An energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction

Abstract

The present application relates to the technical field of energy consumption and carbon emission collaborative optimization, and particularly relates to an energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction, comprising the following steps: constructing a process constraint topology and identifying an active constraint to determine the transmission path of material, energy and control relationship to the process state boundary, and reducing invalid search caused by fixed boundary; reversing the process allowed interval along the active propagation chain to reconstruct the local process boundary of the current working condition and exclude the unexecutable parameter range; selecting the synchronous decline area through the energy consumption and carbon emission decline interval to avoid the deterioration of another index caused by single optimization; and eliminating the multi-constraint conflict by intersecting and comparing the collaborative decline value of the parameter value of the multiple active propagation chains, so as to improve the executability, collaboration and calculation efficiency of the optimization result.

Description

Technical Field

[0001] This invention relates to the technical field of energy consumption and carbon emission synergistic optimization, and in particular to a method for energy consumption and carbon emission synergistic optimization based on process constraint boundary reconstruction. Background Technology

[0002] During the operation of the prebaked anode aluminum electrolytic cell group, alumina enters the electrolyte system through the feeding device, the anode lifting mechanism adjusts the anode position, and the tank shell cooling device changes the heat dissipation state of the tank. All control actions together affect the alumina concentration, electrolyte temperature, anode-cathode distance, tank voltage, tank voltage fluctuation, and equipment load. Reducing the anode-cathode distance can reduce the tank voltage, but may increase the fluctuation within the tank and reduce the safe operating margin. Changing the single alumina feeding amount or feeding interval will change the alumina concentration and the risk of anode effect. Although increasing the heat dissipation intensity of the tank can suppress the rise in electrolyte temperature, it may increase heat loss and power consumption of auxiliary equipment.

[0003] Existing control methods typically set fixed upper and lower limits for each process state, and then adjust the anode position, alumina feeding, and tank heat dissipation parameters based on a single control loop or local empirical rules. This makes it difficult to reflect the chain effect of the same control parameter on multiple process states through material, energy, and control relationships. The resulting control parameter range may include parameter combinations that are not executable under the current operating conditions.

[0004] Given that existing aluminum electrolysis process optimization technologies suffer from problems such as fixed process constraint boundaries, missing constraint action paths, conflicting control parameters among multiple constraints, and fragmented evaluation of energy consumption and carbon emissions, this invention proposes a collaborative optimization method for energy consumption and carbon emissions based on the reconstruction of process constraint boundaries. This method aims to address the challenges of dynamically determining control parameter boundaries and simultaneously optimizing energy consumption and carbon emissions in aluminum electrolysis cell clusters under the coupling of multiple process constraints. Summary of the Invention

[0005] The purpose of this section is to outline some aspects of the embodiments of the present invention and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this section, the abstract and title of the invention. Such simplifications or omissions shall not be used to limit the scope of the present invention.

[0006] In view of the aforementioned existing problems, the present invention is proposed.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A collaborative optimization method for energy consumption and carbon emission based on process constraint boundary reconstruction includes: constructing a process constraint topology based on material, energy and control relationships; converting control parameter intervals into process state intervals; determining constraints that reach or exceed the endpoints of the process allowable intervals as activation constraints; and tracing back control parameters to form an activation propagation chain. The process allowable interval of the activation constraint is converted into the parameter constraint interval in reverse along the activation propagation chain, and the intersection with the control parameter interval is used to form the local process boundary; Within the local process boundary, the control parameters are adjusted bidirectionally, and the energy consumption and carbon emission reduction ranges are calculated separately. The minimum of the two lower limits is taken as the synergistic reduction value, forming a synergistic reduction parameter domain where the lower limits are both greater than zero. The intersection of the value ranges of the same control parameter associated with multiple activation propagation chains in each collaborative descent parameter domain is obtained, and the intersection is combined. The combination with the largest collaborative descent value is taken as the collaborative optimization parameter combination.

[0008] The beneficial effects of this invention are as follows: By constructing a process constraint topology and identifying activation constraints, this invention clarifies the transmission path of material, energy, and control relationships to the process state boundary, reducing invalid searches caused by fixed boundaries; by reconstructing the local process boundary of the current operating condition by reversely converting the process allowable interval along the activation propagation chain, it eliminates the range of unexecutable parameters; by screening synchronously decreasing regions through the energy consumption and carbon emission reduction ranges, it avoids the deterioration of one indicator due to single-item optimization; by finding the intersection of the values ​​of the associated parameters of multiple activation propagation chains and comparing the coordinated reduction values, it eliminates multiple constraint conflicts and improves the executability, coordination, and computational efficiency of the optimization results. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart illustrating the energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction as shown in this invention. Detailed Implementation

[0010] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0011] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort should fall within the scope of protection of this invention.

[0012] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0013] The following describes in detail a specific implementation of an energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction, using a process optimization scenario of a group of prebaked anode aluminum electrolytic cells as an example. The implementation uses cell number 12 out of 32 400kA prebaked anode aluminum electrolytic cells in the electrolysis workshop as the target cell. The controlled objects include the anode lifting mechanism, alumina feeding device, and cell shell cooling fan. Control parameters include anode position adjustment, single alumina feeding amount, alumina feeding interval, and cooling fan operating frequency. Process status includes anode-cathode distance, electrolytic cell operating voltage, electrolyte temperature, alumina concentration, electrolyte superheat, flue gas negative pressure, flue gas flow rate, and electrolytic cell load. The cell controller, voltage acquisition device, series current acquisition device, electrolyte temperature measurement device, alumina concentration analysis device, flue gas pressure transmitter, flue gas flow meter, and power metering device respectively acquire the control parameters and process status, and merge them according to the target electrolytic cell number and acquisition time.

[0014] In this embodiment, a process calculation period of 600s is used. The sampling interval for the electrolytic cell operating voltage and series current is 1s, and the sampling interval for electrolyte temperature, flue negative pressure, flue gas flow rate and auxiliary equipment power is 10s. The anode position, alumina feeding action and cooling fan operating frequency are obtained when the control command changes. The sampled values ​​within the same process calculation period are arranged in chronological order. The voltage, current and auxiliary equipment power are taken as the arithmetic mean, and the electrolyte temperature, flue negative pressure and flue gas flow rate are taken as the effective measurement values ​​adjacent to the start time of the process calculation period. If the interval between adjacent effective measurement values ​​is greater than 30s, the self-diagnosis status of the acquisition device is abnormal, or any necessary process status is missing, the optimization process of that process calculation period is stopped.

[0015] The alumina concentration is estimated using the concentration output from the tank control machine and corrected by the electrolyte sampling and testing results; the anode-cathode distance is determined based on the anode position, the aluminum liquid height, and the electrolyte height; the electrolyte superheat is determined based on the difference between the electrolyte temperature and the primary crystallization temperature corresponding to the current electrolyte composition.

[0016] As an example, in this embodiment, the current is 400kA, the electrolytic cell operating voltage is 4.18V, the electrolyte temperature is 968.5℃, the alumina concentration is 1.82%, the anode-cathode distance is 4.55cm, the electrolyte superheat is 12.5℃, the flue gas negative pressure is -120Pa, the flue gas flow rate is 8500Nm³ / h; the current anode position adjustment is 0mm, the single alumina feeding amount is 1.80kg, the alumina feeding interval is 27s, and the cooling fan operating frequency is 34Hz.

