response trend list includes a first-level response node index, a continuous fluctuation node mark, and a trend fluctuation state identification value; the dynamic excitation table of control variables is specifically an excitation level label, a control variable binding relationship, and a response trend path identifier; the list of fluctuation active intervention points includes a disturbance node identifier, a trend consistency matching state, and a variable excitation association label; the energy efficiency optimization result of the acid separation system is specifically an equipment classification variable status table, concentration amplitude sorting information, and a frequency freeze control label.

[0008] As a further solution of the present invention, the electric energy monitoring module includes: The deviation extraction sub-module obtains the electric power input, electric power output and temperature data of each node in the acid separation system within a specified time period, extracts the electric power input and electric power output of each node at the monitoring time points respectively, calculates the numerical difference between the electric power input and electric power output of each node within the specified time period, summarizes the electric work difference values at all monitoring time points according to the node numbers, and obtains a node electric work deviation value sequence; The electric power fluctuation identification sub-module calls the node electric power deviation value sequence, determines the temperature rise and fall direction according to the temperature data at adjacent time points, divides the node electric power deviation value sequence into intervals in combination with the temperature rise and fall direction, respectively obtains the fluctuation amplitude of the electric power deviation value in each interval and calculates the change intensity within the duration, classifies and constructs a time series according to the node number, and generates a node electric power fluctuation sequence set.

[0009] As a further solution of the present invention, the density difference trigger module includes: The deviation extraction sub-module extracts the electric power difference corresponding to each node in the current cycle of the node electric power fluctuation sequence set, combines the peak value records of the electric power differences of the nodes in multiple cycles, obtains the deviation amount between the current cycle electric power difference and the maximum electric power difference in the records, compares the deviation amount with the electric power deviation threshold, screens the nodes with the deviation amount higher than the electric power deviation threshold and marks them as first-level response items, and generates a first-level response node sequence; The trend identification sub-module calls the electric power direction data of the nodes in the current cycle in the first-level response node sequence, extracts the change situation where the electric power direction of the node continuously rises or continuously falls when the deviation is in the section where the electric power deviation threshold is located, judges the trend continuity, records the nodes that meet the condition of continuous direction fluctuation as candidate items, and establishes a node response trend list.

[0010] As a further solution of the present invention, the variable hierarchical response module includes: The direction extraction sub-module obtains the change direction of the electric power difference of the candidate nodes in the current cycle based on the candidate nodes in the node response trend list, extracts the change direction of the electric power difference of the adjacent upstream nodes, collects the change direction of the fluid concentration of the candidate nodes in the same cycle, integrates the direction data according to the node number, and obtains a node direction comparison set; The excitation identification sub-module calls the node direction comparison set, judges whether the electric power direction of each candidate node is consistent with the concentration change direction, and at the same time judges whether its electric power difference is greater than the electric power difference of the upstream node, and uses the formula:

[0011] Calculate the control variable excitation value of the node , which is used to quantify the dynamic response intensity of each candidate node to the control variable in the current cycle, records the corresponding excitation state according to the node number, integrates the trend and excitation information on the basis of the first-level response, and establishes a control variable dynamic excitation table; Among them, represents the total number of time points of the node in the current cycle, represents the node at the time point electric power difference, represents the The electrical work difference of the adjacent node upstream of the node at the time point , represents the consistency coefficient of the concentration change direction of the th node at the time point , and represents the direction continuity discrimination coefficient.

[0012] As a further solution of the present invention, the flow dissipation identification module includes: The flow extraction sub-module dynamically excites the control variables that are already in the excited state in the control variable dynamic excitation table according to the control variables, extracts the number information of the bound process segments and the monitoring period range, obtains the flow monitoring data of the corresponding process segments in the current period, statistically analyzes the flow changes at the time points and summarizes them into an excited process segment flow data set; The trend consistency identification sub-module calls the excited process segment flow data set, calculates the flow change ratio of each process segment in the current period, compares it with the flow dissipation change threshold, screens the process segment numbers whose change ratio exceeds the flow dissipation change threshold, and backtracks the temperature fluctuation amplitude sequence, concentration direction trend sequence and control variable excitation frequency sequence of the corresponding process segments in the specified period according to the screening results, and judges whether the directions of the three sequences are consistent in the period. If the three trend directions are consistent, the target process segment is marked as a fluctuation active point, and a fluctuation active intervention point list is established.