[0017] Based on the mechanical range of the actuator, the equipment control resolution, and the production procedures of the target electrolytic cell, the control parameter range for the anode position adjustment is -1.20mm to 1.00mm, with a minimum control resolution of 0.10mm; the control parameter range for the single alumina feed amount is 1.60kg to 2.00kg, with a minimum control resolution of 0.05kg; the control parameter range for the alumina feeding interval is 22s to 34s, with a minimum control resolution of 1s; and the control parameter range for the cooling fan operating frequency is 28Hz to 46Hz, with a minimum control resolution of 0.10Hz. The upward direction of the anode corresponds to a positive value for the anode position adjustment, and the downward direction of the anode corresponds to a negative value for the anode position adjustment.

[0018] In a preferred embodiment, this embodiment selects operating condition data from the target electrolytic cell over the most recent 30 production days, which includes no electrode switching operation, no aluminum tapping operation, no anode effect, normal self-diagnostic status of the data acquisition device, and complete process status. The data from the first 20 production days form an interval conversion relationship, and the data from the last 10 production days are used to verify the interval conversion results. The operating condition data are arranged in ascending order of control parameters, and the least squares processing is used to obtain the linear response coefficient for adjacent control parameter segments. When the process status changes in adjacent segments are in opposite directions, the common control parameter value of the two segments is determined as the inflection point, and monotonic interval conversion relationships are established on both sides of the inflection point.

[0019] The absolute deviation between the predicted and measured results is calculated based on the test data. The absolute deviation that 95% of the test data does not exceed is set as the range expansion. In this embodiment, the alumina concentration range expansion is 0.005 percentage points, the electrolyte temperature range expansion is 0.4℃, the electrolytic cell operating voltage range expansion is 0.010V, the auxiliary equipment power range expansion is 0.30kW, and the process direct emission range expansion is... The lower limit obtained from subsequent interval conversion is subtracted from the corresponding interval expansion, and the upper limit is added to the corresponding interval expansion.

[0020] According to an embodiment of the present invention, in combination Figure 1 The flowchart shown illustrates a collaborative optimization method for energy consumption and carbon emissions based on process constraint boundary reconstruction, which specifically includes the following steps: S1. Construct a process constraint topology based on material, energy, and control relationships. Convert the control parameter range into various process state ranges. Determine the constraints that reach or exceed the endpoints of the process's allowable ranges as active constraints, and trace the control parameters in reverse to form an activation propagation chain. Note that the following points should be noted in this step: S1.1 Set the control parameters as control parameter nodes, and set temperature, pressure, flow rate, component content and equipment load as process state nodes. Establish the connection relationship between nodes according to the material flow direction, energy flow direction and control action direction, and associate the process allowable range with the corresponding process state nodes to form a process constraint topology; configure the interval conversion relationship for the connection relationship between nodes.

[0021] Specifically, the process constraint topology includes control parameter nodes, process state nodes, process constraint nodes, and directed association edges. Among them, control parameter nodes correspond to the control variables of production equipment; process state nodes correspond to material state, energy state, or equipment operating state; process constraint nodes correspond to the process allowable range of the process state; and directed association edges include material transfer edges connecting process state nodes according to the material flow direction, energy transfer edges connecting process state nodes according to the energy flow direction, control transfer edges connecting control parameter nodes and process state nodes according to the control action direction, and boundary association edges connecting process state nodes and process constraint nodes.

[0022] Furthermore, the anode position adjustment amount, alumina single feeding amount, alumina feeding interval, and cooling fan operating frequency are respectively set as control parameter nodes; wherein, the control parameter nodes correspond to the displacement control amount of the anode lifting mechanism, the single metering control amount of the alumina feeding device, the action interval control amount of the alumina feeding device, and the frequency conversion control amount of the tank cooling fan.

[0023] The following parameters are set as process state nodes: alumina addition amount, alumina consumption amount, alumina concentration, anode-cathode distance, electrolytic cell operating voltage, electrolytic cell load, electrolyte temperature, electrolyte superheat, heat dissipation of the cell, flue gas flow rate, and flue negative pressure.

[0024] For example, the amount of alumina added, the amount of alumina consumed, and the alumina concentration belong to the material state; the electrolyte temperature, the electrolyte superheat, and the heat dissipation of the tank belong to the energy state; and the anode-cathode distance, the electrolytic cell operating voltage, the electrolytic cell load, the flue gas flow rate, and the flue negative pressure belong to the equipment operating state.

[0025] The material flow direction is as follows: alumina silo, alumina feeding device, electrolyte system, and alumina consumption process. According to the material flow direction, the alumina single feeding amount node and the alumina feeding interval node are connected to the alumina addition amount node, and the alumina addition amount node and the alumina consumption amount node are connected to the alumina concentration node.

[0026] The energy flow direction is as follows: electrical energy input, electrolytic cell Joule heat, electrochemical reaction heat consumption, electrolyte heat storage, and tank heat dissipation. According to the energy flow direction, the anode-cathode distance node is connected to the electrolytic cell working voltage node, the electrolytic cell working voltage node and the series current status are connected to the electrolytic cell load node, and the electrolytic cell load node, electrochemical reaction heat consumption node and tank heat dissipation node are connected to the electrolyte temperature node.

[0027] The control action direction is sequentially connected to the control parameter node, the actuator response node, and the process status node; the anode position adjustment node is connected to the anode-cathode distance node; the alumina single feed amount node and the alumina feed interval node are connected to the alumina feed amount node; the cooling fan operating frequency node is connected to the tank heat dissipation node, the flue gas flow rate node, and the flue negative pressure node.

[0028] The alumina concentration node is connected to the electrolytic cell operating voltage node and the electrolyte temperature node to reflect the effect of alumina concentration changes on the electrolytic cell resistance and alumina dissolution endothermic reaction; the electrolyte temperature node is connected to the electrolytic cell operating voltage node to reflect the effect of electrolyte temperature changes on the electrolyte resistance.

[0029] In this embodiment, the following process allowable ranges are set: alumina concentration 1.80% to 3.20%, electrolyte temperature 950°C to 970°C, anode-cathode distance 4.10cm to 5.00cm, electrolytic cell operating voltage 3.95V to 4.35V, electrolyte superheat 8°C to 18°C, flue gas negative pressure -180Pa to -80Pa, and flue gas flow rate 7000. Up to 11000 And the allowable range of electrolytic cell load technology is 1.55MW to 1.74MW.

[0030] Each process allowable interval forms a process constraint node, and the corresponding process status node is connected to the process constraint node through boundary association edges; the process allowable interval is taken from the production technology specifications adopted at the current cell age stage of the target electrolyzer; the same set of process allowable intervals is used within the process calculation period.

[0031] Interval conversion relationships are configured between control parameter nodes and process status nodes, and between upstream process status nodes and downstream process status nodes. A mechanical displacement relationship is adopted between the anode position adjustment amount and the anode-cathode distance. For every 1 mm increase in the anode position adjustment amount, the anode-cathode distance increases by 1 mm. Within the operating range of this embodiment, for every 1 mm increase in the anode-cathode distance, the electrolytic cell operating voltage increases by 0.035V.

[0032] The amount of alumina added is determined by multiplying the amount of alumina added in a single batch by the equivalent number of feedings within a 600s process calculation period; the equivalent number of feedings is determined by the ratio of 600s to the alumina feeding interval; the heat dissipation of the tank increases monotonically with the operating frequency of the cooling fan in the range of 28Hz to 46Hz; there is a turning point between the alumina concentration and the electrolytic cell operating voltage at an alumina concentration of 2.20%: when the alumina concentration increases from 1.70% to 2.20%, the electrolytic cell operating voltage decreases monotonically; when the alumina concentration increases from 2.20% to 3.20%, the electrolytic cell operating voltage increases monotonically.

[0033] S1.2. Substitute the control parameter intervals into the interval conversion relation according to the transmission order of the connection relationship between nodes; for monotonic interval conversion relations, convert the endpoints of the input intervals to the endpoints of the output intervals; for interval conversion relations containing turning points, divide the monotonic segments according to the turning points, convert them separately, and then find the union of the resulting output intervals.

[0034] In a preferred embodiment, interval conversion is performed in the following order: from control parameter nodes to process state nodes, from upstream process state nodes to downstream process state nodes, and from process state nodes to process constraint nodes. The anode position adjustment interval is first converted into the anode-cathode distance interval, and then into the electrolytic cell operating voltage interval and the electrolytic cell load interval. The alumina single feed amount interval and the alumina feed interval interval are first converted into the alumina addition amount interval, and then combined with the alumina consumption interval to convert into the alumina concentration interval. The cooling fan operating frequency interval is first converted into the tank heat dissipation interval, the flue gas flow interval, and the flue negative pressure interval. The electrolytic cell load interval, the electrochemical reaction heat consumption interval, and the tank heat dissipation interval are all converted into the electrolyte temperature interval.

[0035] For monotonically increasing interval conversion relationships, the lower limit of the input interval corresponds to the lower limit of the output interval, and the upper limit of the input interval corresponds to the upper limit of the output interval; for monotonically decreasing interval conversion relationships, the upper limit of the input interval corresponds to the lower limit of the output interval, and the lower limit of the input interval corresponds to the upper limit of the output interval.

[0036] For interval conversion relationships containing inflection points, first determine whether the input interval crosses the inflection point; if the input interval does not cross the inflection point, convert according to the increasing or decreasing direction of the monotonic segment; if the input interval crosses the inflection point, split the input interval with the inflection point, convert each monotonic segment separately, and retain each output interval; when the output intervals overlap or their endpoints connect, take the union; when there are uncovered ranges between the output intervals, retain them separately.

[0037] S1.3 When the same process state node receives multiple input intervals, substitute the multiple input intervals into the corresponding material conservation equation, energy conservation equation or control superposition equation to obtain the process state interval of the process state node.

[0038] In this embodiment, the material conservation equation, energy conservation equation, or control superposition formula corresponds to different types of process state nodes. When the target process state node belongs to the material state node, the material quantity range entering the material quantity range, the material quantity range leaving the material quantity range, and the initial material inventory range are substituted into the material conservation equation. When the target process state node belongs to the energy state node, the energy input range, the energy consumption range, the energy loss range, and the initial energy state range are substituted into the energy conservation equation. When the target process state node belongs to the equipment operation state node, the multiple control parameter change ranges and the previous process state change ranges are substituted into the control superposition formula. Thus, the multiple input ranges formed by the previous nodes are not directly merged, but are used as joint inputs of the corresponding physical calculation relationships. The process state range of the target process state node is determined based on the minimum and maximum values ​​of the calculation results corresponding to all allowed input combinations.

[0039] For example, suppose the calculation relationship between the target process state and multiple input quantities is as follows: ; in, For the target process state, The material conservation equation, energy conservation equation, or control superposition equation corresponding to the target process state node. For multiple input quantities involved in the calculation of the target process state; when When each input interval has its own corresponding input interval, the input intervals together constitute a joint value range, and the target process state interval is determined according to the following formula: ; in, The target process state range; for The input intervals together form the joint value range; the lower limit of the interval is the minimum value among all allowed input combinations within the joint value range; the upper limit of the interval is the maximum value among all allowed input combinations within the joint value range.

[0040] When the calculation relationship has a definite monotonic direction within each input interval, the combination of input interval endpoints that makes the calculation result reach the minimum and maximum values ​​are selected according to the increasing or decreasing effect of each input quantity on the target process state. When the calculation relationship contains inflection points within the input interval, monotonic segments are divided according to the inflection points, and the output intervals corresponding to each monotonic segment are calculated separately. When the output intervals overlap or their endpoints connect, the union is calculated. When there are uncovered ranges between the output intervals, they are retained as multiple continuous intervals.

[0041] For the alumina concentration node, the alumina concentration node receives the electrolyte mass range, initial alumina concentration range, alumina feeding effectiveness coefficient range, alumina addition range, and alumina consumption range at the start of the process calculation period.

[0042] For example, when the total mass of electrolytes during the process calculation period is considered a constant, the material conservation equation is as follows: ; in, This represents the alumina concentration at the end of the process calculation period. The electrolyte mass at the start of the process calculation period; The alumina concentration at the start of the process calculation period; The effective coefficient for alumina feeding; This refers to the amount of alumina added by the alumina feeding device during the process calculation period; and The product of these two values ​​represents the effective amount of alumina added to the electrolyte system. This represents the amount of alumina consumed less from the electrolyte system due to the electrolytic reaction during the process calculation period. Will , , , and Substituting the corresponding input intervals into the material conservation equation, and under the condition that the denominator is greater than zero, determine... The minimum and maximum values ​​are used; the minimum value is taken as the lower limit of the alumina concentration range, and the maximum value is taken as the upper limit of the alumina concentration range, thus obtaining the process state range of the alumina concentration node.

[0043] For the electrolyte temperature node, the electrolyte temperature node receives the electrolyte temperature range at the beginning of the process calculation period, the electrical energy heat input range, the electrochemical reaction heat consumption range, the tank and flue gas heat exhaust range, the electrolyte mass range participating in heat balance, and the electrolyte specific heat capacity range.

[0044] For example, since the above input quantities collectively determine the net heat change of the electrolyte system, the energy conservation equation is adopted: ; in, The electrolyte temperature at the end of the process calculation period; The electrolyte temperature at the start of the process calculation period; The electrical and thermal input transferred to the electrolyte system during the process calculation period; This represents the heat consumed by the electrochemical reaction during the process calculation period; This represents the heat discharged through the furnace shell, furnace sides, and flue gas during the process calculation period. The mass of the electrolyte participating in the heat balance; The specific heat capacity of the electrolyte; minus and The result obtained is the net heat change of the electrolyte system during the process calculation period; the net heat change divided by and The product of these terms yields the change in electrolyte temperature; , , , , and Substituting the corresponding input intervals into the energy conservation equation, in and Under the condition that all are greater than zero, determine The minimum and maximum values ​​are used; the minimum value is taken as the lower limit of the electrolyte temperature range, and the maximum value is taken as the upper limit of the electrolyte temperature range, thus obtaining the process state range of the electrolyte temperature node.