[0013] As a further solution of the present invention, the synchronous suppression adjustment module includes: The variable classification sub-module classifies the control variables bound to the marked disturbance points in the fluctuation active intervention point list into three main process equipment types: heat exchangers, reactors, and separation towers according to the equipment types corresponding to the process segments to which the variables belong, collects the concentration monitoring data sequences of each variable in the current period of the process segments in each type of equipment, and uses the formula: ; Calculate the concentration change amplitude of the th control variable in the th type of equipment in the current period, which is used to measure the dynamic fluctuation degree of the process segment bound by the variable, and establish an equipment variable concentration amplitude set; Among them, represents the concentration value of the variable at the time point , represents the total number of time points in the current period, and respectively represent the maximum and minimum values in the sequence, represents the set of consecutive integers from the first time point to the Tth time point; The amplitude adjustment sub-module calls the set of device variable concentration amplitudes, constructs a variable reduction magnitude index based on the degree of deviation from the device mean and its own fluctuation intensity, and uses the formula: ; Calculate the reduction magnitude index of the th control variable in the th type of device , which is used to determine the adjustment ratio of the variable excitation amplitude, output the corresponding variable reduction excitation value, and obtain the variable reduction adjustment data table; Among them, represents the concentration change amplitude of the variable in the current period, represents the mean value of the concentration change amplitudes of all variables under the th type of device, represents the variance of the concentration change amplitudes of all variables under the th type of device, is the normalization deviation normalization factor; The alignment output sub-module calls the variable reduction adjustment data table, records the excitation call frequency of each variable in the current period after reduction, compares the frequency with the frequency threshold. If the call frequency exceeds the threshold, the target variable is marked as the frozen state, and all control variable adjustment states are classified and integrated according to the frozen and non-frozen states, and alignment analysis is carried out in combination with the concentration change trend of the current period of the affiliated device to establish the energy efficiency optimization result of the acid separation system.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In the present invention, by constructing the node response trend through the electric work difference and the electric power direction sequence, the sensitive areas with significant electric work fluctuations in the system are excavated, and combined with the upstream electric work conduction logic and the concentration change direction, the dynamic hierarchical binding of key variables is realized, greatly enhancing the time-effect recognition and intervention accuracy of the response variables. With the help of the consistency evaluation of the change trend of the node electric work difference and the concentration fluctuation direction, variables with continuous response ability can be accurately screened out, making the control strategy dynamically plastic at the variable level. By analyzing the flow change ratio of the variable binding process segment in the excitation state and combining the consistency condition of the temperature and concentration fluctuation trends, the process intervention points with cooperative disturbance characteristics are identified, thereby forming a precise definition of the fluctuation source at the system level. Further, according to the device type to which the disturbance variable belongs, combined with the deviation degree of the concentration amplitude and the fluctuation intensity of the variable itself, the variable excitation amplitude is dynamically adjusted to realize the orderly suppression of the adjustment rhythm of the control variable, avoiding the system risks of over-response and frequent disturbance. Finally, the frozen state is set based on the variable excitation frequency threshold and aligned with the concentration trend of the current period of the device, promoting the system to make a smooth transition while maintaining efficiency, taking into account the optimal allocation of separation quality and energy consumption, and promoting the transformation of energy efficiency optimization from static rule-driven to multi-dimensional fluctuation linkage strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is the platform flowchart of the present invention; Figure 2 is the flowchart of the electric energy monitoring module of the present invention; Figure 3 is the flowchart of the density difference trigger module of the present invention; Figure 4 is the flowchart of the variable stratification response module of the present invention; Figure 5 is the flowchart of the flow dissipation identification module of the present invention; Figure 6 is the flowchart of the synchronous suppression adjustment module of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention.

[0017] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0018] Please refer to Figure 1 , the energy efficiency analysis and dynamic optimization management platform of the acid separation system includes: The electric energy monitoring module obtains the electric power input, electric power output and temperature data of each node in the acid separation system, and establishes a node electric work fluctuation sequence set; The density difference trigger module extracts the electric work difference and peak deviation records based on the node electric work fluctuation sequence set, marks the first-level response items and screens the candidate nodes with continuous offset trends to obtain a node response trend list; The variable stratification response module determines whether the electric work difference between the candidate node and the upstream node and the change direction of the fluid concentration are consistent based on the candidate nodes in the node response trend list, binds the second-level excitation item nodes, and generates a control variable dynamic excitation table; The flow dissipation identification module dynamically excites the control variables that are already in the excited state in the control variable dynamic excitation table, identifies the flow change ratio of the process segment bound by the excitation variable, and combines the temperature fluctuation, the concentration direction trend, and the excitation frequency consistency to establish a list of fluctuation active intervention points; The synchronous suppression adjustment module classifies the disturbance variables by device based on the control variables bound to the marked disturbance points in the list of fluctuation active intervention points, sorts the concentration adjustment amplitudes, controls the excitation frequency, and marks the frozen state to generate the energy efficiency optimization result of the acid separation system; The node electric power fluctuation sequence set includes the electric power difference sequence, the electric power direction sequence, and the electric power fluctuation amplitude sequence; the node response trend list includes the primary response node index, the continuous fluctuation node mark, and the trend fluctuation state identification value; the control variable dynamic excitation table specifically includes the excitation level label, the control variable binding relationship, and the response trend path identifier; the fluctuation active intervention point list includes the disturbance node identifier, the trend consistency matching state, and the variable excitation association label; the energy efficiency optimization result of the acid separation system specifically includes the device classification variable status table, the concentration amplitude sorting information, and the frequency freeze control label.