[0045] For the electrolytic cell operating voltage node, the operating voltage node receives the current operating voltage range, the anode-cathode distance variation range, the alumina concentration variation range, and the electrolyte temperature variation range. The anode-cathode distance variation affects the electrolyte conductivity, while the alumina concentration variation and electrolyte temperature variation affect the electrolyte resistance. Within the current monotonic operating condition range, these effects are superimposed according to the local response coefficient; therefore, a control superposition method is adopted. ; in, The adjusted operating voltage of the electrolytic cell; This is the current operating voltage of the electrolytic cell; This is the voltage response coefficient corresponding to the change in anode-cathode distance; This represents the change in the anode-cathode distance. This represents the voltage response coefficient corresponding to the change in alumina concentration. This represents the change in alumina concentration; This is the voltage response coefficient corresponding to the change in electrolyte temperature; This represents the change in electrolyte temperature. , and The determination is based on the stable operating data of the target electrolyzer within the corresponding monotonic operating condition range, and remains unchanged during a single process calculation period; , , and Substituting the corresponding input intervals into the control superposition formula, according to , and The positive and negative directions, choose to make The minimum and maximum values ​​of the input interval endpoints are obtained respectively; the minimum value is used as the lower limit of the electrolytic cell working voltage range, and the maximum value is used as the upper limit of the electrolytic cell working voltage range, thereby obtaining the process state range of the electrolytic cell working voltage node.

[0046] When the alumina concentration variation range crosses the inflection point of the voltage response relationship, the same voltage response should not be used throughout the entire alumina concentration variation range. The alumina concentration variation range should be divided according to the inflection point, and the working voltage output range of the electrolytic cell should be calculated on both sides of the inflection point using the corresponding voltage response coefficient. Then, the union of the output ranges or their endpoints should be determined according to whether they overlap or are connected, or they should be retained separately.

[0047] Therefore, the material conservation equation, energy conservation equation, and control superposition equation respectively convert the input intervals formed by multiple front-end nodes into alumina concentration intervals, electrolyte temperature intervals, and electrolytic cell operating voltage intervals; the three output intervals respectively constitute the process state intervals of the corresponding process state nodes and are transmitted to the corresponding process constraint nodes for endpoint comparison between step S1.4 and the alumina concentration process allowable interval, electrolyte temperature process allowable interval, and electrolytic cell operating voltage process allowable interval.

[0048] S1.4 Compare the upper limit of the process state interval with the upper limit of the process allowable interval, and compare the lower limit of the process state interval with the lower limit of the process allowable interval; when the upper limit of the process state interval is not less than the upper limit of the process allowable interval, or the lower limit of the process state interval is not greater than the lower limit of the process allowable interval, determine the corresponding process allowable interval constraint as the active constraint.

[0049] For example, comparing the alumina concentration range of 1.7395% to 1.9065% with the process-allowed alumina concentration range of 1.80% to 3.20%, the lower limit of the alumina concentration range of 1.7395% is not greater than the lower limit of the process-allowed range of 1.80%, so the lower limit constraint of alumina concentration is determined as the first activation constraint; comparing the electrolyte temperature range of 965.46℃ to 971.12℃ with the process-allowed electrolyte temperature range of 950℃ to 970℃, the upper limit of the electrolyte temperature range of 971.12℃ is not less than the upper limit of the process-allowed range of 970℃, so the upper limit constraint of electrolyte temperature is determined as the second activation constraint.

[0050] S1.5. Tracing back along the process constraint topology from the activation constraint, the preceding process state nodes and control parameter nodes involved in the calculation of the corresponding endpoints of the process state interval are traced until the control parameter node is reached; tracing paths with common nodes and connection relationships are merged to form an activation propagation chain from the control parameter node to the activation constraint.

[0051] Specifically, starting from the lower limit of the alumina concentration range corresponding to the first activation constraint, the endpoint combination participating in the calculation of the lower limit of the alumina concentration range is determined; the endpoint combination includes the lower limit of the alumina addition range, the upper limit of the alumina consumption range, and the current alumina concentration; continuing to trace back to the lower limit of the alumina addition range, the lower limit of the single alumina addition range and the upper limit of the alumina addition interval range are determined; the single alumina addition node, the alumina addition interval node, the alumina addition node, the alumina concentration node, and the first activation constraint constitute the first activation propagation chain.

[0052] Starting from the upper limit of the electrolyte temperature range corresponding to the second activation constraint, determine the upper limit of the electrical energy heat input range, the lower limit of the electrochemical reaction heat consumption range, and the lower limit of the tank heat dissipation range that participate in the calculation of the upper limit of the electrolyte temperature range; continue to trace back the upper limit of the electrical energy heat input range to the electrolytic cell working voltage node, the anode-cathode distance node, and the anode position adjustment node; trace back the lower limit of the tank heat dissipation range to the cooling fan operating frequency node; trace back the alumina dissolution endothermic heat to the alumina concentration node and the alumina feeding interval node; the anode position adjustment node, the alumina feeding interval node, the cooling fan operating frequency node, and the corresponding process state node constitute the second activation propagation chain.

[0053] Both the first and second activation propagation chains contain alumina feeding interval nodes, therefore the alumina feeding interval is the same control parameter associated with multiple activation propagation chains.

[0054] Preferably, step S1 determines the control path that may exceed the lower limit of alumina concentration and the upper limit of electrolyte temperature under the current operating conditions through process constraint topology and interval operation, which solves the problem that the control range of fixed equipment cannot reflect the material balance, heat balance and control coupling relationship; the resulting activation propagation chain limits the processing range of subsequent boundary reconstruction and reduces the participation of nodes unrelated to the critical process state in the reverse conversion.

[0055] S2. Transform the process-allowed interval of the activation constraint into a parameter constraint interval by moving backward along the activation propagation chain, and intersect it with the control parameter interval to form a local process boundary. Note that the following points should be noted in this step: S2.1 Arrange the interval conversion relationships in the activation propagation chain in reverse order from the activation constraint to the control parameter node.

[0056] It should be noted that the first activation propagation chain is arranged in the order of first activation constraint, alumina concentration node, alumina addition amount node, alumina single addition amount node, and alumina addition interval node; the second activation propagation chain is arranged in reverse hierarchy according to the second activation constraint, electrolyte temperature node, electrical energy heat input node and tank heat dissipation node, electrolytic cell operating voltage node, anode-cathode distance node, anode position adjustment amount node, alumina concentration node, alumina addition interval node, and cooling fan operating frequency node; when a downstream node connects to multiple upstream nodes, they form reverse branches respectively; the reverse branches only include nodes that participate in the calculation of the corresponding endpoints of the activation constraint.

[0057] S2.2 Set the process allowable interval of the activation constraint as the current allowable interval; for the monotonic interval conversion relationship, find the intersection of the current allowable interval and the output range of the monotonic interval conversion relationship, and convert the endpoints of the non-empty intersection into the candidate allowable intervals of the previous node.

[0058] In a preferred embodiment, the process allowable range of alumina concentration of 1.80% to 3.20% for the first activation constraint is set as the current allowable range; the intersection of the current allowable range and the output range from alumina addition amount to alumina concentration is calculated, and the endpoints of the non-empty intersection are reverse-calculated to obtain a candidate allowable range of alumina addition amount of 39.56 kg to 54.55 kg within 600 s; when the single alumina addition amount is the current value of 1.80 kg, the candidate allowable range of alumina addition interval is reverse-calculated according to the candidate allowable range of alumina addition amount to obtain a candidate allowable range of alumina addition interval of 22 s to 27 s; when the alumina addition interval is the current value of 27 s, the candidate allowable range of single alumina addition amount is 1.80 kg to 2.00 kg.

[0059] In this embodiment, the allowed range of candidate preceding nodes refers to the range that the preceding node to be converted can obtain when the following node is in the current allowed range and the other preceding nodes are in the value or value range used in this reverse conversion.

[0060] S2.3 For the segmented monotonic interval conversion relationship, find the intersection of the current allowed interval with the output range of each monotonic segment, perform reverse conversion on each non-empty intersection, and find the union of the resulting candidate allowed intervals of the previous node.