[0019] Please refer to Figure 2 , the electric energy monitoring module includes: The deviation extraction sub-module obtains the electric power input, electric power output, and temperature data of each node in the acid separation system within a specified time period, extracts the electric power input and electric power output of each node at the monitoring time points respectively, calculates the numerical difference between the electric power input and electric power output of each node within the specified time period, summarizes the electric power difference values at all monitoring time points according to the node numbers, and obtains the node electric power deviation value sequence; To obtain the electric power input, electric power output and temperature data of each node in the acid separation system, it is necessary to establish data acquisition channels for three typical nodes: heat exchanger, reactor and separation tower. For example, for the reactor node, set the acquisition period to 10 minutes, set the start and end times to 8:00 to 10:00, obtain a total of 13 time point data, extract the electric power input data and electric power output data respectively at each time point, with the unit unified as kilojoule. At the same time, synchronously read the temperature sensor to obtain the surface temperature of the node, forming an electric power input list of 3200, 3400, 3300, 3450, 3550, an electric power output list of 3000, 3100, 3050, 3150, 3220, and a temperature list of 120.5, 121.0, 120.8, 121.5, 122.3. Establish mapping relationships for the data of each time point to form data groups. The electric power input and electric power output are paired one by one according to the time point. For example, at the first time point, the electric power input is 3200 and the electric power output is 3000, with a corresponding difference of 200. Arrange such electric power differences in chronological order to form a sequence of 200, 300, 250, 300, 330. Summarize by node name. The electric power differences of the heat exchanger are 150, 170, 160, 180, 190, and those of the separation tower are 210, 220, 205, 225, 240. The numerical values of the electric power input and output of each node are all taken from the continuous recorded values of the sensor, without using externally preset data or average values. Through this method, the electric power input and output processing processes of the three nodes are completed respectively, and a complete electric power deviation sequence for each node is constructed, and finally a sequence of node electric power deviation values is obtained.

[0020] The electric power fluctuation identification sub-module calls the sequence of node electric power deviation values, determines the temperature rise and fall direction according to the temperature data of adjacent time points, divides the sequence of node electric power deviation values according to the temperature rise and fall direction, obtains the fluctuation amplitude of the electric power deviation value in each interval respectively, calculates the change intensity during the duration, classifies by node number to construct a time series, and generates a set of node electric power fluctuation sequences; Call the node electric power deviation value sequence. Take the heat exchanger as an example. Collect the temperature sequence 95.2, 96.0, 95.7, 96.5, and 97.1 at 10-minute intervals. Calculate the temperature difference between adjacent time points. The temperature rise is marked as rising, the temperature drop is marked as falling, and the temperature remains unchanged as flat. The corresponding temperature change direction is rising, falling, rising, and rising. The above direction information is used as the basis for dividing the electric power deviation value. The corresponding electric power deviation values are 150, 170, 160, 180, and 190. In the "rising" direction, select the electric power deviation values of 170, 180, and 190 in the 2nd, 4th, and 5th minutes. The fluctuation amplitude of this section is calculated as 190 minus 170 to get 20 kilojoules. The duration The corresponding "downward" direction is from the 2nd to the 3rd minute, the electric power deviation values are 170 and 160, the fluctuation amplitude is 10 kilojoules, and the variation intensity is 100. The variation intensity of each direction segment is summarized to form a variation intensity set of 400 and 100. Similarly, the reactor temperature sequence is 120.5, 121.0, 120.8, 121.2, and 122.1, and the corresponding electric power deviation values are 200, 300, 250, 270, and 290. The directions are divided and the segments are extracted to calculate the fluctuation amplitude and variation intensity. Finally, the electric power fluctuation information is combined to establish the electric power fluctuation sequence set of each node, and the node electric power fluctuation sequence set is generated.

[0021] See also Figure 3 , the density difference trigger module includes: The deviation extraction submodule extracts the power difference corresponding to the current cycle of each node in the node power fluctuation sequence set, combines the power difference peak records of the node in multiple cycles, obtains the deviation between the power difference of the current cycle and the maximum power difference in the record, compares the deviation with the power deviation threshold, selects nodes with deviations higher than the power deviation threshold and marks them as first-level response items, and generates a first-level response node sequence; First, set a cycle range of 5 consecutive monitoring cycles for each node. Extract the current cycle's electric energy difference as the electric energy difference corresponding to the 6th cycle. Record the peak value that appears in the electric energy differences within the historical 5 cycles as the historical maximum electric energy difference. The electric energy difference is derived from the difference sequence arranged in time in the node electric energy fluctuation sequence set generated by the previous module. Then, calculate the difference between the current cycle's electric energy difference and the historical maximum electric energy difference. For example, if the current cycle's electric energy difference of node A is 1240W and the peak value of the electric energy differences in the historical five cycles is 1100W, then the deviation value is 1240 - 1100 = 140W. Compare this deviation value with the preset electric energy offset threshold in the system. Assume the electric energy offset threshold is 130W. Then this deviation value exceeds the threshold, and it is determined that this node is a first-level response item. For such judgments, intervals need to be divided and the judgment boundaries need to be clarified. For example, set the threshold range to [0, 150]W, where less than 130W is a non-response item, 130–150W is the transition zone, and more than 150W is a severely offset node. The current node's deviation value of 140W falls within the response interval. During the judgment process, it is necessary to ensure that the threshold is a dynamically set value. This threshold should be set based on 1.2 times the maximum average electric energy difference of all nodes in the historical operation cycle. For example, in 40 monitoring cycles, the average value of the maximum electric energy differences of all nodes is 108W, then the offset threshold is 108×1.2 = 129.6W, which is rounded up to 130W. Furthermore, it can ensure that the set threshold has statistical rationality and accurately compare with the differences of each node. Nodes are marked as first-level response items on the basis of meeting this threshold, and a first-level response node sequence is formed by arranging them according to the node numbers, and finally a first-level response node sequence is generated.