[0061] For example, the second activation propagation chain includes a piecewise monotonic relationship between alumina concentration and electrolytic cell operating voltage; the electrolytic cell operating voltage output ranges corresponding to the alumina concentration ranges of 1.70% to 2.20% and 2.20% to 3.20% are obtained respectively, and the current allowable range of electrolytic cell operating voltage is intersected with the two output ranges respectively.

[0062] When the intersection of a monotonic segment is empty, delete the reverse branch of that monotonic segment; when the intersection is not empty, reverse the intersection endpoints according to the increasing or decreasing direction of the monotonic segment; when two reverse conversion results overlap or their endpoints connect, find the union; when there is an uncovered range between two reverse conversion results, retain them respectively.

[0063] As an example, in this embodiment, the allowable range for the front-end node concentration is 1.80% to 2.68%.

[0064] S2.4. Determine the candidate allowable interval of the predecessor node as the allowable interval of the predecessor node; when the same predecessor node is formed by multiple successor nodes to form the candidate allowable interval of the predecessor node, find the intersection of the candidate allowable intervals of the predecessor node; set the allowable interval of the predecessor node as the new current allowable interval and perform reverse calculation step by step until the parameter constraint interval of the control parameter node is obtained.

[0065] Specifically, when a candidate allowed interval for the same pre-level node is formed by only one subsequent node, the candidate allowed interval is determined as the allowed interval of the pre-level node; when a candidate allowed interval for the same pre-level node is formed by multiple subsequent nodes, the intersection of all candidate allowed intervals is calculated, and the non-empty intersection is determined as the allowed interval of the pre-level node.

[0066] Set the allowed interval of the preceding node as the current allowed interval of the next reverse level, and continue the reverse calculation according to the monotonic relationship or the piecewise monotonic relationship until the control parameter node is reached.

[0067] As an example, the first activation propagation chain forms a single alumina feeding quantity parameter constraint range of 1.80 kg to 2.00 kg and an alumina feeding interval parameter constraint range of 22 s to 27 s.

[0068] As an example, the second activation propagation chain forms an anode position adjustment parameter constraint range of -1.20mm to 0.40mm, an alumina feeding interval parameter constraint range of 24s to 30s, and a cooling fan operating frequency parameter constraint range of 34Hz to 46Hz.

[0069] It should be noted that the parameter constraint interval represents the range that the control parameter can be obtained under the condition that the corresponding activation constraint does not exceed the endpoint of the process allowable interval; the parameter constraint intervals formed by different activation propagation chains for the same control parameter are retained respectively.

[0070] S2.5. Find the intersection of each continuous interval in the parameter constraint interval with the corresponding control parameter interval, and form the local process boundary of the control parameter by all non-empty intersections; when all intersections are empty, the local process boundary is determined as an empty set.

[0071] For example, the first activation propagation chain forms a local process boundary of 1.80 kg to 2.00 kg for the single alumina feeding amount and a local process boundary of 22 s to 27 s for the alumina feeding interval; the second activation propagation chain forms a local process boundary of -1.20 mm to 0.40 mm for the anode position adjustment amount, a local process boundary of 24 s to 30 s for the alumina feeding interval, and a local process boundary of 34 Hz ​​to 46 Hz for the cooling fan operating frequency.

[0072] In this embodiment, a continuous interval means that each value between the lower limit and the upper limit of the interval, arranged according to the minimum control resolution of the corresponding control parameters, belongs to the intersection of the parameter constraint interval and the control parameter interval. There are no unsolvable ranges, equipment disabled ranges, or process conflict ranges within the interval.

[0073] In a preferred embodiment, the anode position adjustment increments by 0.10 mm, the single alumina feeding amount increments by 0.05 kg, the alumina feeding interval increments by 1 s, and the cooling fan operating frequency increments by 0.10 Hz; when the parameter constraint interval contains multiple non-connected continuous intervals, they are retained respectively; when all intersections are empty, the corresponding local process boundary is determined as an empty set, and the candidate parameter processing of the corresponding activation propagation chain is stopped.

[0074] It should be noted that step S2 reverses the process allowable range of alumina concentration and electrolyte temperature to the anode position, alumina feeding and tank heat dissipation control end, which solves the problem that the equipment control parameter range includes values ​​that cannot be executed under the current operating conditions; in particular, by eliminating the range of control parameters that would cause activation constraints to exceed the limit through local process boundaries, the processing space for subsequent energy consumption and carbon emission calculations is reduced.

[0075] S3. Within the local process boundary, adjust the control parameters bidirectionally, calculate the energy consumption and carbon emission reduction ranges respectively, and take the minimum of the two lower limits as the coordinated reduction value, forming a coordinated reduction parameter domain where both lower limits are greater than zero. Note that the following should be noted in this step:

[0076] S3.1 When the local process boundary is not empty, the local process boundary is divided into an increasing adjustment range and a decreasing adjustment range, with the current control parameter value as the dividing line.

[0077] Specifically, the delineation of local process boundaries includes: Select a continuous parameter interval containing the current control parameter value from the local process boundary; when the current control parameter value is inside the continuous parameter interval, the interval from the lower limit of the continuous parameter interval to the current control parameter value and excluding the current control parameter value is determined as the decrease adjustment interval, and the interval from the current control parameter value to the upper limit of the continuous parameter interval and excluding the current control parameter value is determined as the increase adjustment interval; When the current control parameter value is equal to the lower limit of the continuous parameter range, only an increasing adjustment range is formed; when the current control parameter value is equal to the upper limit of the continuous parameter range, only a decreasing adjustment range is formed.

[0078] For example, in the first activation propagation chain, the current value of the single alumina feeding amount of 1.80 kg is located at the lower limit of the local process boundary, thus forming an increased adjustment range of 1.85 kg to 2.00 kg; the current value of the alumina feeding interval of 27 s is located at the upper limit of the local process boundary, thus forming a decreased adjustment range of 22 s to 26 s; in the second activation propagation chain, the anode position adjustment amount forms a decreased adjustment range of -1.20 mm to -0.10 mm and an increased adjustment range of 0.10 mm to 0.40 mm; the alumina feeding interval forms a decreased adjustment range of 24 s to 26 s and an increased adjustment range of 28 s to 30 s; the cooling fan operating frequency forms an increased adjustment range of 34.1 Hz to 46 Hz.

[0079] S3.2. Convert the endpoints of the non-empty increasing and decreasing adjustment intervals along the positive direction of the activation propagation chain to obtain the energy consumption rate intervals, the corresponding process duration intervals, and the process direct emission intervals. When the interval conversion relationship contains inflection points, divide the monotonic segments according to the inflection points, convert them separately, and then find the union of the resulting intervals.

[0080] Specifically, the endpoints of the adjustment interval and the inflection points of the interval conversion relationship within the adjustment interval are converted forward along the corresponding activation propagation chain; the forward conversion sequentially obtains the alumina addition range, alumina concentration range, anode-cathode distance range, electrolytic cell working voltage range, electrolytic cell load range, electrolyte temperature range, cell heat dissipation range, and auxiliary equipment power range.

[0081] The energy consumption rate includes the electrolytic cell power, the average power of the alumina feeding device, and the power of the cooling fan; the corresponding process duration is 600s; the direct emissions from the process include carbon dioxide emissions from carbon anode consumption and perfluorinated carbon emissions converted from the anode effect.

[0082] The carbon anode consumption is determined based on the series current, current efficiency, and net anode consumption coefficient in the most recent anode replacement cycle of the target electrolytic cell. When the lower limit of the alumina concentration range corresponding to the candidate parameter is lower than 1.80% or the electrolytic cell operating voltage range reaches the anode effect judgment voltage, the corresponding anode effect emission range is included in the process direct emission range. Otherwise, the lower limit of the anode effect emission range is 0, and the upper limit is the range extension corresponding to the test data.

[0083] For example, through forward conversion, the control parameters for reducing both energy consumption and carbon emissions in the first activation propagation chain are: a single alumina feeding amount of 1.85 kg to 2.00 kg and an alumina feeding interval of 24 s to 26 s; the control parameters for reducing both energy consumption and carbon emissions in the second activation propagation chain are: anode position adjustment amount of -1.00 mm to -0.20 mm, an alumina feeding interval of 24 s to 27 s, and a cooling fan operating frequency of 36 Hz to 42 Hz.

[0084] S3.3 Perform interval product on each energy consumption rate interval and the corresponding process duration interval to obtain each energy consumption interval, and sum the energy consumption intervals to obtain the adjusted energy consumption interval; perform interval product on each energy consumption interval and the corresponding energy carbon emission factor interval and sum the results, and superimpose the process direct emission interval to obtain the adjusted carbon emission interval.

[0085] The electrolytic cell power range, the alumina feeding device average power range, and the cooling fan power range are each multiplied by the 600s process duration execution range, and then converted to kWh. Finally, the energy consumption ranges of each energy source are summed to obtain the adjusted energy consumption range.

[0086] The range of carbon emission factors for electricity is obtained by the energy management system of the electrolysis workshop based on the power supply configuration of the power supply node where the target electrolyzer is located during the corresponding process calculation period; this embodiment adopts... to The energy consumption range and the electricity carbon emission factor range are multiplied by the range, and then the direct emission range of the process is superimposed to obtain the adjusted carbon emission range.

[0087] The energy consumption baseline value for the current operating condition within the 600s process calculation period is 280.80 kWh; the carbon emission baseline value for the current operating condition is calculated based on the upper limit of the electricity carbon emission factor range and the direct emissions of the current process. .

[0088] The first coordinated reduction parameter range corresponds to adjusting the energy consumption range from 276.20 kWh to 279.00 kWh, and adjusting the carbon emission range. to The second coordinated reduction parameter range corresponds to adjusting the energy consumption range from 275.80 kWh to 278.60 kWh, and adjusting the carbon emission range. to .

[0089] S3.4. Based on the energy consumption and carbon emission values ​​that are greater than zero under the current operating conditions, the ratio of the difference between the energy consumption benchmark value and the upper limit of the energy consumption range to the energy consumption benchmark value is determined as the lower limit of the energy consumption reduction range, and the ratio of the difference between the energy consumption benchmark value and the lower limit of the energy consumption range to the energy consumption benchmark value is determined as the upper limit of the energy consumption reduction range; the carbon emission reduction range is determined according to the same endpoint conversion relationship.

[0090] Specifically, the lower limit of the carbon emission reduction range is determined by the ratio of the difference between the carbon emission benchmark value and the adjusted upper limit of the carbon emission range to the carbon emission benchmark value; the upper limit of the carbon emission reduction range is determined by the ratio of the difference between the carbon emission benchmark value and the adjusted lower limit of the carbon emission range to the carbon emission benchmark value.

[0091] S3.5. The minimum value between the lower limit of the energy consumption reduction range and the lower limit of the carbon emission reduction range shall be determined as the coordinated reduction value; the adjustment range in which both lower limits are greater than zero shall be determined as the coordinated reduction parameter range, and the coordinated reduction parameter range shall constitute the coordinated reduction parameter domain.

[0092] Specifically, the lower limits of the energy consumption reduction range and carbon emission reduction range corresponding to each adjustment range are checked respectively; when both lower limits are greater than zero, the adjustment range is determined as the coordinated reduction parameter range; when either lower limit is not greater than zero, the adjustment range is deleted; the coordinated reduction parameter domain is formed by the combination relationship of all coordinated reduction parameter ranges in the same activation propagation chain; among them, a combination of control parameters belongs to the coordinated reduction parameter domain only if it falls into all coordinated reduction parameter ranges of the corresponding activation propagation chain at the same time.

[0093] For example, the first synergistic reduction parameter range includes a single alumina feeding amount of 1.85 kg to 2.00 kg and an alumina feeding interval of 24 s to 26 s; the synergistic reduction value of the first activation propagation chain is the lower of the energy consumption reduction range of 0.64% and the carbon emission reduction range of 2.02%, i.e., 0.64%; the second synergistic reduction parameter range includes an anode position adjustment amount of -1.00 mm to -0.20 mm, an alumina feeding interval of 24 s to 27 s, and a cooling fan operating frequency of 36 Hz to 42 Hz; the synergistic reduction value of the second activation propagation chain is 0.78%.

[0094] It should be noted that step S3 evaluates the direction of increase and decrease of control parameters within the local process boundary, and deletes the adjustment range where energy consumption or carbon emissions may increase, thus solving the problem that a single energy consumption evaluation may increase direct emissions from the process or power consumption of auxiliary equipment.

[0095] S4. Find the intersection of the value intervals of the same control parameter associated with multiple activation propagation chains within each cooperative descent parameter domain, combine the intersections, and select the combination with the largest cooperative descent value as the cooperative optimization parameter combination. Note that the following should be noted in this step: S4.1 Identify the same control parameter that is associated with two or more activation propagation chains, and extract the value range of the control parameter from each associated cooperative descent parameter domain.

[0096] Specifically, each activation propagation chain is checked according to the unique node identifier of the control parameter node; when the same control parameter node appears in more than two activation propagation chains at the same time, and there are continuous directed paths between the control parameter node and the corresponding activation constraint, the control parameter is determined to be the same control parameter associated with multiple activation propagation chains.

[0097] For example, the alumina feeding interval node is involved in the calculation of both the lower limit of the alumina concentration range and the upper limit of the electrolyte temperature range. Therefore, the alumina feeding interval is the same control parameter that is associated with both the first and second activation propagation chains.

[0098] The alumina feeding interval range of 24s to 26s is extracted from the first collaborative descent parameter domain, and the alumina feeding interval range of 24s to 27s is extracted from the second collaborative descent parameter domain.

[0099] S4.2. Find the intersection of the value ranges of the control parameters, and determine the non-empty intersection as the cross-chain value range of the control parameters; when the intersection is empty, no candidate parameter combination involving the relevant cooperative descent parameter domain is formed.

[0100] S4.3 Select the endpoints of the cross-chain value range and the value ranges of other control parameters, as well as the inflection points of the interval conversion relationship within the value range. Combine them according to the order of the control parameter nodes in the process constraint topology, retain the combinations that fall into all relevant collaborative descent parameter domains, and obtain candidate parameter combinations.

[0101] Specifically, candidate values ​​are obtained from the corresponding coordinated descent parameter interval according to the minimum control resolution of each control parameter; to reduce the number of combinations, the lower limit of the interval, the upper limit of the interval, the most recent allowed value of the current control parameter, and the inflection point of the interval conversion relationship are selected first; when the interval between adjacent candidate values ​​is greater than one-quarter of the width of the corresponding control parameter interval, an additional candidate value is added in the middle of the adjacent candidate values.