[0022] The trend recognition sub-module calls the current cycle's electric power direction data of the nodes in the first-level response node sequence, extracts the change situation where the electric power direction continuously rises or continuously falls in the section where the deviation is near the electric energy offset threshold of the nodes, judges the trend continuity, records the nodes that meet the condition of continuous direction fluctuation as candidates, and establishes a node response trend list; Call the electric power direction information of the nodes included in the first-level response node sequence, extract the changes in the electric power direction of these nodes in the current period and the previous adjacent period. The electric power direction is judged based on the sign of the electric power input minus the electric power output. If the current electric power difference is positive, it is defined as the inflow direction; if it is negative, it is defined as the outflow direction. In this way, an electric power direction sequence can be constructed for each node. For example, if the electric power input of node B in the 6th period is 1850W and the electric power output is 2100W, then the electric power difference is -250W, and the direction is the outflow direction. Combining the direction of the previous period, which is -220W, the direction is continuously the outflow direction. Then, judge whether the deviation value falls within the section near the offset threshold. Here, the nearby section is defined as the floating range of ±10W, that is, if the deviation value falls between 120 - 140W, it is close to the threshold. If the deviation value of a certain node is 135W and the direction is continuously the outflow direction, it meets the candidate conditions, and this node is recorded as a candidate item. Finally, summarize the node numbers and corresponding electric power directions of all nodes that meet the conditions of continuous direction fluctuation and deviation near the threshold, form a record item according to the combination of the node number and the direction value, construct a structured set composed of the node number and the direction identifier, and finally establish a node response trend list.

[0023] Please refer to Figure 4 , the variable hierarchical response module includes: The direction extraction sub-module, based on the candidate nodes in the node response trend list, obtains the change direction of the electric power difference of the candidate nodes in the current period, extracts the change direction of the electric power difference of the adjacent upstream nodes, collects the change direction of the fluid concentration of the candidate nodes in the same period, and integrates the direction data according to the node number to obtain a node direction comparison set; First, the identification numbers of these nodes need to be extracted from the system, and then the change direction data of the electric power difference of these nodes in the current control period are obtained from the database. The electric power difference data are calculated in real time. For example, if the electric power input of a certain node is 1720J and the electric power output is 1610J during the period from 08:00 to 08:10 on April 10, 2025, then the corresponding electric power difference is 110J, and the direction is identified by the change trend (rising, falling, or flat) of the electric power difference at consecutive time points, taking values of 1, -1, or 0. Further, the numbers of its upstream adjacent nodes are extracted and the change direction of the electric power difference in the same period is queried. Suppose the electric power difference of the upstream node is 95J and the corresponding direction is rising, which is 1. After the directions between the nodes are associated in this way, the system collects the change direction of the fluid concentration of the candidate nodes in the current period. For example, if the concentrations at 08:00 and 08:10 are obtained as 5.2% and 5.7% respectively through an online densitometer, then it is judged that the change direction of the concentration is rising, and the direction value is 1. Finally, the electric power difference direction, the electric power direction of the upstream node, and the change direction of the concentration of this node are combined and output as a set of records. For example, the three-direction record of node X is [1, 1, 1], which means that the three directions are consistent, and the acquisition of the node direction comparison set is completed.

[0024] The excitation recognition sub-module calls the node direction comparison set to judge whether the electric power direction of each candidate node is consistent with the concentration change direction, and at the same time judge whether its electric power difference is greater than that of the upstream node. The formula is as follows:

[0025] Calculate the excitation value of the control variable of the th node , which is used to quantify the dynamic response intensity of each candidate node to the control variable in the current cycle. Record the corresponding excitation state according to the node number, integrate the trend and excitation information on the basis of the first-level response, and establish a dynamic excitation table of the control variable; Among them, represents the total number of valid time points of the th node in the current cycle, represents the summation of the time points , represents the electric power difference of the th node at the time point , represents the electric power difference of the adjacent upstream node of the th node at the time point , represents the consistency coefficient of the concentration change direction of the th node at the time point (the value of 1 means that the concentration direction is consistent with the electric power direction, and the value of 0 means inconsistent), represents the direction continuity discrimination coefficient (the value of 1 means that the direction is continuous, and the value of 0 means discontinuous).