[0102] For example, in this embodiment, the anode position adjustment amounts are selected as -1.00mm, -0.80mm, -0.60mm, -0.40mm and -0.20mm; the single alumina feeding amount is 1.85kg, 1.90kg, 1.95kg and 2.00kg; the alumina feeding interval is 24s, 25s and 26s; and the cooling fan operating frequency is 36Hz, 39Hz and 42Hz.

[0103] Candidate values ​​are combined in the order of anode position adjustment, single alumina feeding amount, alumina feeding interval, and cooling fan operating frequency. Each combination is checked to see if it falls into both the first and second collaborative descent parameter domains. Combinations that do not fall into both collaborative descent parameter domains are deleted to obtain candidate parameter combinations.

[0104] S4.4. The candidate parameter combination is converted along each activation propagation chain in the forward direction to obtain the energy consumption reduction range and carbon emission reduction range respectively. The minimum value of the lower limit of the two reduction ranges is determined as the intra-chain synergistic reduction value, and the minimum value of the intra-chain synergistic reduction value of each activation propagation chain is determined as the combined synergistic reduction value of the candidate parameter combination.

[0105] Each candidate parameter combination is converted forward along the first and second activation propagation chains respectively, and the energy consumption reduction range and carbon emission reduction range are determined according to the processing methods of S3.3 and S3.4 respectively. For the same activation propagation chain, the lower limit of the energy consumption reduction range and the lower limit of the carbon emission reduction range are compared, and the one with the lower value is determined as the intra-chain collaborative reduction value. When the two values ​​are the same, the intra-chain collaborative reduction value is the same.

[0106] For the same candidate parameter combination, compare its intra-chain cooperative reduction value in all activation propagation chains, and determine the lowest value as the combined cooperative reduction value; this process uses the degree of cooperative reduction of the candidate parameter combination in the most unfavorable activation propagation chain as the combined evaluation result.

[0107] As an example, candidate parameter combination A includes an anode position adjustment of -0.80 mm, a single alumina feeding amount of 1.95 kg, an alumina feeding interval of 25 s, and a cooling fan operating frequency of 39 Hz. In the first activation propagation chain, candidate parameter combination A has a lower limit of 1.05% for energy consumption reduction, a lower limit of 2.60% for carbon emission reduction, and a synergistic reduction value within the first chain of 1.05%. In the second activation propagation chain, the lower limit of 0.92% for energy consumption reduction, a lower limit of 2.35% for carbon emission reduction, and a synergistic reduction value within the second chain of 0.92%. The combined synergistic reduction value of candidate parameter combination A is 0.92%.

[0108] As an example, candidate parameter combination B includes anode position adjustment amount of -1.00mm, alumina single feed amount of 2.00kg, alumina feed interval of 24s and cooling fan operating frequency of 42Hz. Its first chain synergistic reduction value is 1.12%, the second chain synergistic reduction value is 0.70%, and the combined synergistic reduction value is 0.70%.

[0109] As an example, candidate parameter combination C includes anode position adjustment amount of -0.40mm, alumina single feed amount of 1.90kg, alumina feeding interval of 26s and cooling fan operating frequency of 36Hz. Its first chain synergistic reduction value is 0.68%, the second chain synergistic reduction value is 0.81%, and the combined synergistic reduction value is 0.68%.

[0110] S4.5 Compare the combined synergistic reduction values ​​of each candidate parameter combination, and determine the candidate parameter combination with the largest combined synergistic reduction value as the synergistic optimization parameter combination.

[0111] Specifically, determining the combination of parameters for collaborative optimization includes: Candidate parameter combinations are arranged from largest to smallest according to their combined synergistic reduction value, and the candidate parameter combination ranked first is selected. When the combined synergistic reduction values ​​of multiple candidate parameter combinations are the same, the lower limits of the energy consumption reduction range and the lower limits of the carbon emission reduction range of the candidate parameter combinations in each activation propagation chain are summed, and the candidate parameter combination with the largest sum is selected. When the sums are the same, calculate the absolute value of the difference between the candidate value and the current value of each control parameter, divide the absolute value by the interval width of the continuous local process boundary where the candidate value is located, and obtain the normalized adjustment amount; sum the normalized adjustment amounts in the same candidate parameter combination, and determine the candidate parameter combination with the smallest sum of normalized adjustment amounts as the collaborative optimization parameter combination.

[0112] As an example, the final combination of collaborative optimization parameters determined in this embodiment is: anode position adjustment amount -0.80mm, alumina single feeding amount 1.95kg, alumina feeding interval 25s, and cooling fan operating frequency 39Hz.

[0113] Substituting the combined parameters of the collaborative optimization into the forward topology transformation of the process constraints, the alumina addition during the 600s process calculation period is 46.80 kg; the alumina concentration range is 1.8388% to 1.8801%; the electrolyte temperature range is 966.8℃ to 969.4℃; the anode-cathode distance range is 4.46 cm to 4.48 cm; the electrolytic cell operating voltage range is 4.10V to 4.14V; the flue gas negative pressure range is -151Pa to -108Pa; and the flue gas flow rate range is 8100. Up to 9700 .

[0114] Furthermore, the corresponding adjustment of the collaborative optimization parameter combination corresponds to an energy consumption range of 275.80 kWh to 278.60 kWh, with an energy consumption reduction range of 0.78% to 1.78%; the corresponding adjustment of the carbon emission range... to Carbon emissions are expected to decrease by 2.12% to 5.66%.

[0115] Preferably, when the local process boundary of any active propagation chain is empty, the collaborative descent parameter domain is empty, or the cross-chain value range is empty, the collaborative optimization parameter combination is not output during the current process calculation period, and the current control parameter value is retained; after the complete process state is obtained in the next process calculation period, steps S1 to S4 are re-executed.

[0116] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for collaborative optimization of energy consumption and carbon emissions based on process constraint boundary reconstruction, characterized in that, include: Based on the material, energy and control relationships, a process constraint topology is constructed, the control parameter range is converted into each process state range, the constraints that reach or exceed the endpoints of the process allowable range are identified as activation constraints, and the control parameters are traced in reverse to form an activation propagation chain; The process constraint topology construction and conversion of control parameter intervals into process state intervals includes: setting control parameters as control parameter nodes, setting temperature, pressure, flow rate, component content, and equipment load as process state nodes; establishing inter-node connection relationships according to material flow direction, energy flow direction, and control action direction; and associating process allowable intervals with corresponding process state nodes to form a process constraint topology; configuring interval conversion relationships for the inter-node connection relationships; substituting control parameter intervals into the interval conversion relationships according to the transmission order of the inter-node connection relationships; for monotonic interval conversion relationships, converting the input interval endpoints into the output interval endpoints; for interval conversion relationships containing inflection points, dividing monotonic segments according to the inflection points, converting them separately, and then finding the union of the resulting output intervals; when the same process state node receives multiple input intervals, substituting the multiple input intervals into the corresponding material conservation equation, energy conservation equation, or control superposition equation to obtain the process state interval of the process state node; The process allowable interval of the activation constraint is converted into the parameter constraint interval in reverse along the activation propagation chain, and the intersection with the control parameter interval is used to form the local process boundary; Within the local process boundary, the control parameters are adjusted bidirectionally, and the energy consumption and carbon emission reduction ranges are calculated separately. The minimum of the two lower limits is taken as the synergistic reduction value, forming a synergistic reduction parameter domain where the lower limits are both greater than zero. The intersection of the value ranges of the same control parameter associated with multiple activation propagation chains in each collaborative descent parameter domain is obtained, and the intersection is combined. The combination with the largest collaborative descent value is taken as the collaborative optimization parameter combination.

2. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 1, characterized in that, Determine activation constraints and form activation propagation chains, including: The upper limit of the process state interval is compared with the upper limit of the process allowable interval, and the lower limit of the process state interval is compared with the lower limit of the process allowable interval; when the upper limit of the process state interval is not less than the upper limit of the process allowable interval, or the lower limit of the process state interval is not greater than the lower limit of the process allowable interval, the corresponding process allowable interval constraint is determined as an active constraint. Tracing backwards from the activation constraint along the process constraint topology, we find the preceding process state nodes and control parameter nodes that participate in the calculation of the corresponding endpoints of the process state interval, up to the control parameter node; tracing paths with common nodes and connections are merged to form an activation propagation chain pointing from the control parameter node to the activation constraint.

3. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 1, characterized in that, The process constraint topology includes control parameter nodes, process state nodes, process constraint nodes, and directed associated edges; wherein: The control parameter nodes correspond to the control quantities of the production equipment. The process status nodes correspond to material status, energy status, or equipment operating status. The process constraint nodes correspond to the process allowable range of the process state; The directed association edges include material transfer edges that connect process state nodes according to material flow direction, energy transfer edges that connect process state nodes according to energy flow direction, control transfer edges that connect control parameter nodes and process state nodes according to control action direction, and boundary association edges that connect process state nodes and process constraint nodes.

4. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 2, characterized in that, The conversion to parameter constraint intervals includes: The interval conversion relationships in the activation propagation chain are arranged in reverse order from the activation constraint to the control parameter node; Set the process allowable interval of the activation constraint as the current allowable interval; for the monotonic interval conversion relationship, find the intersection of the current allowable interval and the output range of the monotonic interval conversion relationship, and convert the endpoints of the non-empty intersection into the candidate allowable intervals of the previous node in reverse; For the segmented monotonic interval conversion relationship, the current allowed interval is intersected with the output range of each monotonic segment, the non-empty intersection is converted in reverse, and the union of the resulting candidate allowed intervals of the previous node is calculated. The candidate allowed interval of the preceding node is determined as the allowed interval of the preceding node; when the same preceding node is formed by multiple subsequent nodes to form the candidate allowed interval of the preceding node, the intersection of the candidate allowed interval of the preceding node is calculated; the allowed interval of the preceding node is set as the new current allowed interval and the reverse calculation is performed step by step until the parameter constraint interval of the control parameter node is obtained.

5. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 4, characterized in that, The formation of local process boundaries includes: Each continuous interval in the parameter constraint interval is intersected with the corresponding control parameter interval. All non-empty intersections constitute the local process boundary of the control parameters. When all intersections are empty, the local process boundary is determined as an empty set.

6. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 5, characterized in that, Forming the cooperative descent parameter domain includes: When the local process boundary is not empty, the local process boundary is divided into an increase adjustment range and a decrease adjustment range, with the current control parameter value as the dividing line. The endpoints of the non-empty increasing adjustment interval and the decreasing adjustment interval are converted along the positive direction of the activation propagation chain to obtain each energy consumption rate interval, the corresponding process duration interval, and the process direct emission interval; when the interval conversion relationship contains a turning point, the monotonic segment is divided according to the turning point, and the intervals are converted separately and then the union of the resulting intervals is calculated. Perform interval product on each energy consumption rate interval and the corresponding process duration interval to obtain each energy consumption interval, and sum the energy consumption intervals to obtain the adjusted energy consumption interval; perform interval product on each energy consumption interval and the corresponding energy carbon emission factor interval and sum the sum, and superimpose the process direct emission interval to obtain the adjusted carbon emission interval. Based on the current energy consumption and carbon emission values ​​that are greater than zero, the ratio of the difference between the energy consumption benchmark value and the upper limit of the adjusted energy consumption range to the energy consumption benchmark value is determined as the lower limit of the energy consumption reduction range; the ratio of the difference between the energy consumption benchmark value and the lower limit of the adjusted energy consumption range to the energy consumption benchmark value is determined as the upper limit of the energy consumption reduction range; the carbon emission reduction range is determined according to the same endpoint conversion relationship. The minimum value between the lower limit of the energy consumption reduction range and the lower limit of the carbon emission reduction range is determined as the coordinated reduction value; the adjustment range where both lower limits are greater than zero is determined as the coordinated reduction parameter range, and the coordinated reduction parameter range constitutes the coordinated reduction parameter domain.

7. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 6, characterized in that, The delineation of the local process boundary includes: Select a continuous parameter interval containing the current control parameter value from the local process boundary; when the current control parameter value is located inside the continuous parameter interval, determine the interval from the lower limit of the continuous parameter interval to the current control parameter value and excluding the current control parameter value as the decrease adjustment interval, and determine the interval from the current control parameter value to the upper limit of the continuous parameter interval and excluding the current control parameter value as the increase adjustment interval; When the current control parameter value is equal to the lower limit of the continuous parameter interval, only the increasing adjustment interval is formed; when the current control parameter value is equal to the upper limit of the continuous parameter interval, only the decreasing adjustment interval is formed.

8. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 6, characterized in that, The collaborative optimization parameter combination is obtained by: Identify the same control parameter that is associated with two or more activation propagation chains, and extract the value range of the control parameter from each associated cooperative descent parameter domain; The intersection of the value ranges of the control parameters is determined, and the non-empty intersection is identified as the cross-chain value range of the control parameters; when the intersection is empty, no candidate parameter combination involving the relevant cooperative descent parameter domain is formed. The endpoints of the cross-chain value range and the value ranges of other control parameters, as well as the inflection points of the interval conversion relationship located within the value range, are selected and combined according to the order of the control parameter nodes in the process constraint topology. The combinations that fall into all relevant cooperative descent parameter domains are retained to obtain candidate parameter combinations. The candidate parameter combination is converted along each activation propagation chain in the forward direction to obtain the energy consumption reduction range and carbon emission reduction range respectively. The minimum value of the lower limit of the two reduction ranges is determined as the intra-chain synergistic reduction value, and the minimum value of the intra-chain synergistic reduction value of each activation propagation chain is determined as the combined synergistic reduction value of the candidate parameter combination. Compare the combined collaborative reduction values ​​of each candidate parameter combination, and determine the candidate parameter combination with the largest combined collaborative reduction value as the collaborative optimization parameter combination.

9. The energy consumption and carbon emission collaborative optimization method based on process constraint boundary reconstruction according to claim 8, characterized in that, The determination of the collaborative optimization parameter combination includes: Candidate parameter combinations are arranged from largest to smallest according to their combined synergistic reduction value, and the candidate parameter combination ranked first is selected. When the combined synergistic reduction values ​​of multiple candidate parameter combinations are the same, the lower limits of the energy consumption reduction range and the lower limits of the carbon emission reduction range of the candidate parameter combinations in each activation propagation chain are summed, and the candidate parameter combination with the largest sum is selected. When the sums are the same, calculate the absolute value of the difference between the candidate value and the current control parameter value for each control parameter. Divide the absolute value by the interval width of the continuous local process boundary where the candidate value is located to obtain the normalized adjustment amount. Sum the normalized adjustment amounts in the same candidate parameter combination and determine the candidate parameter combination with the smallest sum of normalized adjustment amounts as the collaborative optimization parameter combination.

Images