[0026] Call the node direction comparison set. For each candidate node, first judge whether its electric power direction is consistent with the concentration change direction. Here, the direction values are all ±1 or 0. Judge whether they are consistent by judging the equality of the direction values. If they are consistent, set the consistency coefficient to 1, otherwise to 0. For example, if the electric power direction of node Y in a certain cycle is 1 and the concentration direction is 1, then the consistency coefficient is 1. Then judge whether the electric power difference of this node is greater than that of its upstream node. Suppose the electric power difference of node Y is 110J and the electric power difference of the upstream node is 95J, then it is determined to be established. The nodes that meet the double conditions enter the excitation calculation link, and the following formula is used:

[0027] Among them, the set parameters are as follows: node number , total number of valid time points , the electric power differences of node 3 at the 1st to 3rd time points are 110J, 115J, and 120J respectively, and the electric power differences of the upstream nodes are 95J, 98J, and 99J respectively, and the consistency coefficient (Same direction), continuous direction consistency discriminant value (Continuous rise).

[0028] Substitute into the formula and calculate as follows: The first term: ; The second term: ; The third term: ; Sum and take the average: .

[0029] This result shows that the excitation value of the control variable of node 3 is 0.1812. If the reference threshold of the excitation value set by the system is 0.15, the excitation value of node 3 exceeds the threshold. The control variable bound to it should be marked as a second-level excitation item, and the excitation status value of "activated" should be appended to the original first-level response item of this node, and the response trend is "continuous rise", so as to form a dynamic excitation table of control variables including four items: number, response status, trend direction, and excitation status.

[0030] The benefit of the formula is that by introducing the consistency coefficient and the continuous direction indicator , it avoids the interference of direction fluctuations on the excitation value and enhances the continuous recognition ability of excitation conditions; through the proportional operation of normalizing the increase in electric work difference , it enables nodes with different electric work benchmarks to have a unified response measurement scale, ensuring the comparability of different nodes under the background difference of electric work.

[0031] Please refer to Figure 5 , the flow dissipation identification module includes: The flow extraction sub-module extracts the number information and monitoring period range of the bound process segment according to the control variables that are already in the excitation state in the dynamic excitation table of control variables, obtains the flow monitoring data of the corresponding process segment in the current period, and statistically analyzes the flow changes at time points and summarizes them into an excitation process segment flow data set; First, each record in the excitation table should be parsed. The control variables in the excitation state should be extracted item by item, and the corresponding process section numbers and monitoring time ranges bound to them should be obtained. In practical engineering applications, if the excitation state of the control variable of a certain reaction section, such as the feed valve A01, is "activated", the flow measurement point number associated with this valve, such as FIC-201, should be extracted, and at the same time, the corresponding time series information should be called. For example, from April 1, 2025 to April 7, 2025, the flow sequence automatically recorded by the corresponding system monitoring platform is once per hour. Subsequently, the flow monitoring values within this time period are exported and preliminarily screened to eliminate missing points and outliers, usually by setting reasonable interval limits for the flow values. For example, the flow interval of the FIC-201 monitoring point is set to 2 m³ / h to 8 m³ / h. If the data at a certain time point is less than 1 m³ / h or higher than 10 m³ / h, it is regarded as abnormal and eliminated; in this way, the cleaned flow data sequence is obtained. For example, if the effective data sequence within a 7-day cycle is [3.2, 3.5, 3.6, 4.0, 3.9, 4.2, 3.8] (unit: m³ / h), it can be regarded as the basic flow data set of the process section bound by this control variable within this cycle. This data set will be used in subsequent links to identify flow changes and dissipation judgment. Therefore, the finally obtained result is the excitation process section flow data set.

[0032] The trend consistency identification sub-module calls the excitation process section flow data set, calculates the flow change ratio of each process section within the current cycle, compares it with the flow dissipation change threshold, screens the process section numbers with a change ratio exceeding the flow dissipation change threshold, and backtracks the temperature fluctuation amplitude sequence, concentration direction trend sequence, and control variable excitation frequency sequence of the corresponding process section within the specified cycle according to the screening results, and judges whether the directions of the three sequences are consistent within the cycle. If the three trend directions are consistent, mark this process section as a fluctuation active point and establish a list of fluctuation active intervention points; Analyze the flow rate change ratio within the cycle of each process segment in time series. Define the average value of the first and last days of the cycle as a reference. For example, the flow rate sequence of the above-mentioned FIC-201 is [3.2, 3.5, 3.6, 4.0, 3.9, 4.2, 3.8]. Assume that the average flow rate of the first segment of the cycle is the average of the first three days: (3.2 + 3.5 + 3.6) / 3 = 3.43 m³ / h, and the average flow rate of the last segment is the average of the last three days: (4.0 + 3.9 + 4.2) / 3 = 4.03 m³ / h. Then the change ratio is: (4.03 - 3.43) / 3.43 = 0.175, that is, the change ratio is 17.5%. If the preset flow dissipation change threshold of the system is set to 15%, then this change ratio exceeds the threshold, and the process segment needs to be traced back. During the tracing back process, extract the temperature fluctuation range, concentration direction trend, and control variable excitation frequency of this process segment in the interval from April 1st to April 7th, 2025. The temperature fluctuation can be calculated by the range of the temperature data sequence in the same cycle. For example, if the temperature sequence is [75.2, 76.1, 77.5, 75.8, 76.3, 78.0, 76.5], then the fluctuation range is 78.0 - 75.2 = 2.8 °C. The concentration direction trend is determined by whether the concentration values at each time point change monotonically. For example, the sequence [3.0%, 3.2%, 3.4%, 3.5%, 3.6%, 3.7%, 3.8%] is a continuously rising trend. The excitation frequency is measured in the number of days the excitation state lasts. For example, if this control variable is continuously in the excitation state for 6 days out of 7 days, then the frequency is 6 / 7 = 0.857. Judge that the three trend directions are consistent, that is, require the temperature fluctuation to be in the increasing direction (the amplitude increases), the concentration direction to be a unidirectional trend, and the excitation frequency to be high (such as higher than 0.75). In this example, all three meet the consistency judgment criteria. Therefore, mark this process segment as a fluctuation active point, and finally establish a list of fluctuation active intervention points.

[0033] Please refer to Figure 6 , and the synchronous suppression adjustment module includes: Based on the control variables bound to the marked disturbance points in the list of fluctuation active intervention points, the variable classification sub-module classifies the variables into three main process equipment types: heat exchangers, reactors, and separation towers according to the equipment types corresponding to the process segments they belong to. Collect the concentration monitoring data sequences of each variable in the current cycle of the process segments in each type of equipment, and use the formula: ; Calculate the concentration change amplitude of the th control variable in the th type of equipment in the current cycle, which is used to measure the dynamic fluctuation degree of the process segment bound by the variable, and establish a set of equipment variable concentration amplitudes; Among them, represents the concentration value of this variable at time point , represents the total number of time points within the current cycle, and respectively represent the maximum and minimum values in the sequence, represents a set of consecutive integers from the 1st time point to the Tth time point.

[0034] First, the device types of the process segments bound to each variable need to be extracted. Usually, this information already exists in the industrial control system tag structure. For example, for the control variable CV_001, the corresponding process segment identifier is R-203, and the system device identifier field is marked as a reactor. In this step, an array mapping variables to device types needs to be established , and the control variables are classified into three device categories according to the device type; subsequently, during the actual operation cycle, data extraction is performed on the concentration monitoring points of the process segments corresponding to each variable. For example, within the current cycle, the sampling frequency of CV_001 is once per minute, and the time period coverage is to , and the concentration sequence that can be obtained is (unit: mol / L). Then, the maximum and minimum values of this sequence are taken, and the concentration change amplitude is calculated according to the formula . If , , then . This operation is performed on all control variables in each type of device, and finally, a set of control variable concentration change amplitudes within each device is established as the basic data for subsequent adjustment. represents the concentration change amplitude of the th variable in the st type of device, is the monitoring value of this variable at time point , is the total number of time points in the current cycle. In a specific example , no thresholds, weights, or fuzzy terms are involved in this step. The data acquisition process is clear, and the result is a set of device variable concentration amplitudes.

[0035] The amplitude adjustment sub-module calls the set of device variable concentration amplitudes, constructs a variable reduction magnitude index based on the degree of deviation from the device mean and its own fluctuation intensity, and uses the formula: ; Calculate the reduction magnitude index of the th control variable in the th type of device to determine the adjustment ratio of the variable excitation amplitude, output the corresponding variable reduction excitation value, and obtain the variable reduction adjustment data table; Among them, represents the concentration change amplitude of this variable in the current cycle, represents the The mean of the concentration change range of all variables under the equipment type, denotes the variance of the concentration change range of all variables under the equipment type, is the normalization deviation standardization factor.

[0036] Call the equipment variable concentration amplitude set, and perform normalization and standard deviation difference analysis on the concentration change amplitude of each variable under the three equipment of heat exchanger, reactor, and separation tower. First, the average concentration change amplitude under each equipment type needs to be calculated , for example, for the reactor equipment type, the concentration change amplitude array is , then its average value is , and the variance calculation is Substitute into the formula:

[0037] Taking the variable concentration amplitude as an example, , , substituting gives: ; Finally, the reduction magnitude index of this variable is obtained as 0.00126, indicating that the adjustment of its excitation amplitude should refer to this value for reduction operations. When actually performing the excitation amplitude adjustment operation, the reduction magnitude index can be used as the reduction ratio factor for variable excitation signal adjustment. The specific reference method is as follows: regard this index as the relative adjustment ratio of the original excitation amplitude, that is, assume the original excitation amplitude is , then the reduced excitation amplitude can be adjusted as follows: ; Taking actual values as an example, if the original excitation amplitude of a certain control variable is (the unit can be specific dimensions such as opening ratio, setting coefficient, etc.), and the reduction magnitude index , then the adjusted excitation amplitude is: ; The operation shows that the original excitation signal will be reduced by about 0.13% to gradually suppress the influence on variables with large concentration fluctuation amplitudes, thereby balancing the equipment system load and stability. Therefore, is the dynamic proportional coefficient of the control variable adjustment amplitude, and the larger the value, the stronger the adjustment.

[0038] The paired output sub-module calls the variable reduction adjustment data table, records the excitation call frequency of each variable in the current cycle after reduction, compares the frequency with the frequency threshold, if the call frequency exceeds the threshold, mark the variable as frozen, classify and integrate the adjustment status of all control variables according to the frozen and unfrozen states, and perform paired analysis in combination with the current cycle concentration change trend of the affiliated equipment to establish the energy efficiency optimization result of the acid separation system; Call the variable reduction adjustment data table to extract the adjusted excitation call frequency of each control variable in the current cycle. This frequency is the number of cycles in which the variable excitation signal is scheduled and controlled by the system to execute. For example, if a certain control variable is triggered by the excitation signal 12 times in total and the total number of minutes in the cycle is 60 minutes, then its frequency is , and call the frequency threshold set by the system as . According to the judgment operation: if the excitation frequency , then the variable status is set to frozen. For example, if the current excitation frequency of this variable is 0.2 and it meets the condition of being greater than the threshold, the system will mark the status of this variable as "frozen", and the identification field in the status classification result is set to a Boolean value (1 means frozen, 0 means not frozen). The status list can be expressed as: . Then, according to the device to which the variable belongs, match the concentration change trend of its device in the current cycle. For example, if the concentration sequence of device E-301 in this cycle is , then the trend is an upward trend, and the association result is [CV_001: frozen, upward]. This type of result constitutes the state set data of the final system, and the result is the energy efficiency optimization result of the acid separation system.

[0039] The above is only a preferred embodiment of the present invention, and it does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. Acid separation system energy efficiency analysis and dynamic optimization management platform, characterized in that, The platform includes: The electric energy monitoring module obtains the electric power input, electric power output and temperature data of each node in the acid separation system, and establishes a node electric work fluctuation sequence set; The density difference triggering module extracts the electric work difference and peak deviation records based on the node electric work fluctuation sequence set, marks the first-level response items and screens the candidate nodes with continuous offset trends to obtain a node response trend list; The variable hierarchical response module determines whether the electric work difference and the fluid concentration change direction between the candidate nodes and the upstream nodes are consistent based on the candidate nodes in the node response trend list, binds the second-level excitation item nodes, and generates a control variable dynamic excitation table; The flow dissipation identification module identifies the flow change ratio of the flow segment bound by the excitation variable according to the control variables that are already in the excitation state in the control variable dynamic excitation table, combines the temperature fluctuation, the concentration direction trend and the excitation frequency consistency, and establishes a fluctuation active intervention point list; The synchronous suppression adjustment module classifies the disturbance variables by equipment based on the control variables bound to the marked disturbance points in the fluctuation active intervention point list, sorts the concentration adjustment amplitude, controls the excitation frequency and marks the frozen state, and generates the energy efficiency optimization result of the acid separation system.

2. The energy efficiency analysis and dynamic optimization management platform for the acid separation system according to claim 1, wherein The node electric work fluctuation sequence set includes an electric work difference sequence, an electric power direction sequence, and an electric power fluctuation amplitude sequence; the node response trend list includes a first-level response node index, a continuous fluctuation node mark, and a trend fluctuation state identification value; The control variable dynamic excitation table is specifically an excitation level label, a control variable binding relationship, and a response trend path identifier; the fluctuation active intervention point list includes a disturbance node identifier, a trend consistency matching state, and a variable excitation association label; the energy efficiency optimization result of the acid separation system is specifically an equipment classification variable status table, a concentration amplitude sorting information, and a frequency freeze control label.

3. The energy efficiency analysis and dynamic optimization management platform for the acid separation system according to claim 1, characterized in that, The electric energy monitoring module includes: The deviation extraction sub-module obtains the electric power input, electric power output and temperature data of each node in the acid separation system within a specified time period, extracts the electric power input and electric power output of the node at each monitoring time point respectively, calculates the numerical difference between the electric power input and electric power output of each node within the specified time period, summarizes the electric work difference values at all monitoring time points according to the node number, and obtains a node electric work deviation value sequence; The electric work fluctuation identification sub-module calls the node electric work deviation value sequence, determines the temperature rise and fall direction according to the temperature data of adjacent time points, divides the node electric work deviation value sequence according to the temperature rise and fall direction, obtains the fluctuation amplitude of the electric work deviation value in each interval respectively and calculates the change intensity within the duration, classifies and constructs a time series according to the node number, and generates a node electric work fluctuation sequence set.

4. The energy efficiency analysis and dynamic optimization management platform for the acid separation system according to claim 3, characterized in that The density difference triggering module includes: The deviation extraction sub-module extracts the electric work difference corresponding to each node in the current cycle of the node electric work fluctuation sequence set, combines the electric work difference peak records of the node in multiple cycles, obtains the deviation amount between the current cycle electric work difference and the maximum electric work difference in the record, compares the deviation amount with the electric work offset threshold, screens the nodes with deviation amounts higher than the electric work offset threshold and marks them as first-level response items, and generates a first-level response node sequence; The trend recognition sub-module calls the electric power direction data of the nodes in the current cycle of the first-level response node sequence, extracts the change conditions where the power direction of the nodes continuously rises or continuously falls in the section where the deviation is within the electric work offset threshold, judges the trend continuity, records the nodes that meet the condition of continuous directional fluctuation as candidates, and establishes a node response trend list.

5. The energy efficiency analysis and dynamic optimization management platform for the acid separation system according to claim 4, characterized in that The variable hierarchical response module includes: The direction extraction sub-module obtains the direction of change in the electric work difference of the candidate nodes in the current cycle based on the candidate nodes in the node response trend list, extracts the direction of change in the electric work difference of the adjacent upstream nodes, collects the direction of change in the fluid concentration of the candidate nodes in the same cycle, integrates the direction data according to the node numbers, and obtains a node direction comparison set; The excitation recognition sub-module calls the node direction comparison set to judge whether the electric power direction of the candidate node is consistent with the concentration change direction, and at the same time judge whether the electric power difference is greater than the electric power difference of the upstream node, using the formula: ; Calculate the control variable excitation value of the node , which is used to quantify the dynamic response intensity of each candidate node to the control variable in the current period, record the corresponding excitation state according to the node number, integrate the trend and excitation information on the basis of the first-level response, and establish a dynamic excitation table of the control variable; Among them, represents the total number of time points of the node within the current cycle, represents the electric power difference of the node at the time point, represents the electric power difference of the adjacent upstream node of the node at the time point, represents the consistency coefficient of the concentration change direction of the node at the time point, represents the direction continuity discrimination coefficient.

6. The energy efficiency analysis and dynamic optimization management platform for the acid separation system according to claim 5, characterized in that The flow dissipation identification module includes: The flow extraction sub-module extracts the number information of the bound process section and the monitoring period range according to the control variables that are already in the excited state in the control variable dynamic excitation table, obtains the flow monitoring data of the corresponding process section in the current cycle, counts the flow changes at each time point and summarizes them into an excited process section flow data set; The trend consistency identification sub-module calls the excited process section flow data set, calculates the flow change ratio of each process section in the current cycle, compares it with the flow dissipation change threshold, screens the process section numbers whose change ratio exceeds the flow dissipation change threshold, and backtracks the temperature fluctuation amplitude sequence, concentration direction trend sequence and control variable excitation frequency sequence of the corresponding process section in the specified cycle according to the screening results, and judges whether the directions of the three sequences are consistent within the cycle. If the three trend directions are consistent, the target process section is marked as a fluctuation active point, and a fluctuation active intervention point list is established.

7. The energy efficiency analysis and dynamic optimization management platform for the acid separation system according to claim 6, wherein The synchronous suppression adjustment module includes: The variable classification sub-module classifies variables into three main process equipment types: heat exchangers, reactors, and separation towers, based on the control variables bound to the marked disturbance points in the fluctuation active intervention point list, according to the equipment types corresponding to the process segments to which the variables belong. It collects the concentration monitoring data sequences of the current cycle of each process segment where each variable in each type of equipment is located, and uses the formula: ; Calculate the concentration change amplitude of the th control variable in the current cycle in the th type of device, which is used to measure the dynamic fluctuation degree of the process segment bound by the variable, and establish a set of concentration amplitudes of device variables; Among them, represents the concentration value of the variable at the time point ; represents the total number of time points within the current cycle, and respectively represent the maximum and minimum values in the sequence, represents the set of consecutive integers from the 1st time point to the Tth time point; The amplitude adjustment sub-module calls the set of device variable concentration amplitudes, and jointly constructs a variable reduction magnitude index according to the degree of deviation from the device mean value and its own fluctuation intensity, using the formula: ; Calculate the reduction magnitude index of the th control variable in the th type of device, to determine the adjustment ratio of the variable excitation amplitude, output the corresponding variable reduction excitation value, and obtain the variable reduction adjustment data table; Among them, represents the amplitude of the concentration change of the variable in the current period, represents the average value of the amplitude of the concentration change of all variables under the represents the variance of the amplitude of the concentration change of all variables under the type of equipment, is the normalization deviation normalization factor; The alignment output sub-module calls the variable reduction adjustment data table, records the excitation call frequency of each variable in the current cycle after reduction, compares the frequency with the frequency threshold. If the call frequency exceeds the threshold, the target variable is marked as the frozen state, classifies and integrates the adjustment states of all control variables according to the frozen and non-frozen states, and performs alignment analysis in combination with the concentration change trend of the affiliated equipment in the current cycle to establish the energy efficiency optimization result of the acid separation system.

Images